AWS Whitepaper

Big Data Analytics Options on AWS

Copyright © 2024 Amazon Web Services, Inc. and/or its affiliates. All rights reserved.
Big Data Analytics Options on AWS AWS Whitepaper

Big Data Analytics Options on AWS: AWS Whitepaper

Copyright © 2024 Amazon Web Services, Inc. and/or its affiliates. All rights reserved.

Amazon's trademarks and trade dress may not be used in connection with any product or service
that is not Amazon's, in any manner that is likely to cause confusion among customers, or in any
manner that disparages or discredits Amazon. All other trademarks not owned by Amazon are
the property of their respective owners, who may or may not be affiliated with, connected to, or
sponsored by Amazon.
Big Data Analytics Options on AWS AWS Whitepaper

Table of Contents
Abstract ............................................................................................................................................ 1
Introduction ..................................................................................................................................... 2
The AWS advantage in big data analytics ..................................................................................... 3
Amazon Kinesis ............................................................................................................................................. 4
Ideal usage patterns ............................................................................................................................... 6
Cost model ............................................................................................................................................... 7
Performance ............................................................................................................................................. 7
Durability and availability ..................................................................................................................... 8
Scalability and elasticity ........................................................................................................................ 8
Interfaces ................................................................................................................................................... 8
Anti-patterns ............................................................................................................................................ 8
Amazon MSK ................................................................................................................................................. 9
Ideal usage patterns ............................................................................................................................. 10
Cost model ............................................................................................................................................. 10
Performance ........................................................................................................................................... 11
Durability and availability ................................................................................................................... 11
Scalability and elasticity ...................................................................................................................... 11
Interfacess .............................................................................................................................................. 12
Anti-patterns .......................................................................................................................................... 12
AWS Lambda ............................................................................................................................................... 12
Ideal usage patterns ............................................................................................................................. 13
Cost model ............................................................................................................................................. 13
Performance ........................................................................................................................................... 13
Durability and availability ................................................................................................................... 14
Scalability and elasticity ...................................................................................................................... 14
Interfaces ................................................................................................................................................ 14
Anti-patterns .......................................................................................................................................... 15
Amazon EMR ............................................................................................................................................... 15
Ideal usage patterns ............................................................................................................................. 16
Cost model ............................................................................................................................................. 16
Performance ........................................................................................................................................... 16
Durability and availability ................................................................................................................... 17
Scalability and elasticity ...................................................................................................................... 17
Interfaces ................................................................................................................................................ 18

iii
Big Data Analytics Options on AWS AWS Whitepaper

Anti-patterns .......................................................................................................................................... 21
AWS Glue ..................................................................................................................................................... 22
Ideal usage patterns ............................................................................................................................. 22
Cost model ............................................................................................................................................. 23
Performance ........................................................................................................................................... 24
Durability and availability ................................................................................................................... 24
Scalability and elasticity ...................................................................................................................... 24
Interfaces ................................................................................................................................................ 24
Anti-patterns .......................................................................................................................................... 25
AWS Lake Formation ................................................................................................................................. 25
Ideal usage patterns ............................................................................................................................. 25
Cost model ............................................................................................................................................. 26
Performance ........................................................................................................................................... 27
Durability and availability ................................................................................................................... 27
Scalability and elasticity ...................................................................................................................... 27
Interfaces ................................................................................................................................................ 27
Anti-patterns .......................................................................................................................................... 28
Amazon Machine Learning ....................................................................................................................... 28
Ideal usage patterns ............................................................................................................................. 30
Cost model ............................................................................................................................................. 31
Performance ........................................................................................................................................... 32
Durability and availability ................................................................................................................... 33
Scalability and elasticity ...................................................................................................................... 33
Interfaces ................................................................................................................................................ 34
Anti-patterns .......................................................................................................................................... 34
Amazon DynamoDB ................................................................................................................................... 35
Ideal usage patterns ............................................................................................................................. 36
Cost model ............................................................................................................................................. 36
Performance ........................................................................................................................................... 37
Durability and availability ................................................................................................................... 38
Scalability and elasticity ...................................................................................................................... 38
Interfaces ................................................................................................................................................ 38
Anti-patterns .......................................................................................................................................... 39
Amazon Redshift ........................................................................................................................................ 39
Ideal usage patterns ............................................................................................................................. 40
Cost model ............................................................................................................................................. 41

iv
Big Data Analytics Options on AWS AWS Whitepaper

Performance ........................................................................................................................................... 41
Durability and availability ................................................................................................................... 41
Scalability and elasticity ...................................................................................................................... 42
Interfaces ................................................................................................................................................ 42
Anti-patterns .......................................................................................................................................... 43
Amazon OpenSearch Service ................................................................................................................... 43
Ideal usage patterns ............................................................................................................................. 44
Cost model ............................................................................................................................................. 44
Performance ........................................................................................................................................... 45
Durability and availability ................................................................................................................... 45
Scalability and elasticity ...................................................................................................................... 46
Interfaces ................................................................................................................................................ 47
Fine-grained access control ................................................................................................................ 47
Anti-patterns .......................................................................................................................................... 48
Amazon QuickSight ................................................................................................................................... 48
Ideal usage patterns ............................................................................................................................. 49
Cost model ............................................................................................................................................. 50
Performance ........................................................................................................................................... 50
Durability and availability ................................................................................................................... 51
Scalability and elasticity ...................................................................................................................... 51
Interfaces ................................................................................................................................................ 51
Anti-patterns .......................................................................................................................................... 51
Amazon Compute Services ....................................................................................................................... 52
Ideal usage patterns ............................................................................................................................. 52
Cost model ............................................................................................................................................. 53
Performance ........................................................................................................................................... 53
Durability and availability ................................................................................................................... 53
Scalability and elasticity ...................................................................................................................... 53
Interfaces ................................................................................................................................................ 54
Anti-patterns .......................................................................................................................................... 54
Amazon Athena .......................................................................................................................................... 54
Ideal usage patterns ............................................................................................................................. 55
Cost model ............................................................................................................................................. 55
Performance ........................................................................................................................................... 56
Durability and availability ................................................................................................................... 56
Scalability and elasticity ...................................................................................................................... 56

v
Big Data Analytics Options on AWS AWS Whitepaper

Security, authorization, and encryption ........................................................................................... 56

Interfaces ................................................................................................................................................ 57
Anti-patterns .......................................................................................................................................... 57
Solving big data problems on AWS ............................................................................................. 58
Example 1: Queries against an Amazon S3 data lake ........................................................................ 59
Example 2: Capturing and analyzing sensor data ............................................................................... 60
Example 3: sentiment analysis of social media ................................................................................... 63
Conclusion ...................................................................................................................................... 66
Contributors ................................................................................................................................... 67
Further reading .............................................................................................................................. 68
Document revisions ....................................................................................................................... 69
Notices ............................................................................................................................................ 70

vi
Big Data Analytics Options on AWS AWS Whitepaper

Big Data Analytics Options on AWS

Publication date: July 26, 2021 (Document revisions)

This whitepaper helps architects, data scientists, and developers understand the big data analytics
options available in the Amazon Web Services (AWS) Cloud. It provides an overview of services,
including:

• Ideal usage patterns

• Cost model
• Performance
• Durability and availability
• Scalability and elasticity
• Interfaces
• Anti-patterns

This paper concludes with scenarios that showcase the analytics options used, as well as additional
resources for getting started with big data analytics on AWS.

1
Big Data Analytics Options on AWS AWS Whitepaper

Introduction
As the world becomes more digital, the amount of data created and collected constantly grows
and accelerates. Analysis of this ever-growing data becomes a challenge with traditional analytical
tools. Innovation is required to bridge the gap between generated data and data that can be
analyzed effectively.

Big data tools and technologies offer opportunities to analyze data efficiently so you can better
understand customer preferences, gain a competitive advantage in the marketplace, and grow your
business. Data management architectures have evolved from the traditional data warehousing
model to more complex architectures that address more requirements, such as real-time and batch
processing, structured and unstructured data, high velocity transactions, and so on.

AWS provides a broad platform of managed services to help you build, secure, and seamlessly scale
end-to-end big data applications quickly and with ease. Whether your applications require real-
time streaming or batch data processing, AWS provides the infrastructure and tools to tackle your
next big data project. There is no hardware to procure, no infrastructure to maintain and scale—
only what you need to collect, store, process, and analyze big data. AWS has a system of analytical
solutions specifically designed to handle this growing amount of data and provide insight into your
business.

2
Big Data Analytics Options on AWS AWS Whitepaper

The AWS advantage in big data analytics

Analyzing large datasets requires significant compute capacity that can vary in size, based on
the amount of input data and the type of analysis. This characteristic of big data workloads is
ideally suited to the pay-as-you-go cloud computing model, where applications can easily scale
up and down based on demand. As requirements change, you can easily resize your environment
(horizontally or vertically) on AWS to meet your needs, without having to wait for additional
hardware or over-investing to provision enough capacity.

For mission-critical applications on a more traditional infrastructure, system designers have no

choice but to over-provision, because a surge in additional data due to an increase in business
needs must be something the system can handle. By contrast, on AWS, you can provision more
capacity and compute in a matter of minutes, meaning that your big data applications grow and
shrink as demand dictates, and your system runs as close to optimal efficiency as possible.

In addition, you get flexible computing on a global infrastructure with access to the many different
geographic Regions that AWS offers, along with the ability to use other scalable services that
augment to build sophisticated big data applications. These other services include:

• Amazon Simple Storage Service (Amazon S3) to store data

• AWS Glue to orchestrate jobs to move and transform the data easily
• AWS IoT, which lets connected devices interact with cloud applications and other connected
devices

As the amount of data being generated continues to grow, AWS has many options to get that
data to the cloud, including secure devices like AWS Snow Family to accelerate petabyte-scale
data transfers, delivery streams with Amazon Data Firehose to load streaming data continuously,
migrating databases using AWS Database Migration Service, and scalable private connections
through AWS Direct Connect.

As mobile continues to rapidly grow in usage, you can use the suite of services within the AWS
Mobile Hub to collect and measure app usage and data, or export that data to another service for
further custom analysis.

These capabilities of AWS make it an ideal fit for solving big data problems, and many customers
have implemented successful big data analytics workloads on AWS. For more information about
case studies, see Big Data Customer Success Stories.

3
Big Data Analytics Options on AWS AWS Whitepaper

The following services for collecting, processing, storing, and analyzing big data are described in
order:

• Amazon Kinesis
• Amazon Managed Streaming for Apache Kafka (Amazon MSK)
• AWS Lambda
• Amazon Elastic Map Reduce (Amazon EMR)
• AWS Glue
• AWS Lake Formation
• Amazon Machine Learning
• Amazon DynamoDB
• Amazon Redshift
• Amazon OpenSearch Service (OpenSearch Service)
• Amazon QuickSight
• Amazon Compute Services (Amazon Elastic Compute Cloud (Amazon EC2) instances, Amazon
Elastic Container Service (Amazon ECS), and Amazon Elastic Kubernetes Service (Amazon EKS)
are available for self-managed big data applications.)
• Amazon Athena

Amazon Kinesis
Amazon Kinesis is a platform for streaming data on AWS that makes it easy to load and analyze
streaming data. Amazon Kinesis also enables you to build custom streaming data applications for
specialized needs. With Kinesis, you can ingest real-time data such as application logs, website
clickstreams, Internet of Things (IoT) telemetry data, and more into your databases, data lakes, and
data warehouses, or build your own real-time applications using this data. Amazon Kinesis enables
you to process and analyze data as it arrives and respond in real-time instead of having to wait
until all your data is collected before the processing can begin.

Currently there are four pieces of the Kinesis platform that can be utilized based on your use case:

• Amazon Kinesis Data Streams enables you to build custom applications that process or analyze
streaming data.

Amazon Kinesis 4
Big Data Analytics Options on AWS AWS Whitepaper

• Amazon Kinesis Video Streams enables you to build custom applications that process or analyze
streaming video.
• Amazon Data Firehose enables you to deliver real-time streaming data to AWS destinations such
as Amazon S3, Amazon Redshift, OpenSearch Service, and Splunk.
• Amazon Managed Service for Apache Flink enables you to process and analyze streaming data
with standard SQL or with Java (managed Apache Flink).

Kinesis Data Streams and Kinesis Video Streams enable you to build custom applications that
process or analyze streaming data in real time. Kinesis Data Streams can continuously capture
and store terabytes of data per hour from hundreds of thousands of sources, such as website
clickstreams, financial transactions, social media feeds, IT logs, and location-tracking events.
Kinesis Video Streams can continuously capture video data from smartphones, security cameras,
drones, satellites, dashcams, and other edge devices.

With the Amazon Kinesis Client Library (KCL), you can build Amazon Kinesis applications and use
streaming data to power real-time dashboards, generate alerts, and implement dynamic pricing
and advertising. You can also emit data from Kinesis Data Streams and Kinesis Video Streams to
other AWS services such as Amazon S3, Amazon Redshift, Amazon EMR, and AWS Lambda.

Provision the level of input and output required for your data stream, in blocks of one megabyte
per second (MB/sec), using the AWS Management Console, API, or SDKs. The size of your stream
can be adjusted up or down at any time without restarting the stream and without any impact on
the data sources pushing data to the stream. Within seconds, data put into a stream is available for
analysis.

With Amazon Data Firehose, you don't need to write applications or manage resources. You
configure your data producers to send data to Firehose and it automatically delivers the data to
the AWS destination or third party (Splunk) that you specified. You can also configure Firehose to
transform your data before data delivery. It is a fully managed service that automatically scales
to match the throughput of your data and requires no ongoing administration. It can also batch,
compress, and encrypt the data before loading it, minimizing the amount of storage used at the
destination and increasing security.

Amazon Managed Service for Apache Flink is the easiest way to process and analyze real-time,
streaming data. With Managed Service for Apache Flink, you just use standard SQL or Java (Flink)
to process your data streams, so you don’t have to learn any new programming languages. Simply
point Managed Service for Apache Flink at an incoming data stream, write your SQL queries,
and specify where you want to load the results. Managed Service for Apache Flink takes care of

Amazon Kinesis 5
Big Data Analytics Options on AWS AWS Whitepaper

running your SQL queries continuously on data while it’s in transit and sending the results to the
destinations.

For complex data processing applications, Amazon Managed Service for Apache Flink provides an
option use open-source libraries such as Apache Flink, Apache Beam, AWS SDK, and AWS service
integrations. It includes more than ten connectors from Apache Flink, and gives you the ability to
build custom integrations. It’s also compatible with the AWS Glue Schema Registry, a serverless
feature of AWS Glue that enables you to validate and control the evolution of streaming data using
registered Apache Avro schemas.

You can use Apache Flink in Amazon Managed Service for Apache Flink to build applications
whose processed records affect the results exactly once, referred to as exactly once processing.
This means that even in the case of an application disruption, like internal service maintenance
or user-initiated application update, the service will ensure that all data is processed and there is
no duplicate data. The service stores previous and in-progress computations, or state, in running
application storage. This enables you to compare real-time and past results over any time period,
and provides fast recovery during application disruptions.

The subsequent sections focus primarily on Amazon Kinesis Data Streams.

Ideal usage patterns

Amazon Kinesis Data Steams is useful wherever there is a need to move data rapidly off producers
(data sources) and continuously process it. That processing can be to transform the data before
emitting into another data store, drive real-time metrics and analytics, or derive and aggregate
multiple streams into more complex streams, or downstream processing. The following are typical
scenarios for using Kinesis Data Streams for analytics:

• Real-time data analytics – Kinesis Data Streams enables real-time data analytics on streaming
data, such as analyzing website clickstream data and customer engagement analytics.
• Log and data feed intake and processing – With Kinesis Data Streams, you can have producers
push data directly into an Amazon Kinesis stream. For example, you can submit system and
application logs to Kinesis Data Streams and access the stream for processing within seconds.
This prevents the log data from being lost if the front-end or application server fails, and reduces
local log storage on the source. Kinesis Data Streams provides accelerated data intake because
you are not batching up the data on the servers before you submit it for intake.
• Real-time metrics and reporting – You can use data ingested into Kinesis Data Streams for
extracting metrics and generating KPIs to power reports and dashboards at real-time speeds.

