0% found this document useful (0 votes)

129 views70 pages

(25D3T2S04) - Amazon DynamoDB

The AWS Data & AI Roadshow 2025 focuses on Amazon DynamoDB, discussing its capabilities, recent updates, and various use cases. It highlights the importance of data modeling, partitioning, and the architecture of DynamoDB, along with its role in Amazon's database migration strategy. The document also showcases successful implementations of DynamoDB across various Amazon services and external companies, emphasizing its scalability and performance benefits.

Uploaded by

이동환

We take content rights seriously. If you suspect this is your content, claim it here.

Available Formats

Download as PDF, TXT or read online on Scribd

0% found this document useful (0 votes)

129 views70 pages

(25D3T2S04) - Amazon DynamoDB

Uploaded by

이동환

We take content rights seriously. If you suspect this is your content, claim it here.

Available Formats

Download as PDF, TXT or read online on Scribd

You are on page 1/ 70

AWS DATA & AI ROADSHOW 2025

Amazon DynamoDB로 무엇이 가능할까?

Hyuk Lee
GTM Database SA

Agenda

• What is Amazon DynamoDB?

• Recent updates from Amazon DynamoDB

• Database Freedom at Amazon.com

• Amazon DynamoDB use cases

What is Amazon DynamoDB?

“Sporadic ‘problems’ block access to Amazon.com”

CNN article, December 6, 2004

“Amazon.com hit with outages”

CNET article, December 7, 2004

2004
Amazon.com
outage
© 2025 Amazon Web Services, Inc. or its affiliates. All rights reserved. 4
AWS DATA & AI ROADSHOW 2025

20% of queries
70% of
queries
operated on a
+ returned multiple
rows, but only
from a single
= 90% of queries
did not use
joins
single record
table

2004
Amazon.com
outage
© 2025 Amazon Web Services, Inc. or its affiliates. All rights reserved. 5
AWS DATA & AI ROADSHOW 2025

Today

2004 2007 2012 2025

Amazon.com Dynamo DynamoDB
outage paper launch
© 2025 Amazon Web Services, Inc. or its affiliates. All rights reserved. 6
AWS DATA & AI ROADSHOW 2025

Amazon DynamoDB

1M+
Customers using DynamoDB

0-9 partition
10-19
20-29
30-39
40-49
50-59

DynamoDB is a distributed
DynamoDB
table hash table

...
...

n-m
© 2025 Amazon Web Services, Inc. or its affiliates. All rights reserved. 8
AWS DATA & AI ROADSHOW 2025

Data modeling
You must consider your access patterns and the
distribution of data

Partitioning
DynamoDB automatically distributes your items
based on the partition key

DynamoDB partition key Hash Function partition

table

Split 1 Split 2 Split 3

0-2
0-2
3
0-4

3-4
0-9 4
5-7
5-7
5-9
8-9
8-9
t

Automatically partitions down to the individual

primary key
© 2025 Amazon Web Services, Inc. or its affiliates. All rights reserved. 12
AWS DATA & AI ROADSHOW 2025
Load Request Storage Auto
balancers routers Streams
Nodes Admin

AZ1

App AZ2
Frontend Backend

AZ3
...

AuthN/AuthZ Metadata Transactions Backup/restore

© 2025 Amazon Web Services, Inc. or its affiliates. All rights reserved.
KMS PITR Import/export13
AWS DATA & AI ROADSHOW 2025

Storage
Node

Database architectures

Single tenant Multi-tenant

AMAZON
DYNAMODB

Storage
Node

Recent Updates from Amazon DynamoDB

Enabling querying with microservices CQRS architecture

Analytics/Query side
OLTP/Command side Amazon Simple Storage
Service (S3)
Amazon
Athena

Amazon API AWS Amazon AWS

Gateway Lambda DynamoDB Lambda

Amazon Redshift
Amazon
DynamoDB
stream
ü Reliable downstream delivery of events
Amazon
via DynamoDB stream and Lambda OpenSearch
ü Independently scalable command and Service

query sides

DynamoDB

Command Query Responsibility Segregation (CQRS)

DynamoDB Global Tables

Multi-Region eventual consistency Multi-Region strong consistency

• Lower write and strongly consistent • Higher write and strongly consistent
read latencies read latencies