Ideal usage patterns 6

Big Data Analytics Options on AWS AWS Whitepaper

This enables data-processing application logic to work on data as it is streaming in continuously,

rather than waiting for data batches to arrive.

Cost model

Amazon Kinesis Data Streams has simple pay-as-you-go pricing, with no upfront costs or minimum
fees, and you pay only for the resources you consume. An Amazon Kinesis stream is made up
of one or more shards. Each shard gives you a capacity of five read transactions per second, up
to a maximum total of 2 MB of data read per second. Each shard can support up to 1,000 write
transactions per second, and up to a maximum total of 1 MB data written per second.

The data capacity of your stream is a function of the number of shards that you specify for the
stream. The total capacity of the stream is the sum of the capacity of each shard. There are two
components to pricing:

• Primary pricing includes an hourly charge per shard and a charge for each one million PUT
transactions.

• Pricing for optional components for extended retention and enhanced fan-out.

For more information, see Amazon Kinesis Data Streams Pricing. Applications that run on Amazon
EC2 and process Amazon Kinesis streams also incur standard Amazon EC2 costs.

Performance

Amazon Kinesis Data Streams enables you to choose the throughput capacity you require in terms
of shards. With each shard in an Amazon Kinesis stream, you can capture up to 1 megabyte per
second of data at 1,000 write transactions per second. Your Amazon Kinesis applications can read
data from each shard at up to 2 megabytes per second. You can provision as many shards as you
need to get the throughput capacity you want; for example, a one gigabyte per second data stream
would require 1024 shards.

Additionally, there is a new feature. Enhanced fan-out enables developers to scale up the number
of stream consumers (applications reading data from a stream in real-time) by offering each stream
consumer their own read throughput. Developers can register stream consumers to use enhanced
fan-out and receive their own 2MB/sec pipe of read throughput per shard. This throughput
automatically scales with the number of shards in a stream.

Cost model 7
Big Data Analytics Options on AWS AWS Whitepaper

Durability and availability

Amazon Kinesis Data Streams synchronously replicates data across three Availability Zones in an
AWS Region, providing high availability and data durability.

Additionally, you can store a cursor in Amazon DynamoDB to durably track what has been read
from an Amazon Kinesis stream. In the event that your application fails in the middle of reading
data from the stream, you can restart your application and use the cursor to pick up from the exact
spot where the failed application left off.

Scalability and elasticity

You can increase or decrease the capacity of the stream at any time according to your business or
operational needs, without any interruption to ongoing stream processing. By using API calls or
development tools, you can automate scaling of your Amazon Kinesis Data Streams environment
to meet demand and ensure you only pay for what you need.

Interfaces

There are two interfaces to Kinesis Data Streams:

• Input which is used by data producers to put data into Kinesis Data Streams

• Output to process and analyze data that comes in

Producers can write data using the Amazon Kinesis PUT API, an AWS Software Development Kit
(SDK) or toolkit abstraction, the Amazon Kinesis Producer Library (KPL), or the Amazon Kinesis
Agent.

For processing data that has already been put into an Amazon Kinesis stream, there are client
libraries provided to build and operate real-time streaming data processing applications. The KCL
acts as an intermediary between Amazon Kinesis Data Streams and your business applications
which contain the specific processing logic. There is also integration to read from an Amazon
Kinesis stream into Apache Spark Streaming running on Amazon EMR.

Anti-patterns

Amazon Kinesis Data Streams has the following anti-patterns:

Durability and availability 8

Big Data Analytics Options on AWS AWS Whitepaper

• Small scale consistent throughput – Even though Kinesis Data Streams works for streaming
data at 200 KB per second or less, it is designed and optimized for larger data throughputs.
• Long-term data storage and analytics – Kinesis Data Streams is not suited for long-term data
storage. By default, data is retained for 24 hours, and you can extend the retention period by up
to 365 days.

Amazon Managed Streaming for Apache Kafka (Amazon MSK)

Amazon MSK is a fully managed service that makes it easy for you to build and run applications
that use Apache Kafka to process streaming data. Apache Kafka is an open-source platform for
building real-time streaming data pipelines and applications. With Amazon MSK, you can use
native Apache Kafka APIs to populate data lakes, stream changes to and from databases, and
power machine learning and analytics applications.

Apache Kafka clusters are challenging to set up, scale, and manage in production. When you run
Apache Kafka on your own, you need to provision servers, configure Apache Kafka manually,
replace servers when they fail, orchestrate server patches and upgrades, architect the cluster for
high availability, ensure data is durably stored and secured, set up monitoring and alarms, and
carefully plan scaling events to support load changes. Amazon MSK makes it easy for you to build
and run production applications on Apache Kafka without needing Apache Kafka infrastructure
management expertise.

With a few clicks in the Amazon MSK console (sign-in required), you can create a fully managed
Apache Kafka cluster that follows Apache Kafka’s deployment best practices, or you can create
your own cluster using your own custom configuration. After you create your desired configuration,
Amazon MSK automatically provisions, configures, and manages the operations of your Apache
Kafka cluster and Apache ZooKeeper nodes.

An Amazon MSK cluster is the primary Amazon MSK resource that you can create in your account.

Following are the primary components that work together in MSK:

• Broker nodes — When creating an Amazon MSK cluster, you specify how many broker nodes you
want Amazon MSK to create in each Availability Zone. Each Availability Zone has its own virtual
private cloud (VPC) subnet.
• ZooKeeper nodes — Amazon MSK also creates the Apache ZooKeeper nodes for you. Apache
ZooKeeper is an open-source server that enables highly reliable distributed coordination.

Amazon MSK 9
Big Data Analytics Options on AWS AWS Whitepaper

• Producers, consumers, and topic creators — Amazon MSK enables you to use Apache Kafka
data-plane operations to create topics and to produce and consume data.
• Cluster operations — You can use the AWS Management Console, the AWS Command Line
Interface (AWS CLI), or the APIs in the SDK to perform control-plane operations. For example,
you can create or delete an Amazon MSK cluster, list all the clusters in an account, view the
properties of a cluster, and update the number and type of brokers in a cluster.

Amazon MSK detects and automatically recovers from the most common failure scenarios for
clusters, so that your producer and consumer applications can continue their write and read
operations with minimal impact. When Amazon MSK detects a broker failure, it mitigates the
failure or replaces the unhealthy or unreachable broker with a new one. In addition, where possible,
it reuses the storage from the older broker to reduce the data that Apache Kafka needs to replicate.
Your availability impact is limited to the time required for Amazon MSK to complete the detection
and recovery. After a recovery, your producer and consumer apps can continue to communicate
with the same broker IP addresses that they used before the failure.

Ideal usage patterns

The AWS Streaming Data Solution for Amazon MSK enables you to capture, store, process, and
deliver real-time streaming data. This service helps you address real-time streaming use cases; for
example:

• Capture high volume application log files

• Analyze website clickstreams
• Process database event streams
• Track financial transactions
• Aggregate social media feeds
• Collect IT log files
• Continuously deliver to a data lake

Cost model

Amazon MSK has a simple “pay only for what you use” model. There are no minimum fees or
upfront commitments. There are charges for the time your broker instances run, the storage used
monthly, and standard data transfer fees for data in and out of the cluster. Apache Kafka broker

Ideal usage patterns 10

Big Data Analytics Options on AWS AWS Whitepaper

instance usage is billed on an hourly basis (billed at one second resolution), with varying fees
depending on the size of the Apache Kafka broker instance and active brokers in your Amazon
MSK clusters. Amazon MSK charges for the amount of storage you provision in the cluster. This is
calculated by adding up the GB per broker each hour and dividing by the number of hours in the
month, resulting in a value in "GB-Month."

There are no additional charges for data transfer between brokers or between Apache ZooKeeper
nodes and brokers. Standard AWS data transfer charges are charged for data transferred in and out
of Amazon MSK clusters. See Amazon MSK pricing for further details.

Performance

Amazon MSK allows you to choose the right type and number of brokers for your cluster. You
can size your cluster based on your ingestion rate, hours of retention and data output rates. The
number of partitions per broker is affected by use case and configuration. For more information
about the different broker types, see Broker types.

Durability and availability

Use the following recommendations so that your MSK cluster can be highly available during an
update (such as when you're updating the broker type or Apache Kafka version, for example) or
when Amazon MSK is replacing a broker.

• Ensure that the replication factor (RF) is at least two for two-AZ clusters and at least three for
three-AZ clusters. An RF of one can lead to offline partitions during a rolling update.

• Set minimum in-sync replicas (minISR) to at most RF-1. A minISR that is equal to the RF can
prevent producing to the cluster during a rolling update. A minISR of two allows three-way
replicated topics to be available when one replica is offline.

• Ensure client connection strings include multiple brokers. Having multiple brokers in a client’s
connection string allows for failover when a specific broker is offline for an update.

Scalability and elasticity

You can increase the capacity of the cluster at any time according to your business or operational
needs. You can use Amazon MSK operation to increase the number of brokers in your MSK cluster.
To expand a cluster, make sure that it is in the ACTIVE state.

Performance 11
Big Data Analytics Options on AWS AWS Whitepaper

You can also increase storage per broker. You can increase the amount of EBS storage per broker,
but you can't decrease the storage. Storage volumes remain available during this scaling-up
operation. You can use:

• Automatic scaling — You can configure Amazon Managed Streaming for Apache Kafka to
automatically expand your cluster's storage in response to increased usage using Application
Auto Scaling policies. Your automatic scaling policy sets the target disk utilization and the
maximum scaling capacity.
• Manual scaling — To increase storage, wait for the cluster to be in the ACTIVE state. Storage
scaling has a cooldown period of at least six hours between events. Even though the operation
makes additional storage available right away, the service performs optimizations on your cluster
that can take up to 24 hours or more. The duration of these optimizations is proportional to your
storage size.

Interfaces
Amazon MSK has deep AWS service integrations with Amazon EMR, AWS Lambda, Amazon
Managed Service for Apache Flink, and AWS Glue Streaming ETL. It also works with Kafka Connect,
Mirror Maker, Kafka Streams, and a number of 3rd party frameworks like Apache Spark, Apache
Storm, and so on. The producer side APIs add messages to the cluster for a topic. The consumer
side APIs get messages for a topic as a stream of messages.

Anti-patterns
Amazon MSK has the following anti-patterns:

• Ad hoc queries — MSK is a stream of unbounded data. It is not used for ad hoc queries.
• Long-term data storage and analytics — MSK is not suited for long-term data storage.

AWS Lambda
AWS Lambda enables you to run code without provisioning or managing servers. You pay only
for the compute time you consume – there is no charge when your code is not running. With
Lambda, you can run code for virtually any type of application or backend service – all with zero
administration. Just upload your code and Lambda takes care of everything required to run and
scale your code with high availability. You can set up your code to automatically trigger from other
AWS services or call it directly from any web or mobile app.

Interfacess 12
Big Data Analytics Options on AWS AWS Whitepaper

Ideal usage patterns

AWS Lambda enables you to run code in response to triggers such as changes in data, shifts
in system state, or actions by users. Lambda can be directly triggered by AWS services such as
Amazon S3, DynamoDB, Amazon Kinesis Data Streams, Amazon Simple Notification Service
(Amazon SNS), and Amazon CloudWatch, enabling you to build a variety of real-time data
processing systems:

• Real-time file processing – You can trigger Lambda to invoke a process where a file has been
uploaded to Amazon S3 or modified. For example, to change an image from color to gray scale
after it has been uploaded to Amazon S3.
• Real-time stream processing – You can use Kinesis Data Streams and Lambda to process
streaming data for click stream analysis, log filtering, and social media analysis.
• Extract, transform, load (ETL) – You can use Lambda to run code that transforms data and loads
that data into one data repository to another.
• Replace Cron – Use schedule expressions to run a Lambda function at regular intervals as a
cheaper and more available solution than running cron on an EC2 instance.
• Process AWS Events – Many other services, such as AWS CloudTrail, can act as event sources
simply by logging to Amazon S3 and using S3 bucket notifications to trigger Lambda functions.

Cost model

With AWS Lambda you only pay for what you use. You are charged based on the number of
requests for your functions and the time your code runs. The Lambda free tier includes 1M free
requests per month and 400,000 GB-seconds of compute time per month. You are charged $0.20
per 1 million requests thereafter ($0.0000002 per request). Additionally, the duration of your code
running is priced in relation to memory allocated. You are charged $0.00001667 for every GB-
second used.

See AWS Lambda Pricing for more details.

Performance

After deploying your code into Lambda for the first time, your functions are typically ready to
call within seconds of upload. Lambda is designed to process events within milliseconds. Latency
will be higher immediately after a Lambda function is created, updated, or if it has not been used

Ideal usage patterns 13

Big Data Analytics Options on AWS AWS Whitepaper

recently. To improve performance, Lambda may choose to retain an instance of your function and
reuse it to serve a subsequent request, rather than creating a new copy.

The Lambda provisioned concurrency feature provides customers greater control over performance
of their serverless applications at any scale. Functions using provisioned concurrency run with
consistent start-up latency, making them ideal for building interactive mobile or web backends,
latency sensitive microservices, and synchronously invoked APIs. With provisioned concurrency,
functions can instantaneously serve a burst of traffic with consistent start-up latency for every
invoke up to the specified scale.

To learn more about how Lambda reuses function instances, see Getting started with Lambda. Your
code should not assume that this reuse will always happen.

Durability and availability

AWS Lambda is designed to use replication and redundancy to provide high availability for both
the service itself and for the Lambda functions it operates. There are no maintenance windows
or scheduled downtimes for either. On failure, Lambda functions being invoked synchronously
respond with an exception. Lambda functions being invoked asynchronously are retried at least
three times, after which the event may be rejected.

Scalability and elasticity

There is no limit on the number of Lambda functions that you can run. However, Lambda has a
default safety throttle of 1,000 concurrent runs per account per Region. A member of the AWS
support team can increase this limit. Lambda is designed to scale automatically on your behalf.
There are no fundamental limits to scaling a function. Lambda dynamically allocates capacity to
match the rate of incoming events.

Interfaces
Lambda functions can be managed in a variety of ways. You can easily list, delete, update, and
monitor your Lambda functions using the dashboard in the Lambda console. You also can use the
AWS CLI and AWS SDK to manage your Lambda functions.

You can trigger a Lambda function from an AWS event, such as Amazon S3 bucket notifications,
Amazon DynamoDB Streams, Amazon CloudWatch logs, Amazon Simple Email Service (Amazon
SES), Amazon Kinesis Data Streams, Amazon SNS, Amazon Cognito, and more. Any API call in any
service that supports AWS CloudTrail can be processed as an event in Lambda by responding to
CloudTrail audit logs. For more information about event sources, see Lambda Event Sources.

Durability and availability 14

Big Data Analytics Options on AWS AWS Whitepaper

AWS Lambda supports code written in Node.js (JavaScript), Python, Java (Java 8 compatible), C#
(.NET Core), Go, PowerShell and Ruby. Your code can include existing libraries, even native ones.
See AWS documentation on using Node.js, Python, Java, C#, Go, PowerShell, and Ruby.

You can package your Lambda function code and dependencies as a container image, using
tools such as the Docker CLI. You can then upload the image to your container registry hosted
on Amazon ECR. Note that you must create the Lambda function from the same account as the
container registry in Amazon ECR.

Anti-patterns
AWS Lambda has the following anti-patterns:

• Long running application — Each Lambda function must complete within 900 seconds. For
long running applications that may require jobs to run longer than fifteen minutes, Amazon EC2,
Amazon EKS, or Amazon ECS is recommended. Alternately, you can use Step Functions.
• Dynamic websites — While it is possible to run a static website with AWS Lambda, running a
highly dynamic and large volume website can be performance prohibitive. Utilizing Amazon EC2,
Amazon EKS, or Amazon ECS and AWS CloudFormation is a recommended use-case.
• Stateful applications — Lambda code must be written in a “stateless” style, meaning it should
assume there is no affinity to the underlying compute infrastructure. Local file system access,
child processes, and similar artifacts may not extend beyond the lifetime of the request, and
any persistent state should be stored in Amazon S3, DynamoDB, or another internet-available
storage service.

Amazon EMR
Amazon EMR is the industry-leading cloud big data platform for processing vast amounts of
data using open source tools such as Apache Spark, Apache Hive, Apache HBase, Apache Flink,
Apache Hudi, and Presto. Amazon EMR makes it easy to set up, operate, and scale your big data
environments by automating time-consuming tasks like provisioning capacity and tuning clusters.
With EMR you can run petabyte-scale analysis at less than half of the cost of traditional on-
premises solutions and over 3x faster than standard Apache Spark. You can run workloads on
Amazon EC2 instances, on Amazon EKS clusters, or on-premises using EMR on AWS Outposts.

Amazon EMR does all the work involved with provisioning, managing, and maintaining the
infrastructure and software of a Hadoop cluster. Amazon EMR now supports Amazon EC2 M6g
instances to deliver the best price performance for cloud workloads. Amazon EC2 M6g instances

Anti-patterns 15
Big Data Analytics Options on AWS AWS Whitepaper

are powered by AWS Graviton2 processors that are custom designed by AWS, utilizing 64-bit Arm
Neoverse cores. Amazon EMR provides up to 35% lower cost and up to 15% improved performance
for Spark workloads on Graviton2-based instances versus previous generation instances.

Ideal usage patterns

Amazon EMR’s flexible framework reduces large processing problems and data sets into smaller
jobs and distributes them across many compute nodes in a Hadoop cluster. This capability lends
itself to many usage patterns with big data analytics. Here are a few examples:

• Log processing and analytics

• Large ETL data movement
• Risk modeling and threat analytics
• Ad targeting and click stream analytics
• Genomics
• Predictive analytics
• Ad hoc data mining and analytics

For more information, see the documentation for Amazon EMR.

Cost model
With Amazon EMR, you can launch a persistent cluster that stays up indefinitely, or a temporary
cluster that ends after the analysis is complete. In either scenario, you pay only for the hours the
cluster is up.

Amazon EMR supports a variety of Amazon EC2 instance types (standard, high CPU, high memory,
high I/O, and so on) and all Amazon EC2 pricing options (On-Demand, Reserved, and Spot). When
you launch an Amazon EMR cluster (also called a "job flow"), you choose how many and what type
of Amazon EC2 instances to provision. The Amazon EMR price is in addition to the Amazon EC2
price.

For more information, see Amazon EMR pricing.

Performance
Amazon EMR performance is driven by the type of EC2 instances on which you choose to run your
cluster, and how many you chose to run your analytics. You should choose an instance type suitable

Ideal usage patterns 16

Big Data Analytics Options on AWS AWS Whitepaper

for your processing requirements, with sufficient memory, storage, and processing power. With the
introduction of Graviton2 instances, you can see improved performance of up to 15% relative to
equivalent previous generation instances. For more information about EC2 instance specifications,
see Amazon EC2 Instance Types.

An important consideration when you create an EMR cluster is how you configure Amazon EC2
instances. EC2 instances in an EMR cluster are organized into node types.

The node types in Amazon EMR are as follows:

• Primary node — A node that manages the cluster by running software components to
coordinate the distribution of data and tasks among other nodes for processing. The primary
node tracks the status of tasks and monitors the health of the cluster. Every cluster has a primary
node, and it's possible to create a single-node cluster with only the primary node.
• Core node — A node with software components that run tasks and store data in the Hadoop
Distributed File System (HDFS) on your cluster. Multi-node clusters have at least one core node.
• Task node — A node with software components that only runs tasks and does not store data in
HDFS. Task nodes are optional.

Durability and availability

By default, Amazon EMR is fault tolerant for core node failures and continues job execution if a
dependent node goes down. Amazon EMR will also provision a new node when a core node fails.
However, Amazon EMR will not replace nodes if all nodes in the cluster are lost.

You can monitor the health of nodes and replace failed nodes with Amazon CloudWatch. When you
launch an Amazon EMR cluster, you can choose to have one or three primary nodes in your cluster.
Launching a cluster with three primary nodes is only supported by Amazon EMR version 5.23.0 and
later. EMR can take advantage of EC2 placement groups to ensure primary nodes are placed on
distinct underlying hardware to further improve cluster availability.

For more information, see EMR integration with EC2 placement groups.

Scalability and elasticity

With Amazon EMR, it's easy to resize a running cluster. You can add core nodes which hold the
HDFS at any time to increase your processing power and increase the HDFS storage capacity (and
throughput). Additionally, you can use Amazon S3 natively, or using EMRFS along with or instead

Durability and availability 17

Big Data Analytics Options on AWS AWS Whitepaper

of local HDFS, which enables you to decouple your memory and compute from your storage
providing greater flexibility and cost efficiency.

You can also add and remove task nodes at any time which can process Hadoop jobs, but do not
maintain HDFS. Some customers add hundreds of instances to their clusters when their batch
processing occurs, and remove the extra instances when processing completes. For example, you
may not know how much data your clusters will be handling in six months, or you may have spiky
processing needs.

With Amazon EMR, you don't need to guess your future requirements or provision for peak demand
because you can easily add or remove capacity at any time.

You can add all new clusters of various sizes and remove them at any time with a few clicks in the
console or by a programmatic API call.

Additionally, you can configure instance fleets for a cluster to choose a wide variety of provisioning
options for EC2 instances. With instance fleets you can specify target capacities on On-Demand
Instances, and Spot Instances within each fleet. Amazon EMR tries to provide the capacity you need
with the best mix of capacity and price based on your selection of Availability Zones.

While a cluster is running, if Amazon EC2 reclaims a Spot Instance because of a price increase, or
an instance fails, Amazon EMR tries to replace the instance with any of the instance types that you
specify. This makes it easier to regain capacity if a node is lost for any reason.

Interfaces
Amazon EMR supports many tools on top of Hadoop that can be used for big data analytics and
each has their own interfaces. Here is a brief summary of the most popular options:

Hive

Hive is an open source data warehouse and analytics package that runs on top of Hadoop. Hive
is operated by Hive QL, a SQL-based language, which enables users to structure, summarize,
and query data. Hive QL goes beyond standard SQL, adding first-class support for map/reduce
functions and complex extensible user-defined data types like JSON and Thrift. This capability
allows processing of complex and unstructured data sources such as text documents and log files.

Hive allows user extensions via user-defined functions written in Java. Amazon EMR has made
numerous improvements to Hive, including direct integration with DynamoDB and Amazon S3.
For example, with Amazon EMR you can load table partitions automatically from S3, you can write

Interfaces 18
Big Data Analytics Options on AWS AWS Whitepaper

data to tables in S3 without using temporary files, and you can access resources in S3, such as
scripts for custom map and/or reduce operations and additional libraries.

For more information, see Apache Hive in the Amazon EMR Release Guide.

Pig

Pig is an open-source analytics package that runs on top of Hadoop. Pig is operated by Pig Latin,
an SQL-like language which enables users to structure, summarize, and query data. Pig Latin also
adds first-class support for map and reduce functions and complex extensible user-defined data
types. This capability allows processing of complex and unstructured data sources such as text
documents and log files.

Pig allows user extensions via user-defined functions written in Java. Amazon EMR has made
numerous improvements to Pig, including the ability to use multiple file systems (normally, Pig can
only access one remote file system), the ability to load customer JARs and scripts from S3 (such
as “REGISTER s3://my-bucket/piggybank.jar”), and additional functionality for String and
DateTime processing.

For more information, see Apache Pig in the Amazon EMR Release Guide.

Spark

Spark is an open-source data analytics engine built on Hadoop with the fundamentals for in-
memory MapReduce. Spark provides additional speed for certain analytics and is the foundation
for other power tools such as Shark (SQL driven data warehousing), Spark Streaming (streaming
applications), GraphX (graph systems) and MLlib (machine learning).

EMR features Amazon EMR runtime for Apache Spark, a performance-optimized runtime
environment for Apache Spark that is active by default on Amazon EMR clusters. Amazon EMR
runtime for Apache Spark can be over 3x faster than clusters without the EMR runtime, and
has 100% API compatibility with standard Apache Spark. This improved performance means
your workloads run faster and saves you compute costs, without making any changes to your
applications.

For more information, see Apache Spark on Amazon EMR .

HBase

HBase is an open-source, non-relational, distributed database modeled after Google's Bigtable.

It was developed as part of Apache Software Foundation's Hadoop project and runs on top of

Interfaces 19
Big Data Analytics Options on AWS AWS Whitepaper

Hadoop Distributed File System (HDFS) to provide BigTable-like capabilities for Hadoop. HBase
provides you a fault-tolerant, efficient way of storing large quantities of sparse data using column-
based compression and storage. In addition, HBase provides fast lookup of data, because data is
stored in-memory instead of on disk.

HBase is optimized for sequential write operations, and it is highly efficient for batch inserts,
updates, and deletes. HBase works seamlessly with Hadoop, sharing its file system and serving
as a direct input and output to Hadoop jobs. HBase also integrates with Apache Hive, enabling
SQL-like queries over HBase tables, joins with Hive-based tables, and support for Java Database
Connectivity (JDBC). With Amazon EMR, you can back up HBase to Amazon S3 (full or incremental,
manual or automated) and you can restore from a previously created backup.

For more information, see Apache HBase in the Amazon EMR Release Guide.

Presto

Presto is an open-source distributed SQL query engine optimized for low-latency, ad hoc analysis
of data. It supports the ANSI SQL standard, including complex queries, aggregations, joins, and
window functions. Presto can process data from multiple data sources, including HDFS and
Amazon S3.

Hudi

Apache Hudi is an open-source data management framework used to simplify incremental data
processing and data pipeline development by providing record-level insert, update, upsert, and
delete capabilities. Upsert refers to the ability to insert records into an existing dataset if they do
not already exist or to update them if they do. By efficiently managing how data is laid out in S3,
Hudi allows data to be ingested and updated in near-real-time. Hudi carefully maintains metadata
of the actions performed on the dataset to help ensure that the actions are atomic and consistent.
Hudi is integrated with Apache Spark, Apache Hive, and Presto. In Amazon EMR release versions
6.1.0 and later, Hudi is also integrated with Trino (PrestoSQL).

Kinesis Connector

The Kinesis Connector enables EMR to directly read and query data from Kinesis Data Streams.
You can perform batch processing of Kinesis streams using existing Hadoop tools such as Hive, Pig,
MapReduce, Hadoop Streaming, and Cascading. Some use cases enabled by this integration are:

• Streaming log analysis — You can analyze streaming web logs to generate a list of top ten error
types every few minutes by Region, browser, and access domains.

Interfaces 20
Big Data Analytics Options on AWS AWS Whitepaper

• Complex data processing workflows — You can join Kinesis stream with data stored in Amazon
S3, Dynamo DB tables, and HDFS. You can write queries that join clickstream data from Kinesis
with advertising campaign information stored in a DynamoDB table to identify the most
effective categories of ads that are displayed on particular websites.
• Ad-hoc queries — You can periodically load data from Kinesis into HDFS and make it available as
a local Impala table for fast, interactive, analytic queries.

Other third-party tools

Amazon EMR also supports a variety of other popular applications and tools in the Hadoop system,
such as R (statistics), Mahout (machine learning), Ganglia (monitoring), Accumulo (secure NoSQL
database), Hue (user interface to analyze Hadoop data), HCatalog (table and storage management),
and more.

Additionally, you can install your own software on top of Amazon EMR to help solve your business
needs. AWS provides the ability to quickly move large amounts of data from S3 to HDFS, from
HDFS to S3, and between S3 buckets using Amazon EMR’s S3DistCp, an extension of the open
source tool DistCp that uses MapReduce to efficiently move large amounts of data.

You can optionally use the EMR File System (EMRFS), an implementation of HDFS which allows
Amazon EMR clusters to store data on S3. You can enable S3 server-side and client-side encryption.

EMR Studio

EMR Studio is an integrated development environment (IDE) that makes it easy for data scientists
and data engineers to develop, visualize, and debug data engineering and data science applications
written in R, Python, Scala, and PySpark.

EMR Studio provides fully managed Jupyter Notebooks, and tools like Spark UI and YARN Timeline
Service to simplify debugging. Data scientists and analysts can install custom kernels and libraries,
collaborate with peers using code repositories such as GitHub and BitBucket, or run parameterized
notebooks as part of scheduled workflows using orchestration services like Apache Airflow or
Amazon Managed Workflows for Apache Airflow. Administrators can set up EMR Studio such that
analysts can run their applications on existing EMR clusters, or create new clusters using pre-
defined AWS CloudFormation templates for EMR.

Anti-patterns
Amazon EMR has the following anti-pattern:

Anti-patterns 21
Big Data Analytics Options on AWS AWS Whitepaper

• Small datasets – Amazon EMR is built for massive parallel processing; if your data set is small
enough to run quickly on a single machine, in a single thread, the added overhead to map and
reduce jobs may not be worth it for small datasets that can easily be processed in memory on a
single system. Amazon EMR or a relational database running on Amazon EMR may be a better
option for workloads with stringent requirements.

AWS Glue
AWS Glue is a serverless data integration service that makes it easy to discover, prepare, and
combine data for analytics, machine learning, and application development. AWS Glue provides
all of the capabilities needed for data integration. It uses both visual and code-based interfaces to
make data integration easier.

Users can easily find and access data using the AWS Glue Data Catalog. Data engineers and ETL
developers can visually create, run, and monitor ETL workflows with a few clicks in AWS Glue
Studio. Data analysts and data scientists can use AWS Glue DataBrew to visually enrich, clean, and
normalize data without writing code. With AWS Glue Elastic Views, application developers can
use familiar Structured Query Language (SQL) to combine and replicate data across different data
stores. You pay only for the resources consumed while your jobs are running.

Ideal usage patterns

AWS Glue is designed to easily prepare data for extract, transform, and load (ETL) jobs. Using AWS
Glue gives you the following benefits:

• Data discovery
• Automatic schema discovery, using AWS Glue crawlers
• Manage and enforce schemas for data streams with AWS Glue Schema Registry
• Data Catalog
• The AWS Glue Data Catalog is a central repository to store structural and operational
metadata for all your data assets. For a given dataset, you can store its table definition,
physical location, add business relevant attributes, as well as track how this data has changed
over time.
• The AWS Glue Data Catalog is Apache Hive Metastore-compatible and is a drop-in replacement
for the Apache Hive Metastore for Big Data applications running on Amazon EMR, and third-
party applications such as Databricks.

AWS Glue 22
Big Data Analytics Options on AWS AWS Whitepaper

• Data transformation
• Visually transform data with a drag and drop interface using AWS Glue Studio. AWS Glue
automatically generates the reusable and portable code using familiar technology – Python (or
Scala) and Spark.
• Serverless streaming ETL jobs in AWS Glue continuously consume data from streaming sources
including Amazon Kinesis and Amazon MSK, clean and transform it in-flight, and make it
available for analysis in seconds in your target data store.
• Data replication
• AWS Glue Elastic Views enables you to create views over data stored in multiple types of AWS
data stores, and materialize the views in a target data store of your choice by writing queries in
PartiQL, an open-source SQL-compatible query language.
• Data preparation
• Deduplicate and cleanse data with built-in machine learning. The FindMatches feature
deduplicates and finds records that are imperfect matches of each other.
• Normalize data without code using a visual interface. AWS Glue DataBrew provides an
interactive, point-and-click visual interface for users like data analysts and data scientists to
clean and normalize data without writing code. Choose from over 250 built-in transformations.
• Integration
• Integration with data access services like Amazon Athena, Amazon EMR, and Amazon Redshift.
Also with third parties.
• Serverless
• No infrastructure to provision or manage.

Cost model

With AWS Glue jobs (crawler and ETL), AWS Glue DataBrew jobs, and AWS Glue Elastic Views,
you pay an hourly rate, billed by the second with a 1-minute minimum billing duration. For the
AWS Glue Data Catalog, you pay a simple monthly fee for storing and accessing the metadata.
The first million objects stored are free, and the first million accesses are free. If you provision a
development endpoint to interactively develop your ETL code, you pay an hourly rate, billed per
minute. AWS Glue DataBrew interactive sessions bill for the total number of the sessions used.
Each session is 30 minutes. A session is initiated when you open a DataBrew project.