• Strongly consistent reads can return • Strongly consistent reads never

stale data return stale data

• Conflicts are resolved with • Conflicts return a

last writer wins ReplicatedWriteConflictException

• Typical RPO is single-digit seconds • RPO is zero

Eventually consistent reads are always eventually consistent

Database Freedom at Amazon.com

Recommendations of data stores

Key to the project's success was an AWS technical core team of experienced
solutions architects and database engineers. This team made recommendations as
to which AWS services would be best suited for each category of Amazon data
being migrated from Oracle, including:
• Amazon DynamoDB: recommended for critical services requiring high availability,
mutating schemas, and consistent, single-digit millisecond latency at scale.
• Amazon Aurora: recommended for services with stable schemas requiring high
availability and strong referential integrity.
• Amazon Simple Storage Service (Amazon S3): recommended for inexpensive,
long term storage of relational and non-relational data.

https://aws.amazon.com/solutions/case-studies/amazon-database-migration/
© 2025 Amazon Web Services, Inc. or its affiliates. All rights reserved. 24
AWS DATA & AI ROADSHOW 2025

Recommendations of data stores

• Amazon Relational Database Service (Amazon RDS) for PostgreSQL or MySQL:

recommended for non-critical services and ease of set up, operation, and scaling.
• AWS Database Migration Service (AWS DMS): helps migrate databases to AWS
quickly and securely. The source database remains fully operational during the
migration, minimizing downtime to applications that rely on the database. DMS
can migrate data to and from most widely used commercial and open-source
databases.
• AWS Schema Conversion Tool (AWS SCT): makes heterogeneous database
migrations predictable by automatically converting the source database schema
and a majority of the database code objects, including views, stored procedures,
and functions, to a format compatible with the target database.

Database usage during Prime Day 2024

Amazon Aurora – On Prime Day, 6,311 database instances running the PostgreSQL-
compatible and MySQL-compatible editions of Amazon Aurora processed more
than 376 billion transactions, stored 2,978 terabytes of data, and transferred 913
terabytes of data.
Amazon DynamoDB – DynamoDB powers multiple high-traffic Amazon properties
and systems including Alexa, the Amazon.com sites, and all Amazon fulfillment
centers. Over the course of Prime Day, these sources made tens of trillions of calls
to the DynamoDB API. DynamoDB maintained high availability while delivering
single-digit millisecond responses and peaking at 146 million requests per second.

Amazon ElastiCache – ElastiCache served more than quadrillion requests on a single

day with a peak of over 1 trillion requests per minute.

https://aws.amazon.com/blogs/aws/how-aws-powered-prime-day-2024-for-record-breaking-sales/
© 2025 Amazon Web Services, Inc. or its affiliates. All rights reserved. 26
AWS DATA & AI ROADSHOW 2025

Amazon Wallet

https://d1.awsstatic.com/events/reinvent/2019/Amazon_Wallet_Increasing_performance_with_DynamoDB_AMZ302.pdf

Amazon Wallet
Benefits from Business
• Improved latency by 50%
• Increased request per second per host by 40%
• Reduced operations costs by 90%

Benefits from DynamoDB

• Autoscaling (including on-demand)
• Time to live (TTL)
• Point-in-time recovery
• On-demand backup and restore
• Global tables
© 2025 Amazon Web Services, Inc. or its affiliates. All rights reserved. 28
AWS DATA & AI ROADSHOW 2025

Amazon Prime Video

https://aws.amazon.com/solutions/case-studies/amazon-primevideo-dynamodb/

Amazon Item Master Service (Catalog)

https://aws.amazon.com/solutions/case-studies/itemsandoffers/

Amazon Item Master Service (Catalog)

https://aws.amazon.com/solutions/case-studies/itemsandoffers/

Amazon Ads

https://www.youtube.com/watch?v=BS4MPpEO66I

Amazon Ads

https://www.youtube.com/watch?v=BS4MPpEO66I

Amazon Ads

https://www.youtube.com/watch?v=BS4MPpEO66I

Amazon DynamoDB use cases

Disney+

https://www.youtube.com/watch?v=TCnmtSY2dFM

Disney+

https://www.youtube.com/watch?v=TCnmtSY2dFM

Disney+

https://www.youtube.com/watch?v=TCnmtSY2dFM

Disney+

https://www.youtube.com/watch?v=TCnmtSY2dFM

Disney+

https://www.youtube.com/watch?v=TCnmtSY2dFM

Mercardo Libre (MELI)