See AWS Glue pricing for more details.

Cost model 23
Big Data Analytics Options on AWS AWS Whitepaper

Performance
AWS Glue uses a scale-out Apache Spark environment to load your data into its destination. You
can simply specify the number of Data Processing Units (DPUs) that you want to allocate to your
ETL job. An AWS Glue ETL job requires a minimum of 2 DPUs. By default, AWS Glue allocates 10
DPUs to each ETL job. Additional DPUs can be added to increase the performance of your ETL job.
Multiple jobs can be triggered in parallel or sequentially by triggering them on a job completion
event. You can also trigger one or more AWS Glue jobs from an external source such as AWS Step
Functions or Amazon Managed Workflows for Apache Airflow.

Durability and availability

AWS Glue connects to the data source of your preference, whether it is in an Amazon S3 file, an
Amazon RDS table, or another set of data. As a result, all your data is stored and available as it
pertains to that data stores durability characteristics. The AWS Glue service provides status of each
job and pushes all notifications to Amazon CloudWatch events. You can setup SNS notifications
using CloudWatch actions to be informed of job failures or completions.

Scalability and elasticity

AWS Glue provides a managed ETL service that runs on a Serverless Apache Spark environment.
This enables you to focus on your ETL job and not worry about configuring and managing the
underlying compute resources. AWS Glue works on top of the Apache Spark environment to
provide a scale-out run environment for your data transformation jobs.

Interfaces
AWS Glue provides a number of ways to populate metadata into the AWS Glue Data Catalog.
AWS Glue crawlers scan various data stores you own to automatically infer schemas and partition
structure and populate the AWS Glue Data Catalog with corresponding table definitions and
statistics. You can also schedule crawlers to run periodically so that your metadata is always up-to-
date and in-sync with the underlying data.

Alternately, you can add and update table details manually by using the AWS Glue Console or by
calling the API. You can also run Hive DDL statements via the Amazon Athena Console or a Hive
client on an Amazon EMR cluster.

Finally, if you already have a persistent Apache Hive Metastore, you can perform a bulk import of
that metadata into the AWS Glue Data Catalog by using the import script.

Performance 24
Big Data Analytics Options on AWS AWS Whitepaper

Anti-patterns

AWS Glue has the following anti-patterns:

• Multiple ETL engines – AWS Glue ETL jobs are Spark-based. If your use case requires you to use
an engine other than Apache Spark or if you want to run a heterogeneous set of jobs that run on
a variety of engines like Hive or Pig, Amazon EMR is a better choice.
• Configurable Spark environment – AWS Glue is a fully managed service. While you can change
some configuration parameters in your cluster, if your use case requires extensive configuration
changes, Amazon EMR or Amazon EMR on EKS is a better choice.

AWS Lake Formation

AWS Lake Formation is an integrated data lake service that makes it easy for you to ingest,
clean, catalog, transform, and secure your data and make it available for analysis and machine
learning. Lake Formation gives you a central console where you can discover data sources, set up
transformation jobs to move data to an S3 data lake, remove duplicates and match records, catalog
data for access by analytic tools, configure data access and security policies, and audit and control
access from AWS analytic and machine learning services.

Lake Formation automatically manages access to the registered data in S3 via services including
AWS Glue, Amazon Athena, Amazon Redshift, Amazon EMR Notebooks and Zeppelin notebooks
with Apache Spark, and Amazon QuickSight to ensure compliance with your defined policies. If
you’ve set up transformation jobs spanning AWS services, Lake Formation configures the flows,
centralizes their orchestration, and lets you monitor the running of your jobs. With Lake Formation,
you can configure and manage your data lake without manually integrating multiple underlying
AWS services.

Ideal usage patterns

AWS Lake Formation helps you collect and catalog data from databases and object storage,
move the data into your new S3 data lake, clean and classify your data using machine learning
algorithms, and secure access to your sensitive data. Your users can access a centralized data
catalog which describes available datasets and their appropriate usage. Using AWS Lake Formation
gives you the following benefits:

• Build data lakes quickly

Anti-patterns 25
Big Data Analytics Options on AWS AWS Whitepaper

• Lake Formation blueprints simplify the deployment of ingestion workflow via a simple
interface.
• Simply point Lake Formation at your data sources, and Lake Formation crawls those sources
and moves the data into your new S3 data lake.
• Lake Formation has built-in machine learning to deduplicate and find matching records (two
entries that refer to the same thing) to increase data quality.
• Simplify security management
• Centrally define security, governance, and auditing policies in one place.
• You can define security policy-based rules for your users and applications by role in
Lake Formation, and integration with AWS Identity and Access Management (AWS IAM)
authenticates those users and roles.
• With tag-based access control, data stewards can define a tag ontology based on data
classification, and grant access based on tags to IAM principals and SAML principals or groups.
• Enforce encryption leveraging the encryption capabilities of S3 for data in your data lake.
• Provides comprehensive audit logs with CloudTrail to monitor access and show compliance
with centrally defined policies.
• Provide self-service access to data
• Label your data with business metadata. Designate data owners, such as data stewards and
business units, by adding a field in table properties as “custom attributes”.
• Specify, grant, and revoke permissions on tables defined in the central Data Catalog. The same
Data Catalog is available for multiple accounts, groups, and services.
• Search for relevant data by name, contents, sensitivity, or other any other custom labels you
have defined.
• Integration
• Integration with data access services like Amazon Athena, Amazon EMR, Amazon Redshift, and
Amazon QuickSight.
• Serverless
• No infrastructure to provision or manage.

Cost model

There is no additional charge for using features in Lake Formation. Lake Formation helps you build
and manage data lakes where your data in stored in S3. It builds on capabilities available in AWS

Cost model 26
Big Data Analytics Options on AWS AWS Whitepaper

Glue and uses the AWS Glue Data Catalog, jobs, and crawlers. It also integrates with services like
AWS CloudTrail, AWS IAM, Amazon CloudWatch, Amazon Athena, Amazon EMR, Amazon Redshift,
and others.

Standard usage rates for these services will apply based on pricing for these services.

Performance
AWS Lake Formation uses AWS Glue as a transformation engine. Refer to the AWS Lake Formation:
How it Works section of the Lake Formation Developer Guide.

Durability and availability

AWS Lake Formation leverage S3 as data lake storage, designed to provide 99.999999999% of
data durability of objects over a given year. It creates and stores copies of all S3 objects across
multiple systems. This means your data is available when you need it, and protected against
failures, errors, and threats. The AWS Glue service is fully managed. It provides status for each job
and pushes all notifications to Amazon CloudWatch Events. You can setup SNS notifications using
CloudWatch actions to be informed of job failures or completions.

Scalability and elasticity

AWS Lake Formation uses AWS Glue, which provides a managed ETL service that runs on a
Serverless Apache Spark environment. This enables you to focus on your ETL job and not worry
about configuring and managing the underlying compute resources. AWS Glue works on top of the
Apache Spark environment to provide a scale-out run environment for your data transformation
jobs.

Interfaces
Lake Formation provides a console to manage your data lake locations, data catalog, access
privileges, and deploy blueprints. It also provides APIs and a CLI to integrate Lake Formation
functionality into your custom applications. Java and C++ SDKs are available to enable you to
integrate your own data engines with Lake Formation. Lake Formation manages query access
from AWS Glue, Athena, Amazon Redshift, EMR Notebooks, and Zeppelin notebooks for EMR with
Apache Spark through a unified security model and permissions.

You can use third-party business applications like Tableau and Looker to connect to your AWS
data sources through services like Athena or Amazon Redshift. Access to data is managed by the

Performance 27
Big Data Analytics Options on AWS AWS Whitepaper

underlying data catalog, so regardless of which application you use, you are assured that access to
your data is governed and controlled.

Anti-patterns

AWS Lake Formation has the following anti-patterns:

• Managing unstructured data – AWS Lake Formation does not manage security for unstructured
objects as document (PDF, Word), images, or videos.

• Business catalog – Lake Formation helps label your data with business metadata and designate
data owners such as data stewards and business units by adding a field in table properties as
“custom attributes”. For advanced use cases, consider integrating a Partner solution from the
AWS Marketplace.

• Managing permission for open source and third-party big data components — AWS Lake
Formation is not integrated with Presto, HBase and other EMR components not specified in the
previous Interface section of this chapter. It is also not integrated with third party applications
such as Trino or Databricks.

• Real-time analytics – AWS Lake Formation is not integrated with Amazon Kinesis or
OpenSearch Service. If you need to manage a real-time analytics pipeline or dashboard, you will
have to manage privileges in a traditional way.

Amazon Machine Learning

AWS offers the broadest and deepest set of machine learning services and supporting cloud
infrastructure, putting machine learning in the hands of every developer, data scientist, and expert
practitioner. When you build an ML-based workload in AWS, you can choose from three different
levels of ML services to balance speed-to-market with level of customization and ML skill level:

• Artificial Intelligence (AI) services

• ML services

• ML frameworks and infrastructure

The AI Services level provides fully managed services that enable you to quickly add ML capabilities
to your workloads using API calls. This gives you the ability to build powerful, intelligent
applications with capabilities such as computer vision, speech, natural language, chatbots,

Anti-patterns 28
Big Data Analytics Options on AWS AWS Whitepaper

predictions, and recommendations. Services at this level are based on pre-trained or automatically
trained machine learning and deep learning models, so you don’t need ML knowledge to use them.

You can use:

• Amazon Translate to translate or localize text content

• Amazon Polly for text-to-speech conversion
• Amazon Lex for building conversational chat bots
• Amazon Comprehend to extract insights and relationships from unstructured data
• Amazon Forecast to build accurate forecasting models
• Amazon Fraud Detector to identify potentially fraudulent online activities,
• Amazon CodeGuru to automate code reviews and to identify most expensive lines of code
• Amazon Textract to extract text and data from documents automatically
• Amazon Rekognition to add image and video analysis to your applications
• Amazon Kendra to reimagines enterprise search for your websites and applications
• Amazon Personalize for real-time personalized recommendations
• Amazon Transcribe to add speech to text capabilities to your applications

The ML Services level provides managed services and resources for machine learning to developers,
data scientists, and researchers.

• Amazon SageMaker enables developers and data scientists to quickly and easily build, train, and
deploy ML models at any scale.
• Amazon SageMaker Ground Truth helps you build highly accurate ML training datasets quickly.
• Amazon SageMaker Studio is the first integrated development environment for machine learning
to build, train, and deploy ML models at scale.
• Amazon SageMaker Autopilot automatically builds, trains, and tunes the best ML models based
on your data, while enabling you to maintain full control and visibility.
• Amazon SageMaker JumpStart helps you quickly and easily get started with ML.
• Amazon SageMaker Data Wrangler reduces the time it takes to aggregate and prepare data for
ML from weeks to minutes.
• Amazon SageMaker Feature Store is a fully managed, purpose-built repository to store, update,
retrieve, and share ML features.

Amazon Machine Learning 29

Big Data Analytics Options on AWS AWS Whitepaper

• Amazon SageMaker Clarify provides ML developers with greater visibility into your training data
and models so you can identify and limit bias and explain predictions.
• Amazon SageMaker Debugger optimizes ML models with real-time monitoring of training
metrics and system resources.
• Amazon SageMaker's distributed training libraries automatically split large deep learning models
and training datasets across AWS graphics processing unit (GPU) instances in a fraction of the
time it takes to do manually.
• Amazon SageMaker Pipelines is the first purpose-built, easy-to-use continuous integration and
continuous delivery (CI/CD) service for ML.
• Amazon SageMaker Neo enables developers to train ML models once, and then run them
anywhere in the cloud or at the edge.

The ML Frameworks and Infrastructure level is intended for expert ML practitioners. These people
are comfortable with designing their own tools and workflows to build, train, tune, and deploy
models, and are accustomed to working at the framework and infrastructure level. In AWS, you
can use open-source ML frameworks such as TensorFlow, PyTorch, and Apache MXNet. The Deep
Learning AMI and Deep Learning Containers in this level have multiple ML frameworks preinstalled
that are optimized for performance. This optimization means that they are always ready to be
launched on powerful, ML-optimized compute infrastructure, such as Amazon EC2 P3 and P3dn
instances, that provides a boost of speed and efficiency to ML workloads.

Amazon ML can create ML models based on data stored in S3, Amazon Redshift, or Amazon RDS.
Built-in wizards guide you through the steps of interactively exploring your data to training the ML
model, evaluate the model quality, and adjust outputs to align with business goals. After a model is
ready, you can request predictions in batches, or using the low-latency real-time API.

Workloads often use services from multiple levels of the ML stack. Depending on the business
use case, services and infrastructure from the different levels can be combined to satisfy multiple
requirements and achieve multiple business goals. For example, you can use AI services for
sentiment analysis of customer reviews on your retail website, and use managed ML services to
build a custom model using your own data to predict future sales.

Ideal usage patterns

Amazon ML is ideal for discovering patterns in your data and using these patterns to create ML
models that can generate predictions on new, unseen data points. For example, you can:

Ideal usage patterns 30

Big Data Analytics Options on AWS AWS Whitepaper

• Enable applications to flag suspicious transactions – Build an ML model that predicts whether a
new transaction is legitimate or fraudulent.
• Forecast product demand – Input historical order information to predict future order quantities.
• Media intelligence – Maximize the value of media content by adding machine learning to media
workflows such as search and discovery, content localization, compliance, monetization, and
more.
• Personalize application content – Predict which items a user will be most interested in, and
retrieve these predictions from your application in real-time.
• Predict user activity – Analyze user behavior to customize your website and provide a better
user experience.
• Listen to social media – Ingest and analyze social media feeds that potentially impact business
decisions.
• Intelligent contact center – Enhance your customer service experience and reduce costs by
integrating ML into your contact center.
• Intelligent search – Boost business productivity and customer satisfaction by delivering accurate
and useful information faster from siloed and unstructured information sources across the
organization.

Cost model

With Amazon Machine Learning services, you pay only for what you use. There are no minimum
fees and no upfront commitments.

AWS pre-trained AI Services cost model varies depending upon the AI service you are planning to
integrate with your applications. For details, see pricing details of the respective AI services:

• Amazon Comprehend
• Amazon Forecast
• Amazon Fraud Detector
• Amazon Translate
• Amazon CodeGuru
• Amazon Textract
• Amazon Rekognition

Cost model 31
Big Data Analytics Options on AWS AWS Whitepaper

• Amazon Polly
• Amazon Lex
• Amazon Kendra
• Amazon Personalize
• Amazon Transcribe

With Amazon SageMaker, you have two choices to pay, and you only pay for what you use:

• On-demand pricing is billed by the second, with no minimum fees and no upfront
commitments.
• SageMaker Savings Plans offer a flexible, usage-based pricing model in exchange for a
commitment to a consistent amount of usage. For details, see Amazon SageMaker pricing.

The ML Frameworks and Infrastructure level is intended for expert ML practitioners. ML training
and inference workloads can exhibit characteristics that are steady state (such as hourly batch
tagging of photos for a large population), spiky (such as kicking off new training jobs or search
recommendations during promotional periods), or both. AWS has pricing options and solutions to
help you optimize your infrastructure performance and costs.

For details, see the AWS Machine Learning Infrastructure.

Performance

The time it takes to create models and to request predictions from ML models depends on the
number of input data records, and the types and distribution of attributes that you specify. There
are a number of principles designed to help increase performance specifically for ML workloads:

• Optimize compute for your ML workload — Most ML workloads are very compute-intensive,
because large amounts of vector multiplications and additions need to be performed on a
multitude of data and parameters. Especially in Deep Learning, there is a need to scale to
chipsets that provide larger queue depth, higher Arithmetic Logic Units and Register counts, to
allow for massively parallel processing. Because of that, GPUs are the preferred processor type to
train a Deep Learning model.
• Define latency and network bandwidth performance requirements for your models —
Some of your ML applications might require near-instantaneous inference results to satisfy
your business requirements. Offering the lowest latency possible may require the removal of

Performance 32
Big Data Analytics Options on AWS AWS Whitepaper

costly round trips to the nearest API endpoints. This reduction in latency can be achieved by
running the inference directly on the device itself. This is known as Machine Learning at the Edge.
A common use case for such a requirement is predictive maintenance in factories. This form
of low latency and near-real-time inference at the edge allows for early indications of failure,
potentially mitigating costly repairs of machinery before the failure actually happens.

• Continuously monitor and measure system performance — The practice of identifying and
regularly collecting key metrics related to the building, training, hosting, and running predictions
against a model ensures that you are able to continuously monitor the holistic success across
key evaluation criteria. To validate system level resources used to support the phases of ML
workloads, it’s key to continuously collect and monitor system level resources such as compute,
memory, and network. Requirements for ML workloads change in different phases, as training
jobs are more memory intensive, while inference jobs are more compute intensive.

Durability and availability

There are key principles designed to help increase availability and durability specifically for ML
workloads:

• Manage changes to model inputs through automation — ML workloads have additional

requirements to manage changes to the data that is used to train a model to be able to recreate
the exact version of a model in the event of failure or human error. Managing versions and
changes through automation provides for a reliable and consistent recovery method.

• Train once and deploy across environments — When deploying the same version of an ML
model across multiple accounts or environments, the same practice of build once that is applied
to application code should be applied for model training. A specific version of a model should
only be trained once and the output model artifacts should be utilized to deploy across multiple
environments to avoid bringing in any unexpected changes to the model across environments.

Scalability and elasticity

There are key principles designed to help increase availability and durability specifically for ML
workloads:

• Identify the end-to-end architecture and operational model early — Early in the ML
development lifecycle, identify the end-to-end architecture and operational model for model
training and hosting. This allows for early identification of architectural and operational

Durability and availability 33

Big Data Analytics Options on AWS AWS Whitepaper

considerations that will be required for the development, deployment, management and
integration of ML workloads.
• Version machine learning inputs and artifacts — Versioned inputs and artifacts enable you to
recreate artifacts for previous versions of your ML workload. Version inputs are used to create
models, including training data and training source code, and model artifacts.

• Automate machine learning deployment pipelines — Minimize human touch points in ML

deployment pipelines to ensure that ML models are consistently and repeatedly deployed
using a pipeline that defines how models move from development to production. Identify and
implement a deployment strategy that satisfies the requirements of your use case and business
problem. If required, include human quality gates in your pipeline to have humans evaluate if a
model is ready to deploy to a target environment.

Interfaces

Creating a data source is as simple as adding your data to S3. To ingest data, you can use AWS
Direct Connect to privately connect your data center directly to an AWS Region. To physically
transfer petabytes of data in batches, use AWS Snowball, or, if you have exabytes of data, use
AWS Snowmobile. You can integrate your existing on-premises storage using Storage Gateway, or
add cloud capabilities using AWS Snowball Edge. Use Amazon Data Firehose to collect and ingest
multiple streaming-data sources.

Anti-patterns

Amazon ML has the following anti-patterns:

• Big data processing – Data processing activities are well suited for tools like Apache Spark,
which provide SQL support for data discovery, among other useful utilities. On AWS, Amazon
EMR facilitates the management of Spark clusters, and enables capabilities like elastic scaling
while minimizing costs through Spot Instance pricing.

• Real-time analytics – Collecting, processing, and analyzing the streaming data to respond in
real time are well suited for tools like Kafka. On AWS, Amazon Kinesis makes it easy to collect,
process, and analyze real-time, streaming data so you can get timely insights and react quickly to
new information and Amazon MSK is a fully managed service that makes it easy for you to build
and run applications that use Apache Kafka to process streaming data. Apache Kafka is an open-
source platform for building real-time streaming data pipelines and applications.

Interfaces 34
Big Data Analytics Options on AWS AWS Whitepaper

Amazon DynamoDB
Amazon DynamoDB is a fast, fully-managed NoSQL database service that makes it simple
and cost effective to store and retrieve any amount of data, and serve any level of request
traffic. DynamoDB helps offload the administrative burden of operating and scaling a highly-
available distributed database cluster. This storage alternative meets the latency and throughput
requirements of highly demanding applications by providing single-digit millisecond latency and
predictable performance with seamless throughput and storage scalability.

DynamoDB stores structured data in tables, indexed by primary key, and allows low-latency read
and write access to items ranging from 1 byte up to 400 KB. DynamoDB supports three data types
(number, string, and binary), in both scalar and multi-valued sets. It supports document stores such
as JSON, XML, or HTML in these data types. Tables do not have a fixed schema, so each data item
can have a different number of attributes. The primary key can either be a single-attribute hash key
or a composite hash-range key.

DynamoDB offers both global and local secondary indexes provide additional flexibility for
querying against attributes other than the primary key. DynamoDB provides both eventually-
consistent reads (by default), and strongly-consistent reads (optional), as well as implicit item-level
transactions for item put, update, delete, conditional operations, and increment/decrement.

With DynamoDB, you can create database tables that can store and retrieve any amount of data
and serve any level of request traffic. You can scale up or scale down your tables' throughput
capacity without downtime or performance degradation.

DynamoDB provides on-demand backup capability. It allows you to create full backups of your
tables for long-term retention and archival for regulatory compliance needs. You can create on-
demand backups and enable point-in-time recovery for your Amazon DynamoDB tables. Point-in-
time recovery helps protect your tables from accidental write or delete operations. With point-in-
time recovery, you can restore a table to any point in time during the last 35 days.

DynamoDB enables you to delete expired items from tables automatically to help you reduce
storage usage and the cost of storing data that is no longer relevant. For more information, see
Expiring Items By Using DynamoDB Time to Live (TTL).

DynamoDB is integrated with other services, such as Amazon EMR, Amazon Redshift, AWS Data
Pipeline, and S3 for analytics, data warehouse, data import/export, backup, and archive.

Amazon DynamoDB 35
Big Data Analytics Options on AWS AWS Whitepaper

Ideal usage patterns

DynamoDB is ideal for existing or new applications that need a flexible NoSQL database with low
read and write latencies, and the ability to scale storage and throughput up or down as needed
without code changes or downtime.

Common use cases include:

• Mobile apps
• Gaming
• Digital ad serving
• Live voting
• Audience interaction for live events
• Sensor networks
• Log ingestion
• Access control for web-based content
• Metadata storage for Amazon S3 objects
• Ecommerce shopping carts
• Web session management

Many of these use cases require a highly available and scalable database because downtime or
performance degradation has an immediate negative impact on an organization’s business.

Cost model

With DynamoDB, you pay only for what you use and there is no minimum fee.

DynamoDB charges for reading, writing, and storing data in your DynamoDB tables, along with any
optional features you choose to enable. DynamoDB has two capacity modes and those come with
specific billing options for processing reads and writes on your tables: on-demand and provisioned.

• With on-demand capacity mode, DynamoDB charges you for the data reads and writes
your application performs on your tables. You do not need to specify how much read and
write throughput you expect your application to perform, because DynamoDB instantly
accommodates your workloads as they ramp up or down.

Ideal usage patterns 36

Big Data Analytics Options on AWS AWS Whitepaper

• With provisioned capacity mode, you specify the number of reads and writes per second that
you expect your application to require. You can use auto scaling to automatically adjust your
table’s capacity based on the specified utilization rate to ensure application performance while
reducing costs.

New customers can start using DynamoDB for free as part of the AWS Free Usage Tier. For more
information, see Amazon DynamoDB pricing.

Performance

DynamoDB is a key-value and document database that can support tables of virtually any size with
horizontal scaling. This enables DynamoDB to scale to more than ten trillion requests per day with
peaks greater than 20 million requests per second, over petabytes of storage.

DynamoDB supports both key-value and document data models. This enables DynamoDB to have
a flexible schema, so each row can have any number of columns at any point in time. This enables
you to easily adapt the tables as your business requirements change, without having to redefine
the table schema as you would in relational databases.

DynamoDB Accelerator (DAX) is an in-memory cache that delivers fast read performance for your
tables at scale by enabling you to use a fully managed in-memory cache. Using DAX, you can
improve the read performance of your DynamoDB tables by up to ten times—taking the time
required for reads from milliseconds to microseconds, even at millions of requests per second.

DynamoDB global tables replicate your data automatically across your choice of AWS Regions,
and automatically scale capacity to accommodate your workloads. With global tables, your
globally distributed applications can access data locally in the selected Regions to get single-digit
millisecond read and write performance.

Amazon Kinesis Data Streams for DynamoDB captures item-level changes in your DynamoDB
tables as a Kinesis data stream. This feature enables you to build advanced streaming applications
such as real-time log aggregation, real-time business analytics, and IoT data capture. Through
Kinesis Data Streams, you also can use Amazon Data Firehose to deliver DynamoDB data
automatically to other AWS services.

AWS Glue Elastic Views supports DynamoDB as a source to combine and replicate data
continuously across multiple databases in near-real-time.

Performance 37
Big Data Analytics Options on AWS AWS Whitepaper

Durability and availability

DynamoDB automatically spreads the data and traffic for your tables over a sufficient number of
servers to handle your throughput and storage requirements, while maintaining consistent and fast
performance. All of your data is stored on solid-state disks (SSDs) and is automatically replicated
across multiple Availability Zones in an AWS Region, providing built-in high availability and data
durability. You can use global tables to keep DynamoDB tables in sync across AWS Regions.

Amazon DynamoDB Streams captures all data activity that happens on your table and enables
you to set up Regional replication from one geographic Region to another to provide even greater
availability.

Scalability and elasticity

DynamoDB is both highly scalable and elastic. There is no limit to the amount of data that you
can store in a DynamoDB table, and the service automatically allocates more storage as you store
more data using the DynamoDB write API operations. Data is automatically partitioned and re-
partitioned as needed, while the use of SSDs provides predictable low-latency response times at
any scale. The service is also elastic, in that you can simply “dial-up” or “dial-down” the read and
write capacity of a table as your needs change.

When you create a DynamoDB table, auto scaling is the default capacity setting, but you can also
enable auto scaling on any table that does not have it active. Behind the scenes, DynamoDB auto
scaling uses a scaling policy in Application Auto Scaling. To configure auto scaling in DynamoDB,
you set the minimum and maximum levels of read and write capacity in addition to the target
utilization percentage. Auto scaling uses Amazon CloudWatch to monitor a table’s read and write
capacity metrics by creating CloudWatch alarms that track consumed capacity.

Interfaces

DynamoDB provides a low-level RESTAPI, as well as higher-level SDKs for Java, ET, and PHP that
wrap the low-level REST API and provide some object-relational mapping (ORM) functions. These
APIs provide both a management and data interface for DynamoDB. The API currently offers
operations that enable table management (creating, listing, deleting, and obtaining metadata) and
working with attributes (getting, writing, and deleting attributes; query using an index, and full
scan).

Durability and availability 38

Big Data Analytics Options on AWS AWS Whitepaper

While standard SQL isn’t available, you can use the DynamoDB select operation to create SQL-like
queries that retrieve a set of attributes based on criteria that you provide. You can also work with
DynamoDB using the console.

Anti-patterns
DynamoDB has the following anti-patterns:

• Prewritten application tied to a traditional relational database – If you are attempting to port
an existing application to the AWS Cloud and need to continue using a relational database, you
can use either Amazon RDS (Amazon Aurora, MySQL, PostgreSQL, Oracle, or SQL Server), or
one of the many pre-configured Amazon EC2 database AMIs. You can also install your choice of
database software on an EC2 instance that you manage.
• Joins or complex transactions – While many solutions are able to leverage DynamoDB to
support their users, it’s possible that your application may require joins, complex transactions,
and other relational infrastructure provided by traditional database platforms. If this is the case,
you may want to explore Amazon Redshift, Amazon RDS, or Amazon EC2 with a self-managed
database.
• Binary large objects (BLOB) data – If you plan on storing large (greater than 400 KB) BLOB data,
such as digital video, images, or music, you’ll want to consider Amazon S3. However, DynamoDB
can be used in this scenario for keeping track of metadata (such as item name, size, date created,
owner, location, and so on) about your binary objects.
• Large data with low I/O rate – DynamoDB uses SSD drives and is optimized for workloads with a
high I/O rate per GB stored. If you plan to store very large amounts of data that are infrequently
accessed, other storage options may be a better choice, such as Amazon S3.

Amazon Redshift
Amazon Redshift is a fast, fully-managed, petabyte-scale data warehouse service that makes
it simple and cost-effective to analyze all your data efficiently using your existing business
intelligence tools. It is optimized for data sets ranging from a few hundred gigabytes to a petabyte
or more, and is designed to cost less than a tenth of the cost of most traditional data warehousing
solutions.

Amazon Redshift delivers fast query and I/O performance for virtually any size dataset by using
columnar storage technology while parallelizing and distributing queries across multiple nodes.
It automates most of the common administrative tasks associated with provisioning, configuring,

Anti-patterns 39
Big Data Analytics Options on AWS AWS Whitepaper

monitoring, backing up, and securing a data warehouse, making it easy and inexpensive to manage
and maintain. This automation enables you to build petabyte-scale data warehouses in minutes
instead of weeks or months.

Amazon Redshift Spectrum enables you to run queries against exabytes of unstructured data
in Amazon S3, with no loading or ETL required. When you issue a query, it goes to the Amazon
Redshift SQL endpoint, which generates and optimizes a query plan. Amazon Redshift determines
what data is local and what is in S3, generates a plan to minimize the amount of S3 data that
needs to be read, and then requests Redshift Spectrum workers out of a shared resource pool to
read and process the data from S3.

With the federated query feature in Amazon Redshift, you can query and analyze data across
operational databases, data warehouses, and data lakes. It also enables you to integrate queries
from Amazon Redshift on live data in external databases with queries across your Amazon Redshift
and S3 environments. With cross-database queries, you can query data from any database in the
Amazon Redshift cluster, regardless of which database you are connected to. Cross-database
queries eliminate data copies and simplify your data organization to support multiple business
groups from the same data warehouse.

AQUA (Advanced Query Accelerator) is a new distributed and hardware-accelerated cache

that enables Amazon Redshift to run up to ten times faster than other enterprise cloud data
warehouses by automatically boosting certain types of queries. AQUA is included with certain node
types in the Amazon Redshift RA3 cluster.

There is also a feature called Data sharing which enables you to share data across Amazon Redshift
clusters without needing to manually copy or move it for read purposes. You can have live access
to data, and users can see the most up-to-date and consistent information as it’s updated in the
Amazon Redshift cluster.

Ideal usage patterns

Amazon Redshift is ideal for online analytical processing (OLAP) using your existing business
intelligence tools. Organizations are using Amazon Redshift to:

• Analyze global sales data for multiple products

• Store historical stock trade data
• Analyze ad impressions and clicks
• Aggregate gaming data

Ideal usage patterns 40

Big Data Analytics Options on AWS AWS Whitepaper

• Analyze social trends

• Measure clinical quality, operation efficiency, and financial performance in health care
• Analyze data across the data lake (S3) and Amazon Redshift.

Cost model

An Amazon Redshift data warehouse cluster requires no long-term commitments or upfront

costs. This frees you from the capital expense and complexity of planning and purchasing data
warehouse capacity ahead of your needs. Charges are based on the size and number of nodes of
your cluster.

There is no additional charge for backup storage up to 100% of your provisioned storage. For
example, if you have an active cluster with 2 XL nodes for a total of 4 TB of storage, AWS provides
up to 4 TB of backup storage on Amazon S3 at no additional charge. Backup storage beyond the
provisioned storage size, and backups stored after your cluster is terminated, are billed at standard
Amazon S3 rates. There is no data transfer charge for communication between S3 and Amazon
Redshift.

For more information, see Amazon Redshift pricing.

Performance

Amazon Redshift uses a variety of innovations to obtain very high performance on data sets
ranging in size from hundreds of gigabytes to a petabyte or more. It uses columnar storage, data
compression, and zone maps to reduce the amount of I/O needed to perform queries.

Amazon Redshift has a massively parallel processing (MPP) architecture, parallelizing and
distributing SQL operations to take advantage of all available resources. The underlying hardware
is designed for high performance data processing, using local attached storage to maximize
throughput between the CPUs and drives, and a 10 GigE mesh network to maximize throughput
between nodes. Performance can be tuned based on your data warehousing needs: AWS offers
Dense Compute (DC) with SSD drives as well as Dense Storage (DS) options.