https://www.youtube.com/watch?v=pA6nET3YAqU

Mercardo Libre (MELI)

https://www.youtube.com/watch?v=pA6nET3YAqU

Mercardo Libre (MELI)

https://www.youtube.com/watch?v=pA6nET3YAqU

Mercardo Libre (MELI)

https://www.youtube.com/watch?v=pA6nET3YAqU

Mercardo Libre (MELI)

https://www.youtube.com/watch?v=pA6nET3YAqU

Dropbox

https://www.youtube.com/watch?v=EqtkKttteyI

Dropbox

https://www.youtube.com/watch?v=EqtkKttteyI

TMAP

One of the 10 largest banks in the

U.S. serving 45 million customers

Challenge:
Monolithic mainframe applications were difficult to maintain,
scale, and innovate.

Solution:
Migrated from mainframe to Amazon DynamoDB.

Result:
• Serving millions of transactions with the new
serverless architecture
• Reduced latency: end-to-end requests less than 55ms
• Unbound scale for app developers

Lyft is a ridesharing company founded

in 2012 and with 315,000 active
drivers in more than 190 U.S. cities
Challenge:
Had to handle massive scale and bursts gracefully.

Solution:
Lyft has chosen Amazon DynamoDB to handle scale, critical
transactional data, and rapid growth with a small company.

Result:
Scaled to 200+ cities, 14M+ rides-per-month, 8X traffic
during peak hours. Mission-critical transactional data like
the GPS location for all rides are tracked in the ride location
tracking system. It also allowed a small team to handle
rapid growth cycles.
“DynamoDB is so simple to scale out. We only have two knobs.
One for reads and one for writes.”
—Chris Lambert,
CTO, Lyft

Fortnite—125+ million players

Challenge:
Fortnite’s revenue depends on continuously capturing the
attention of gamers through new content and continuous
innovation. To operate this way, Epic Games needs an up-to-
the-minute understanding of gamer satisfaction to provide
engaging experiences.

Solution:
Real-time: They have Amazon EMR Spark, Amazon DynamoDB,
Grafana, Scoreboards, and to do real-time query
Batch: They store everything in an Amazon S3 Data Lake.
On top of it, they are running a large EMR cluster for the bulk
of our data processing.

Result:
They guaranteed gamers are engaged resulting in
the most successful game currently in the market.

Amazon Amazon Amazon

EMR DynamoDB
© 2025 Amazon Web Services, Inc. or its affiliates. All rights reserved. S3 51
AWS DATA & AI ROADSHOW 2025
Duolingo is the world's most popular way to
learn a language, with 300M total users and
7B exercises per month

Challenge:
Wanted to enable anyone to learn a language for free.

Solution:
Used purpose-built databases from AWS:
• Amazon DynamoDB: 31B items tracking which language
exercises completed
• Amazon Aurora: primary transactional database
for user data
• Amazon ElastiCache: instant access to common words
and phrases

Result:
More people learning a language on Duolingo than entire
U.S. school system.

Amazon Amazon Amazon

A community marketplace that allows

property owners and travelers to
connect with each other for the purpose
of renting unique vacation spaces
around the world

Challenge:
IT experienced service challenges with its original provider
and wanted to scale their internet business to the next level.

Solution:
Airbnb moved its main MySQL database to Amazon RDS
and more purpose-built databases for different needs:
• Relational data: Amazon RDS for referential integrity and
primary transactional database
• User search history: Amazon DynamoDB for massive data
volume, and quick lookups for personalized search
• Session state: Amazon ElastiCache for in-memory store for
sub millisecond site rendering

Amazon Amazon Amazon

Daangn (Karrot)

https://www.youtube.com/watch?v=aciRo1geadk

Daangn (Karrot)

https://www.youtube.com/watch?v=aciRo1geadk

Daangn (Karrot)

https://www.youtube.com/watch?v=aciRo1geadk

Figma

https://www.figma.com/blog/the-infrastructure-behind-ai-search-in-figma/

Figma

https://www.figma.com/blog/the-infrastructure-behind-ai-search-in-figma/

GrabFood

https://engineering.grab.com/how-we-store-millions-orders

Canva