Durability and availability

Amazon Redshift automatically detects and replaces a failed node in your data warehouse cluster.
The data warehouse cluster is read-only until a replacement node is provisioned and added to

Cost model 41
Big Data Analytics Options on AWS AWS Whitepaper

the DB, which typically only takes a few minutes. Amazon Redshift makes your replacement node
available immediately and streams your most frequently accessed data from S3 first, to allow you
to resume querying your data as quickly as possible.

Additionally, your data warehouse cluster remains available in the event of a drive failure; because
Amazon Redshift mirrors your data across the cluster, it uses the data from another node to rebuild
failed drives. Amazon Redshift clusters reside within one Availability Zone, but if you wish to have a
multi-AZ set up for Amazon Redshift, you can set up a mirror and then self-manage replication and
failover.

Scalability and elasticity

With a few clicks in the console or an API call, you can change the number, or type, of nodes in your
data warehouse as your performance or capacity needs change. Amazon Redshift enables you to
start with a single 160 GB node and scale up to a petabyte or more of compressed user data using
many nodes. For more information, see Clusters and Nodes in Amazon Redshift in the Amazon
Redshift Management Guide.

While resizing, Amazon Redshift places your existing cluster into read-only mode, provisions a new
cluster of your chosen size, and then copies data from your old cluster to your new one in parallel.
During this process, you pay only for the active Amazon Redshift cluster. You can continue running
queries against your old cluster while the new one is being provisioned. After your data has been
copied to your new cluster, Amazon Redshift automatically redirects queries to your new cluster
and removes the old cluster.

Interfaces

Amazon Redshift has custom JDBC and ODBC drivers that you can download from the Connect
Client tab of the console, which enables you to use a wide range of familiar SQL clients. You
can also use standard PostgreSQL JDBC and ODBC drivers. For more information about Amazon
Redshift drivers, see Amazon Redshift and PostgreSQL.

There are numerous examples of validated integrations with many popular BI and ETL vendors.
Loads and unloads are attempted in parallel into each compute node to maximize the rate at which
you can ingest data into your data warehouse cluster as well as to and from S3 and DynamoDB.
You can easily load streaming data into Amazon Redshift using Amazon Data Firehose, enabling
near real-time analytics with existing business intelligence tools and dashboards you’re already
using today. Metrics for compute utilization, memory utilization, storage utilization, and read/write

Scalability and elasticity 42

Big Data Analytics Options on AWS AWS Whitepaper

traffic to your Amazon Redshift data warehouse cluster are available free of charge via the console
or CloudWatch API operations.

Anti-patterns
Amazon Redshift has the following anti-patterns:

• Small datasets – Amazon Redshift is built for parallel processing across a cluster. If your data
set is less than a hundred gigabytes, you won't get all the benefits that Amazon Redshift has to
offer, and Amazon RDS may be a better solution.
• Online transaction processing (OLTP) – Amazon Redshift is designed for data warehouse
workloads producing extremely fast and inexpensive analytic capabilities. If you require a fast
transactional system, you may want to choose a traditional relational database system built on
Amazon RDS or a NoSQL database offering, such as DynamoDB.
• Unstructured data – Data in Amazon Redshift must be structured by a defined schema, rather
than supporting arbitrary schema structure for each row. If your data is unstructured, you can
perform extract, transform, and load (ETL) on Amazon EMR to get the data ready for loading
into Amazon Redshift.
• BLOB data – If you plan on storing large binary files (such as digital video, images, or music), you
may want to consider storing the data in S3 and referencing its location in Amazon Redshift. In
this scenario, Amazon Redshift keeps track of metadata (such as item name, size, date created,
owner, location, and so on) about your binary objects, but the large objects themselves are
stored in S3.

Amazon OpenSearch Service

Amazon OpenSearch Service (OpenSearch Service) makes it easy to deploy, operate, and
scale OpenSearch Service for log analytics, full text search, application monitoring, and more.
OpenSearch Service is a fully managed service that delivers the OpenSearch Service easy-
to-use APIs and real-time capabilities along with the availability, scalability, and security
required by production workloads. The service offers built-in integrations with OpenSearch
Dashboards, Logstash, and AWS services including Amazon Data Firehose, AWS Lambda,
and Amazon CloudWatch so that you can go from raw data to actionable insights quickly.

It’s easy to get started with OpenSearch Service. You can set up and configure your OpenSearch
Service domain in minutes from the AWS Management Console. If you prefer programmatic access,
you can use the AWS CLI or the AWS SDKs. OpenSearch Service provisions all the resources for your

Anti-patterns 43
Big Data Analytics Options on AWS AWS Whitepaper

domain and launches it. The service automatically detects and replaces failed OpenSearch Service
nodes, reducing the overhead associated with self-managed infrastructure and OpenSearch Service
software.

OpenSearch Service enables you to scale your cluster via asingle API call, or a few clicks in the
console. With OpenSearch Service, you get direct access to the OpenSearch Service open-source
API, so that code and applications you’re already using with your existing OpenSearch Service
environments will work seamlessly.

In addition to X86 based instances, Amazon OpenSearch Service offers instances from the
Graviton2 instance family. The instance family includes general purpose, compute-optimized,
and memory-optimized instances for you to choose from. OpenSearch Service Service Graviton2
instances support OpenSearch Service versions 7.9 and above.

Ideal usage patterns

OpenSearch Service is ideal for querying and searching large amounts of data. Organizations can
use OpenSearch Service to do the following:

• Analyze activity logs, such logs for customer facing applications or websites
• Analyze CloudWatch Logs with OpenSearch Service
• Analyze product usage data coming from various services and systems
• Analyze social media sentiments, CRM data, and find trends for your brand and products
• Analyze data stream updates from other AWS services, such as Amazon Kinesis Data Streams and
Amazon DynamoDB
• Utilize a rich search and navigation experience.
• Monitor usage for mobile applications

Cost model

With Amazon OpenSearch Service, you pay only for what you use. There are no minimum fees
or upfront commitments. You are charged for OpenSearch Service instance hours, Amazon EBS
storage (if you choose this option), and standard data transfer fees.

You can get started with our free tier, which provides free usage of up to 750 hours per month of
a single-AZ t2.micro.elasticsearch or t2.small.elasticsearch instance and 10 GB per
month of optional Amazon EBS storage (Magnetic or General Purpose). With OpenSearch Service

Ideal usage patterns 44

Big Data Analytics Options on AWS AWS Whitepaper

Reserved Instances, you can reserve instances for a one- or three-year term and get significant
savings on usage costs compared to On-Demand Instances.

OpenSearch Service enables you to add data durability through automated and manual snapshots
of your cluster. OpenSearch Service provides storage space for automated snapshots free of charge
for each Amazon OpenSearch Service domain. Automated snapshots are retained for a period
of 14 days. Manual snapshots are charged according to Amazon S3 storage rates. Data transfer
for using the snapshots is free of charge. At the time of this writing, with Graviton2 instances for
OpenSearch Service, you can realize up to 44% price/performance improvement over previous
generation instances.

For more information, see Amazon OpenSearch Service Pricing.

UltraWarm is a tier for OpenSearch Service that provides a cost-effective way to store large
amounts of read-only data on OpenSearch Service. Rather than attached storage, UltraWarm
nodes use S3 and a sophisticated caching solution to improve performance. For indexes that you
are not actively writing to, query less frequently, and don't need the same performance from,
UltraWarm offers significantly lower costs per GiB of data. Because warm indices are read-only
unless you return them to hot storage, UltraWarm is best-suited to immutable data, such as logs.

Performance
Performance of OpenSearch Service depends on multiple factors including instance type,
workload, index, number of shards used, read replicas, and storage configurations – instance
storage or EBS storage (general purpose SSD). Indexes are made up of shards of data which can be
distributed on different instances in multiple Availability Zones.

Read replica of the shards are maintained by OpenSearch Service in a different Availability
Zone if zone awareness is checked. OpenSearch Service can use either the fast SSD instance
storage for storing indexes or multiple EBS volumes. A search engine makes heavy use of storage
devices and making disks faster will result in faster query and search performance. Leveraging
Graviton2 instances can also improve indexing throughput, indexing latency reduction, and
query performance, in comparison with the corresponding x86-based instances from the current
generation.

Durability and availability

You can configure your OpenSearch Service domains for high availability by enabling the Zone
Awareness option either at domain creation time or by modifying a live domain. When Zone

Performance 45
Big Data Analytics Options on AWS AWS Whitepaper

Awareness is enabled, OpenSearch Service distributes the instances that support the domain across
two different Availability Zones. Then, if you enable replicas in OpenSearch Service, the instances
are automatically distributed in such a way as to deliver cross-zone replication. You can build data
durability for your OpenSearch Service domain through automated and manual snapshots.

You can use snapshots to recover your domain with preloaded data or to create a new domain with
preloaded data. Snapshots are stored in Amazon S3, which is a secure, durable, highly-scalable
object storage. By default, S3 automatically creates daily snapshots of each domain. In addition,
you can use the S3 snapshot APIs to create additional manual snapshots. The manual snapshots
are stored in S3. Manual snapshots can be used for cross-Region disaster recovery and to provide
additional durability.

Scalability and elasticity

You can add or remove instances, and easily modify Amazon EBS volumes to accommodate data
growth. You can write a few lines of code that will monitor the state of your domain through
Amazon CloudWatch metrics and call the OpenSearch Service API to scale your domain up or down
based on thresholds you set. The service will run the scaling without any downtime. OpenSearch
Service supports 1 EBS volume (max size of 1.5 TB) per instance associated with a domain. With
the default maximum of 20 data nodes allowed per OpenSearch Service domain, you can allocate
about 30 TB of EBS storage to a single domain. You can request a service limit increase up to 100
instances per domain by creating a case with the AWS Support Center (sign-in required) With 100
instances, you can allocate about 150 TB of EBS storage to a single domain.

UltraWarm enables you to dramatically extend your data retention period and reduce costs by up
to 90% over hot storage. Best of all, the interactive analytics experience remains. Query your warm
indexes just like any other index, or use them to build OpenSearch Dashboards. UltraWarm uses a
combination of S3 and nodes powered by the AWS Nitro System to provide a hot-like experience
for aggregations and visualizations.

Cold storage lets you store any amount of infrequently accessed or historical data on your
OpenSearch Service domain and analyze it on demand, at a lower cost than other storage tiers.
Cold storage is appropriate if you need to do periodic research or forensic analysis on your older
data. Practical examples of data suitable for cold storage include infrequently accessed logs, data
that must be preserved to meet compliance requirements, or logs that have historical value.

Similar to UltraWarm storage, cold storage is backed by S3. When you need to query cold data, you
can selectively attach it to existing UltraWarm nodes. You can manage the migration and lifecycle
of your cold data manually or with Index State Management policies.

Scalability and elasticity 46

Big Data Analytics Options on AWS AWS Whitepaper

Interfaces
OpenSearch Service supports many of the commonly used OpenSearch APIs, so code, applications,
and popular tools that you're already using with your current OpenSearch environments will
work seamlessly. For a full list of supported OpenSearch operations, see OpenSearch Service
documentation .

The AWS CLI, API, or the AWS Management Console can be used for creating and managing your
domains as well.

OpenSearch Service supports integration with several AWS services, including streaming data
from S3 buckets, Amazon Kinesis Data Streams, and DynamoDB Streams. Both integrations use a
Lambda function as an event handler in the cloud that responds to new data in Amazon S3 and
Amazon Kinesis Data Streams by processing it and streaming the data to your OpenSearch Service
domain. OpenSearch Service also integrates with Amazon CloudWatch for monitoring OpenSearch
Service domain metrics and CloudTrail for auditing configuration API calls to OpenSearch Service
domains.

OpenSearch Service includes built-in integration with OpenSearch Dashboards, an open-source

analytics and visualization platform and supports integration with Logstash, an open-source data
pipeline that helps you process logs and other event data. You can set up your OpenSearch Service
domain as the backend store for all logs coming through your Logstash implementation to easily
ingest structured and unstructured data from a variety of sources.

Fine-grained access control

Fine-grained access control offers additional ways of controlling access to your data on OpenSearch
Service. For example, depending on who makes the request, you might want a search to return
results from only one index. You might want to hide certain fields in your documents or exclude
certain documents altogether. Fine-grained access control offers the following benefits:

• Role-based access control

• Security at the index, document, and field level
• OpenSearch Dashboards multi-tenancy
• HTTP basic authentication for OpenSearch Service and OpenSearch Dashboards

SAML authentication for OpenSearch Dashboards lets you use your existing identity provider to
offer single sign-on (SSO) for OpenSearch Dashboards on OpenSearch Service domains running

Interfaces 47
Big Data Analytics Options on AWS AWS Whitepaper

OpenSearch Service 6.7 or later. To use SAML authentication, you must enable fine-grained access
control.

Rather than authenticating through Amazon Cognito or the internal user database, SAML
authentication for OpenSearch Dashboards lets you use third-party identity providers to log in
to OpenSearch Dashboards, manage fine-grained access control, search your data, and build
visualizations. OpenSearch Service supports providers that use the SAML 2.0 standard, such
as Okta, Keycloak, Active Directory Federation Services (AD FS), and Auth0. Requests from
OpenSearch Service to third-party providers aren't explicitly encrypted with a service provider
certificate.

SAML authentication for OpenSearch Dashboards is only for accessing OpenSearch Dashboards
through a web browser. Your SAML credentials do not let you make direct HTTP requests to the
OpenSearch Service or OpenSearch Dashboards APIs.

Anti-patterns
Amazon OpenSearch Service has the following anti-patterns:

• Online transaction processing (OLTP) - OpenSearch Service is a real-time distributed search

and analytics engine. There is no support for transactions or processing on data manipulation. If
your requirement is for a fast transactional system, then a traditional relational database system
built on Amazon RDS, or a NoSQL database offering functionality such as DynamoDB, is a better
choice.
• Ad hoc data querying – While OpenSearch Service takes care of the operational overhead
of building a highly scalable OpenSearch Service cluster, if running ad hoc queries or one-off
queries against your data set is your use-case, Amazon Athena is a better choice. Amazon Athena
is an interactive query service that makes it easy to analyze data in Amazon S3 using standard
SQL, without provisioning servers.

Amazon QuickSight
Amazon QuickSight is a scalable, serverless, embeddable, machine learning-powered business
intelligence (BI) service built for the cloud. It makes it easy for all employees within an organization
to build visualizations, perform ad hoc analysis, and quickly get business insights from their
data, anytime, on any device. It can connect to a wide variety of data sources including flat files
such as CSV and Excel, access on-premises databases including Oracle, SQL Server, MySQL, and
PostgreSQL, AWS resources such as Amazon RDS databases, Amazon Redshift, Amazon Athena,

Anti-patterns 48
Big Data Analytics Options on AWS AWS Whitepaper

and Amazon S3, and SaaS solutions such as Salesforce, ServiceNow, and Adobe Analytics. Amazon
QuickSight enables organizations to scale their business analytics capabilities to hundreds of
thousands of users, and delivers fast and responsive query performance by using a robust in-
memory engine (SPICE).

Amazon QuickSight is built with "SPICE" – a Super-fast, Parallel, In-memory Calculation Engine.
Built from the ground up for the cloud, SPICE uses a combination of columnar storage, in-memory
technologies enabled through the latest hardware innovations, and machine code generation to
run interactive queries on large datasets and get rapid responses. SPICE supports rich calculations
to help you derive valuable insights from your analysis without worrying about provisioning or
managing infrastructure. Data in SPICE is persisted until it is explicitly deleted by the user. SPICE
also automatically replicates data for high availability and enables Amazon QuickSight to scale to
hundreds of thousands of users who can all simultaneously perform fast interactive analysis across
a wide variety of AWS data sources.

QuickSight leverages the AWS proven ML capabilities, making it easy for BI teams to perform
advanced analytics (for example, what-if analyses, anomaly detection, ML-based forecasting, churn
prediction, and so on) without prior data science experience. You can use QuickSight’s pre-built
models, or bring your own ML models from Amazon SageMaker.

Ideal usage patterns

Amazon QuickSight is an ideal Business Intelligence tool, allowing end users to create
visualizations that provide insight into their data to help them make better business decisions.
Amazon QuickSight can be used to do the following:

• Quick interactive ad hoc exploration and optimized visualization of data

• Create and share dashboards and KPIs to provide insight into your data
• Create Stories, which are guided tours through specific views of an analysis and enable you
to share insights and collaborate with others. They are used to convey key points, a thought
process, or the evolution of an analysis for collaboration
• Analyze and visualize data coming from logs and stored in S3
• Analyze and visualize data from on premise databases like SQL Server, Oracle, PostGreSQL, and
MySQL
• Analyze and visualize data in various AWS resources such as Amazon RDS databases, Amazon
Redshift, Amazon Athena, and Amazon S3.
• Analyze and visualize data in software as a service (SaaS) applications such as Salesforce.

Ideal usage patterns 49

Big Data Analytics Options on AWS AWS Whitepaper

• Analyze and visualize data in data sources that can be connected to using JDBC/ODBC
connection.
• Enhance with your QuickSight dashboard with Machine Learning Insights and Forecasts

• Ask natural language questions on your data and have Amazon QuickSight Q automatically build
your visualization

Cost model

Amazon QuickSight has two different editions for pricing; Standard edition and Enterprise edition.

• The Standard Edition is ideal for personal data analysis and exploration, it offers a full set
of features for creating and sharing data visualizations, but it does not provide some of the
Enterprise security and advanced options, a comparison of the features of the two editions can
be found on the Amazon QuickSight Pricing page. The Standard Edition is priced at $9/user/
month for an annual subscription ($12/GB/month for the month-to-month option) with ten GB
of SPICE capacity included. You can get additional SPICE capacity for $.25/GB/month.

• The Enterprise edition delivers insights at scale and offers embedding options, advanced security
(data encryption at rest, row and column level security, access on private VPC and on premises),
ML insights, folders, and templates.

It is priced at $18/author/month ($24/GB/month for the month-to-month option) with ten GB

of SPICE capacity included. Readers have a per-session pricing up to $5/user/month ($.30 per
session; 1 session = 30 minutes from login).

For large-scale deployments and public embedding, there is a Session capacity pricing starting at
$250 per month for 500 sessions.

Performance

Amazon QuickSight is built with SPICE. Built from the ground up for the cloud, SPICE uses a
combination of columnar storage, in-memory technologies enabled through the latest hardware
innovations, and machine code generation to run interactive queries on large datasets and get
rapid responses.

Cost model 50
Big Data Analytics Options on AWS AWS Whitepaper

Durability and availability

SPICE automatically replicates data for high availability and enables Amazon QuickSight to scale to
hundreds of thousands of users who can all simultaneously perform fast interactive analysis across
a wide variety of AWS data sources.

Scalability and elasticity

Amazon QuickSight is a fully managed service and it internally takes care of scaling to meet the
demands of your end users. With Amazon QuickSight you don’t need to worry about scale. You
can seamlessly grow your data from a few hundred megabytes to many terabytes of data without
managing any infrastructure.

Interfaces

Amazon QuickSight can connect to a wide variety of data sources including flat files (CSV, TSV, CLF,
ELF), connect to on-premises databases like SQL Server, MySQL, and PostgreSQL, and AWS data
sources including Amazon RDS, Amazon Aurora, Amazon Redshift, Amazon Athena and Amazon
S3, and SaaS, applications like Salesforce. You can also export data from any type of chart or
graph and the export contains only the data in the fields that are currently visible in the selected
visualization.

You can share an analysis, dashboard, or story by choosing Share with users and groups in your
account on the application bar from the Amazon QuickSight service interface. You can select the
recipients (email address, username, or group name), permission levels, and other options, before
sharing the content with others.

Anti-patterns

Amazon QuickSight has the following anti-patterns:

• Highly formatted canned Reports — Amazon QuickSight is used for ad hoc query, analysis,
and visualization of data. For highly formatted reports such as formatted financial statements,
consider using a different tool.

• ETL — While Amazon QuickSight can perform some transformations, it is not a full-fledged ETL
tool. AWS offers AWS Glue, which is a fully managed ETL service that makes it easy for customers
to prepare and load their data for analytics.

Durability and availability 51

Big Data Analytics Options on AWS AWS Whitepaper

Amazon Compute Services (EC2, EKS, and ECS)

Amazon EC2, with instances acting as AWS virtual machines, provides an ideal platform for
operating your own self-managed big data analytics applications on AWS infrastructure. Almost
any software you can install on Linux or Windows virtualized environments can be run on Amazon
EC2, and you can use the pay-as-you-go pricing model.

AWS Graviton processors are custom built by AWS using 64-bit Arm Neoverse cores to deliver the
best price performance for your cloud workloads running in Amazon EC2. Amazon EC2 provides
the broadest and deepest portfolio of compute instances, including many that are powered by
latest-generation Intel and AMD processors. AWS Graviton processors add even more choice to help
customers optimize performance and cost for their workloads.

What you don’t get are the application-level managed services that come with the other services
mentioned in this whitepaper. There are many options for self-managed big data analytics:

• A NoSQL offering, such as MongoDB

• A data warehouse or columnar store like Vertica

• A Hadoop cluster

• An Apache Storm cluster

• An Apache Kafka environment

Any self-managed big data workload that runs on EC2 can also run on an AWS fully managed
container orchestration service such as Amazon ECS, Amazon EKS, and AWS Fargate. Fargate is a
serverless compute engine for containers that works with ECS and EKS.

Ideal usage patterns

• Specialized environment – When running a custom application, a variation of a standard
Hadoop set or an application not covered by another AWS offering, Amazon EC2 provides the
flexibility and scalability to meet your computing needs.

• Compliance requirements – Certain compliance requirements may require you to run

applications yourself on Amazon EC2 instead of using a managed service offering.

Amazon Compute Services 52

Big Data Analytics Options on AWS AWS Whitepaper

Cost model
Amazon EC2 has a variety of instance types in a number of instance families (standard, high CPU,
high memory, high I/O, and so on), and different pricing options (On-Demand, Compute Savings
plan, Reserved, and Spot). At the time of this writing, when running applications on ECS, you pay
only for underlying EC2 instances, with no additional charge for using ECS. However, for EKS,
you pay an additional $0.10 per hour for each EKS cluster you have, along with underlying EC2
instances. AWS Fargate pricing is calculated based on the vCPU, memory, and storage resources
used from the time you start to download your container image until the Amazon ECS task or
Amazon EKS2 pod finishes, rounded up to the nearest second.

While cost is dependent on various factors based on the use case, Graviton2 instances have
in general been able to provide better price/performance over previous generation instances.
Depending on your application requirements, you may want to use additional services along with
Amazon EC2, EKS, or ECS, such as Amazon Elastic Block Store (Amazon EBS) for directly attached
persistent storage, or S3 as a durable object store; each comes with its own pricing model. If you do
run your big data application on Amazon EC2, EKS, or ECS, you are responsible for any license fees
just as you would be in your own data center. The AWS Marketplace offers many different third-
party, big data software packages pre-configured to launch with a simple click of a button.

Performance
Performance in Amazon EC2, EKS, or ECS is driven by the instance type that you choose for
your big data platform. Each instance type has a different amount of CPU, RAM, storage, IOPs,
and networking capability so that you can pick the right performance level for your application
requirements.

Durability and availability

Critical applications should be run in a cluster across multiple Availability Zones within an AWS
Region so that any instance or data center failure does not affect application users. For non-uptime
critical applications, you can back up your application to Amazon S3 and restore to any Availability
Zone in the Region if an instance or zone failure occurs. Other options exist, depending on which
application you are running and the requirements, such as mirroring your application.

Scalability and elasticity

Auto Scaling is a service that enables you to automatically scale your Amazon EC2 capacity up or
down according to conditions that you define. With Auto Scaling, you can ensure that the number

Cost model 53
Big Data Analytics Options on AWS AWS Whitepaper

of EC2 instances you’re using scales up seamlessly during demand spikes to maintain performance,
and scales down automatically during demand lulls to minimize costs. Auto Scaling is particularly
well suited for applications that experience hourly, daily, or weekly variability in usage. Auto
Scaling is enabled by CloudWatch and available at no additional charge beyond CloudWatch fees.

Interfaces

Amazon EC2, EKS, and ECS can be managed programmatically via API, SDK, or the AWS
Management Console. Metrics for compute utilization, memory utilization, storage utilization,
network consumption, and read/write traffic to your instances are free of charge using the console
or CloudWatch API operations.

The interfaces for your big data analytics software that you run on top of Amazon EC2 varies based
on the characteristics of the software you choose.

Anti-patterns

Amazon EC2 has the following anti-patterns:

• Managed service – If your requirement is a managed service offering where you abstract the
infrastructure layer and administration from the big data analytics, then this “do it yourself”
model of managing your own analytics software on Amazon EC2 may not be the correct choice.

• Lack of expertise or resources – If your organization does not have, or does not want to expend,
the resources or expertise to install and manage a high-availability installation for the system
in question, you should consider using the AWS equivalent such as Amazon EMR, DynamoDB,
Amazon Kinesis Data Streams, or Amazon Redshift.

Amazon Athena
Amazon Athena is an interactive query service that makes it easy to analyze data in Amazon S3
using standard SQL. Athena is serverless, so there is no infrastructure to setup or manage, and you
can start analyzing data immediately. You don’t need to load your data into Athena, as it works
directly with data stored in S3. Just log into the Athena Console, define your table schema, and
start querying. Amazon Athena uses Presto with full ANSI SQL support and works with a variety of
standard data formats, including CSV, JSON, ORC, Apache Parquet, and Apache Avro.

Interfaces 54
Big Data Analytics Options on AWS AWS Whitepaper

Ideal usage patterns

• Interactive ad hoc querying for web logs — Athena is a good tool for interactive one-time SQL
queries against data on Amazon S3. For example, you could use Athena to run a query on web
and application logs to troubleshoot a performance issue. You simply define a table for your
data and start querying using standard SQL. Athena integrates with Amazon QuickSight for easy
visualization.

• To query staging data before loading into Redshift — You can stage your raw data in S3 before
processing and loading it into Amazon Redshift, and then use Athena to query that data.

• Send AWS service logs to S3 for analysis with Athena — CloudTrail, CloudFront, ELB/ALB and
VPC flow logs can be analyzed with Athena. AWS CloudTrail logs include details about any API
calls made to your AWS services, including from the console. CloudFront logs can be used to
explore users’ surfing patterns across web properties served by CloudFront. Querying ELB/ALB
logs allows you to see the source of traffic, latency, and bytes transferred to and from Elastic
Load Balancing instances and backend applications. VPC flow logs capture information about
the IP traffic going to and from network interfaces in VPCs in the Amazon Virtual Private Cloud
(Amazon VPC). The logs enable you to investigate network traffic patterns and identify threats
and risks across your VPC estate.

• Building Interactive Analytical Solutions with notebook-based solutions such as RStudio,

Jupyter, or Zeppelin — Data scientists and Analysts are often concerned about managing
the infrastructure behind big data platforms while running notebook-based solutions such as
RStudio, Jupyter, and Zeppelin. Amazon Athena makes it easy to analyze data using standard
SQL without the need to manage infrastructure. Integrating these notebook-based solutions
with Amazon Athena gives data scientists a powerful platform for building interactive analytical
solutions.
• Query data in relational, non-relational, object and custom data sources leveraging Athena
Federated Query — The Amazon Athena federated query allows the user to run SQL queries
across data in relational, non-relational, object and custom data sources, with options to either
query the data in place or extract the data from these data sources and store it in S3.

Cost model

Amazon Athena has simple pay-as-you-go pricing, with no up-front costs or minimum fees, and
you’ll only pay for the resources you consume. It is priced per query, $5 per TB of data scanned,
and charges based on the amount of data scanned by the query. You can save from 30% to 90%

Ideal usage patterns 55

Big Data Analytics Options on AWS AWS Whitepaper

on your per-query costs and get better performance by compressing, partitioning, and converting
your data into columnar formats. Converting data to the columnar format allows Athena to read
only the columns it needs to process the query.

You are charged for the number of bytes scanned by Amazon Athena, rounded up to the nearest
megabyte, with a 10 MB minimum per query. There are no charges for Data Definition Language
(DDL) statements like CREATE/ALTER/DROP TABLE, statements for managing partitions, or failed
queries. Canceled queries are charged based on the amount of data scanned.

Because federated queries invoke Lambda functions in your account, you are charged for Lambda
when a Federated query is made.

Performance

You can improve the performance of your query by compressing, partitioning, and converting your
data into columnar formats. Amazon Athena supports open source columnar data formats such as
Apache Parquet and Apache ORC. Converting your data into a compressed, columnar format lowers
your cost and improves query performance by enabling Athena to scan less data from S3 when
running your query.

Durability and availability

Amazon Athena is highly available and executes queries using compute resources across multiple
facilities, automatically routing queries appropriately if a particular facility is unreachable. Athena
uses Amazon S3 as its underlying data store, making your data highly available and durable.
Amazon S3 provides durable infrastructure to store important data and is designed for durability
of 99.999999999% of objects. Your data is redundantly stored across multiple facilities and
multiple devices in each facility.

Scalability and elasticity

Athena is serverless, so there is no infrastructure to setup or manage, and you can start analyzing
data immediately. Because it is serverless it can scale automatically, as needed.

Security, authorization, and encryption

Amazon Athena allows you to control access to your data by using AWS IAM policies, Access
Control Lists (ACLs), and Amazon S3 bucket policies. With IAM policies, you can grant users fine-

Performance 56
Big Data Analytics Options on AWS AWS Whitepaper

grained control to your S3 buckets. By controlling access to data in S3, you can restrict users from
querying it using Athena. You can query data that’s been protected by:

• Server-side encryption with an S3-managed key

• Server-side encryption with an AWS KMS-managed key
• Client-side encryption with an AWS KMS-managed key

Amazon Athena also can directly integrate with AWS Key Management System (KMS) to encrypt
your result sets.

Interfaces
Querying can be done by using the Athena Console. Athena also supports CLI, API via SDK and
JDBC. Athena also integrates with Amazon QuickSight for creating visualizations based on the
Athena queries.

Athena Federated Query leverages Lambda as data source connectors as its extension to make
queries in sources other than S3. Sources such as Amazon CloudWatch Logs, DynamoDB, Amazon
DocumentDB, Amazon RDS, JDBC-compliant Postgres, and MySQL databases are natively
supported by Athena Federated Query. For others, you can use Athena Query Federation SDKs to
write custom connectors.

Anti-patterns
Amazon Athena has the following anti-patterns:

• Enterprise Reporting and Business Intelligence Workloads – Amazon Redshift is a better

tool for enterprise reporting and BI workloads involving iceberg queries or cached data at the
nodes. Data warehouses pull data from many sources, format and organize it, store it, and
support complex, high speed queries that produce business reports. The query engine in Amazon
Redshift has been optimized to perform especially well on data warehouse workloads.
• ETL Workloads – You should use Amazon EMR/AWS Glue if you are looking for an ETL tool
to process extremely large datasets and analyze them with the latest big data processing
frameworks such as Spark, Hadoop, Presto, or Hbase.
• RDBMS – Athena is not a relational/transactional database. It is not meant to be a replacement
for SQL engines like MySQL.

Interfaces 57
Big Data Analytics Options on AWS AWS Whitepaper

Solving big data problems on AWS

This whitepaper has examined some tools available on AWS for big data analytics. This paper
provides a good reference point when starting to design your big data applications. However, there
are additional aspects you should consider when selecting the right tools for your specific use case.
In general, each analytical workload has certain characteristics and requirements that dictate which
tool to use, such as:

• How quickly do you need analytic results: in real time, in seconds, or is an hour a more
appropriate time frame?
• How much value will these analytics provide your organization and what budget constraints
exist?
• How large is the data and what is its growth rate?
• How is the data structured?
• What integration capabilities do the producers and consumers have?
• How much latency is acceptable between the producers and consumers?
• What is the cost of downtime or how available and durable does the solution need to be?
• Is the analytic workload consistent or elastic?

Each one of these questions helps guide you to the right tool. In some cases, you can simply
map your big data analytics workload into one of the services based on a set of requirements.
However, in most real-world, big data analytic workloads, there are many different, and sometimes
conflicting, characteristics and requirements on the same data set.

For example, some result sets may have real-time requirements as a user interacts with a system,
while other analytics could be batched and run on a daily basis. These different requirements
over the same data set should be decoupled and solved by using more than one tool. If you try
to solve both of these examples using the same toolset, you end up either over-provisioning or
therefore overpaying for unnecessary response time, or you have a solution that does not respond
fast enough to your users in real time. Matching the best-suited tool to each analytical problem
results in the most cost-effective use of your compute and storage resources.

Big data doesn’t need to mean “big costs”. So, when designing your applications, it’s important to
make sure that your design is cost efficient. If it’s not, relative to the alternatives, then it’s probably
not the right design. Another common misconception is that using multiple tool sets to solve a big

58
Big Data Analytics Options on AWS AWS Whitepaper

data problem is more expensive or harder to manage than using one big tool. If you take the same
example of two different requirements on the same data set, the real-time request may be low on
CPU but high on I/O, while the slower processing request may be very compute intensive.

Decoupling can end up being much less expensive and easier to manage, because you can build
each tool to exact specifications and not overprovision. With the AWS pay-as-you-go model, this
equates to a much better value because you could run the batch analytics in just one hour and
therefore only pay for the compute resources for that hour. Also, you may find this approach
easier to manage rather than leveraging a single system that tries to meet all of the requirements.
Solving for different requirements with one tool results in attempting to fit a square peg (real-time
requests) into a round hole (a large data warehouse).

The AWS platform makes it easy to decouple your architecture by having different tools analyze
the same data set. AWS services have built-in integration so that moving a subset of data from one
tool to another can be done very easily and quickly using parallelization. Following are some real
world, big data analytics problem scenarios, and an AWS architectural solution for each.

Example 1: Queries against an Amazon S3 data lake

Data lakes are an increasingly popular way to store and analyze both structured and unstructured
data. If you use an Amazon S3 data lake, AWS Glue can make all your data immediately available
for analytics without moving the data. AWS Glue crawlers can scan your data lake and keep the
AWS Glue Data Catalog in sync with the underlying data. You can then directly query your data lake
with Amazon Athena and Amazon Redshift Spectrum. You can also use the AWS Glue Data Catalog
as your external Apache Hive Metastore for big data applications running on Amazon EMR.

Queries against an Amazon S3 data lake

Example 1: Queries against an Amazon S3 data lake 59

Big Data Analytics Options on AWS AWS Whitepaper

1. An AWS Glue crawler connects to a data store, progresses through a prioritized list of classifiers
to extract the schema of your data and other statistics, and then populates the AWS Glue Data
Catalog with this metadata. Crawlers can run periodically to detect the availability of new data
as well as changes to existing data, including table definition changes. Crawlers automatically
add new tables, new partitions to existing table, and new versions of table definitions. You can
customize AWS Glue crawlers to classify your own file types.
2. The AWS Glue Data Catalog is a central repository to store structural and operational metadata
for all your data assets. For a given data set, you can store its table definition, physical location,
add business relevant attributes, as well as track how this data has changed over time. The AWS
Glue Data Catalog is Apache Hive Metastore compatible and is a drop-in replacement for the
Apache Hive Metastore for Big Data applications running on Amazon EMR. For more information
on setting up your EMR cluster to use AWS Glue Data Catalog as an Apache Hive Metastore, see
AWS Glue documentation.
3. The AWS Glue Data Catalog also provides out-of-box integration with Amazon Athena, Amazon
EMR, and Amazon Redshift Spectrum. After you add your table definitions to the AWS Glue Data
Catalog, they are available for ETL and also readily available for querying in Amazon Athena,
Amazon EMR, and Amazon Redshift Spectrum so that you can have a common view of your data
between these services.
4. Using a BI tool like Amazon QuickSight enables you to easily build visualizations, perform ad hoc
analysis, and quickly get business insights from your data. Amazon QuickSight supports data
sources such as Amazon Athena, Amazon Redshift Spectrum, Amazon S3 and many others. See
Supported Data Sources.

Example 2: Capturing and analyzing sensor data

An international air conditioner manufacturer has many large air conditioners that it sells to
various commercial and industrial companies. Not only do they sell the air conditioner units but, to
better position themselves against their competitors, they also offer add-on services where you can
see real-time dashboards in a mobile app or a web browser. Each unit sends its sensor information
for processing and analysis. This data is used by the manufacturer and its customers. With this
capability, the manufacturer can visualize the dataset and spot trends.

Currently, they have a few thousand pre-purchased air conditioning (A/C) units with this capability.
They expect to deliver these to customers in the next couple of months and are hoping that, in
time, thousands of units throughout the world will use this platform. If successful, they would like
to expand this offering to their consumer line as well, with a much larger volume and a greater

Example 2: Capturing and analyzing sensor data 60

Big Data Analytics Options on AWS AWS Whitepaper

market share. The solution needs to be able to handle massive amounts of data and scale as they
grow their business without interruption. How should you design such a system?

First, break it up into two work streams, both originating from the same data:

• A/C unit’s current information with near-real-time requirements and a large number of
customers consuming this information
• All historical information on the A/C units to run trending and analytics for internal use

The data-flow architecture in the following figure shows how to solve this big data problem.

Capturing and analyzing sensor data

1. The process begins with each A/C unit providing a constant data stream to Amazon Kinesis Data
Streams. This provides an elastic and durable interface the units can talk to that can be scaled
seamlessly as more and more A/C units are sold and brought online.
2. Using the Amazon Kinesis Data Streams-provided tools such as the Kinesis Client Library or SDK,
a simple application is built on Amazon EC2 to read data as it comes into Amazon Kinesis Data
Streams, analyze it, and determine if the data warrants an update to the real-time dashboard. It
looks for changes in system operation, temperature fluctuations, and any errors that the units
encounter.
3. This data flow needs to occur in near real time so that customers and maintenance teams can
be alerted quickly if there is an issue with the unit. The data in the dashboard does have some
aggregated trend information, but it is mainly the current state as well as any system errors. So,

Example 2: Capturing and analyzing sensor data 61

Big Data Analytics Options on AWS AWS Whitepaper

the data needed to populate the dashboard is relatively small. Additionally, there will be lots of
potential access to this data from the following sources:
• Customers checking on their system via a mobile device or browser
• Maintenance teams checking the status of its fleet
• Data and intelligence algorithms and analytics in the reporting platform spot trends that can
be then sent out as alerts, such as if the A/C fan has been running unusually long with the
building temperature not going down.

DynamoDB was chosen to store this near real-time data set because it is both highly available
and scalable; throughput to this data can be easily scaled up or down to meet the needs of its
consumers as the platform is adopted and usage grows.
4. The reporting dashboard is a custom web application that is built on top of this data set and run
on Amazon EC2. It provides content based on the system status and trends as well as alerting
customers and maintenance crews of any issues that may come up with the unit.
5. The customer accesses the data from a mobile device or a web browser to get the current status
of the system and visualize historical trends.

The data flow (steps 2-5) that was just described is built for near real-time reporting of
information to human consumers. It is built and designed for low latency and can scale very
quickly to meet demand. The data flow (steps 6-9) that is depicted in the lower part of the
diagram does not have such stringent speed and latency requirements. This allows the architect
to design a different solution stack that can hold larger amounts of data at a much smaller cost
per byte of information and choose less expensive compute and storage resources.
6. To read from the Amazon Kinesis stream, there is a separate Amazon Kinesis-enabled application
that probably runs on a smaller EC2 instance that scales at a slower rate. While this application
is going to analyze the same data set as the upper data flow, the ultimate purpose of this data
is to store it for long-term record and to host the data set in a data warehouse. This data set
ends up being all data sent from the systems and allows a much broader set of analytics to be
performed without the near real-time requirements.
7. The data is transformed by the Amazon Kinesis-enabled application into a format that is suitable
for long-term storage, for loading into its data warehouse, and storing on Amazon S3. The data
on Amazon S3 not only serves as a parallel ingestion point to Amazon Redshift, but is durable
storage that will hold all data that ever runs through this system; it can be the single source of
truth. It can be used to load other analytical tools if additional requirements arise. Amazon S3
also comes with native integration with Amazon Glacier, if any data needs to be cycled into long-
term, low-cost storage.

Example 2: Capturing and analyzing sensor data 62

Big Data Analytics Options on AWS AWS Whitepaper

8. Amazon Redshift is again used as the data warehouse for the larger data set. It can scale easily
when the data set grows larger, by adding another node to the cluster.
9. For visualizing the analytics, one of the many partner visualization platforms can be used via
the OBDC/JDBC connection to Amazon Redshift. This is where the reports, graphs, and ad hoc
analytics can be performed on the data set to find certain variables and trends that can lead to
A/C units underperforming or breaking.

This architecture can start off small and grow as needed. Additionally, by decoupling the two
different work streams from each other, they can grow at their own rate without upfront
commitment, allowing the manufacturer to assess the viability of this new offering without a large
initial investment. You could imagine further additions, such as adding Amazon ML to predict how
long an A/C unit will last and preemptively sending out maintenance teams based on its prediction
algorithms to give their customers the best possible service and experience. This level of service
would be a differentiator to the competition and lead to increased future sales.

Example 3: sentiment analysis of social media

A large toy maker has been growing very quickly and expanding their product line. After each
new toy release, the company wants to understand how consumers are enjoying and using their
products. Additionally, the company wants to ensure that their consumers are having a good
experience with their products. As the toy system grows, the company wants to ensure that their
products are still relevant to their customers and that they can plan future roadmaps items based
on customer feedback. The company wants to capture the following insights from social media:

• Understand how consumers are using their products

• Ensure customer satisfaction

• Plan future roadmaps

Capturing the data from various social networks is relatively easy but the challenge is building
the intelligence programmatically. After the data is ingested, the company wants to be able to
analyze and classify the data in a cost-effective and programmatic way. To do this, they can use the
architecture in the following figure.

Example 3: sentiment analysis of social media 63

Big Data Analytics Options on AWS AWS Whitepaper

Sentiment analysis of social media

1. First, deploy an Amazon EC2 instance in an Amazon VPC that ingests tweets from Twitter.

2. Next, create an Amazon Data Firehose delivery stream that loads the streaming tweets into the
raw prefix in the solution's S3 bucket.

3. S3 invokes a Lambda function to analyze the raw tweets using Amazon Translate to translate
non-English tweets into English, and Amazon Comprehend to use natural language-processing
(NLP) to perform entity extraction and sentiment analysis.

4. A second Firehose delivery stream loads the translated tweets and sentiment values into the
sentiment prefix in the S3 bucket. A third delivery stream loads entities in the entities prefix in
the S3 bucket.

5. This architecture also deploys a data lake that includes AWS Glue for data transformation,
Amazon Athena for data analysis, and Amazon QuickSight for data visualization. AWS Glue Data
Catalog contains a logical database used to organize the tables for the data in S3. Athena uses

Example 3: sentiment analysis of social media 64

Big Data Analytics Options on AWS AWS Whitepaper

these table definitions to query the data stored in S3 and return the information to an Amazon
QuickSight dashboard.

By using ML and BI services from AWS including Amazon Translate, Amazon Comprehend, Amazon
Kinesis, Amazon Athena, and Amazon QuickSight, you can build meaningful, low-cost social media
dashboards to analyze customer sentiment, which can lead to better opportunities for acquiring
leads, improve website traffic, strengthen customer relationships, and improve customer service.

This example solution automatically provisions and configures the AWS services necessary to
capture multi-language tweets in near-real-time, translate them, and display them on a dashboard
powered by Amazon QuickSight. You can also capture both the raw and enriched datasets and
durably store them in the solution's data lake. This enables data analysts to quickly and easily
perform new types of analytics and ML on this data. For more information, see the AI-Driven Social
Media Dashboard solution.

Example 3: sentiment analysis of social media 65

Big Data Analytics Options on AWS AWS Whitepaper

Conclusion
As more and more data is generated and collected, data analysis requires scalable, flexible, and
high performing tools to provide insights in a timely fashion. However, organizations are facing
a growing big data environment, where new tools emerge and become outdated very quickly.
Therefore, it can be very difficult to keep pace and choose the right tools.

This whitepaper offers a first step to help you solve this challenge. With a broad set of managed
services to collect, process, and analyze big data, AWS makes it easier to build, deploy, and scale
big data applications. This enables you to focus on business problems instead of updating and
managing these tools.

AWS provides many solutions to address your big data analytic requirements. Most big data
architecture solutions use multiple AWS tools to build a complete solution. This approach helps
meet stringent business requirements in the most cost-optimized, performant, and resilient way
possible. The result is a flexible big data architecture that is able to scale along with your business.

66
Big Data Analytics Options on AWS AWS Whitepaper

Contributors
The following individuals and organizations contributed to this document:

• Erik Swensson, Manager, Solutions Architecture, Amazon Web Services

• Erick Dame, Solutions Architect, Amazon Web Services
• Shree Kenghe, Solutions Architect, Amazon Web Services
• Changbin Gong, Solutions Architect, Amazon Web Services
• Raghavarao Sodabathina, EnterpriseSolutions Architect, Amazon Web Services
• Fabrizio Napolitano, Solutions Architect, Amazon Web Services
• Chinmayi Narasimhadevara, Solutions Architect, Amazon Web Services
• Aishwarya Subramaniam, Solutions Architect, Amazon Web Services
• Pavitra Krishnan, Solutions Architect, Amazon Web Services

67
Big Data Analytics Options on AWS AWS Whitepaper

Further reading
The following resources can help you get started in running big data analytics on AWS:

• Analytics on AWS — View the comprehensive portfolio of big data services as well as links to
other resources such AWS big data partners, tutorials, articles, and AWS Marketplace offerings on
big data solutions.
• Read the AWS Big Data Blog — The blog features real life examples and ideas updated regularly
to help you collect, store, clean, process, and visualize big data.
• Try one of the Analytics Quick Starts — Explore the rich environment of products designed to
address big data challenges using AWS. Test Drives are developed by AWS Partner Network
(APN) Consulting and Technology partners and are provided free of charge for education,
demonstration, and evaluation purposes.
• Take an AWS training course on big data — The Big Data on AWS course introduces cloud-based
big data solutions and Amazon EMR. AWS shows how to use Amazon EMR to process data using
the broad system of Hadoop tools such as Pig and Hive. AWS also teaches how to create big data
environments, work with DynamoDB and Amazon Redshift, understand the benefits of Amazon
Kinesis Streams, and leverage best practices to design big data environments for security and
cost-effectiveness.
• View the Big Data Customer Case Studies — Learn from the experience of other customers who
have built powerful and successful big data platforms on the AWS Cloud.

68
Big Data Analytics Options on AWS AWS Whitepaper

Document revisions
To be notified about updates to this whitepaper, subscribe to the RSS feed.

Change Description Date

Minor update Fix non-inclusive language. April 6, 2022

Whitepaper updated Revised to add informati July 26, 2021

on on Amazon SageMaker
, Amazon EMR Studio, AWS
Glue DataBrew, AWS Lake
Formation, and general
updates throughout.

Whitepaper updated Revised to add information December 1, 2018

on Amazon Athena, AWS
QuickSight, AWS Glue, and
general updates throughout.

Whitepaper updated Revised to add information January 1, 2016

on Amazon Machine Learning,
AWS Lambda, Amazon
OpenSearch Service; general
update.

Initial publication Whitepaper first published. December 1, 2014

69
Big Data Analytics Options on AWS AWS Whitepaper

Notices
Customers are responsible for making their own independent assessment of the information in
this document. This document: (a) is for informational purposes only, (b) represents current AWS
product offerings and practices, which are subject to change without notice, and (c) does not create
any commitments or assurances from AWS and its affiliates, suppliers or licensors. AWS products or
services are provided “as is” without warranties, representations, or conditions of any kind, whether
express or implied. The responsibilities and liabilities of AWS to its customers are controlled by
AWS agreements, and this document is not part of, nor does it modify, any agreement between
AWS and its customers.

© 2021 Amazon Web Services, Inc. or its affiliates. All rights reserved.

70