Scribd
Upload
44 views

32-Serverless Computing

Serverless computing is an approach that allows developers to focus on code and business logic by offloading responsibility for infrastructure management tasks like scaling, scheduling, patching and provisioning to cloud providers. With serverless, resources are allocated on-demand and developers only pay for resources used, avoiding costs for idle capacity. Code executes in stateless containers on a per-request basis, scaling transparently based on request volume. Serverless is well-suited for short-running stateless functions, event-based processing, and microservices, but may not be optimal for all applications or workloads characterized by long-running processes.

Uploaded by

Yesmine Makkes
Copyright
© © All Rights Reserved
Available Formats
Download as TXT, PDF, TXT or read online on Scribd
44 views

32-Serverless Computing

Serverless computing is an approach that allows developers to focus on code and business logic by offloading responsibility for infrastructure management tasks like scaling, scheduling, patching and provisioning to cloud providers. With serverless, resources are allocated on-demand and developers only pay for resources used, avoiding costs for idle capacity. Code executes in stateless containers on a per-request basis, scaling transparently based on request volume. Serverless is well-suited for short-running stateless functions, event-based processing, and microservices, but may not be optimal for all applications or workloads characterized by long-running processes.

Uploaded by

Yesmine Makkes
Copyright
© © All Rights Reserved
Available Formats
Download as TXT, PDF, TXT or read online on Scribd
You are on page 1/ 2

Serverless is an approach to computing that offloads responsibility for common

infrastructure
management tasks such as scaling, scheduling, patching, and provisioning
application stacks
to cloud providers, allowing developers to focus their time and effort on the code
and
business logic specific to their applications or process.
Serverless doesn’t mean there are no servers; only that the management of the
underlying
physical or virtual servers is removed from their users.
The serverless computing environment allocates resources as needed for the
applications.
Let’s look at some key attributes that distinguish serverless computing from other
compute models.
The serverless model requires no provisioning of servers, installation of
application stacks
and software, or operation of the infrastructure by the developer.
Serverless computing runs code only on-demand on a per-request basis, scaling
transparently
with the number of requests being served.
Serverless enables end users to pay only for resources being used, never paying for
idle
capacity, which is unlike virtual servers on the cloud—where end users pay for VMs
as long as they are running even if idle.
Effectively, serverless abstracts the infrastructure away from developers.
Code is executed as individual functions where each function runs inside a
stateless container.
No prior execution context is required to serve a request; and with each new
request,
a new instance of the function is invoked.
Let’s look at a scenario.
You could, for example, have a serverless platform between the front-end of your
website
and your storage layer, running individual functions.
The serverless app could be translating text files and storing it in a cloud-based
storage
service.
Using the front-end of your website, you send text files to a serverless app; the
app creates
translations in different languages, and then stores these translated files in
cloud storage,
and sends their links back to you.
Some of the key serverless computing services today include IBM Cloud Functions
(which is
based on Apache OpenWhisk), AWS Lamb-da, and Microsoft Azure Functions.
It is important to note that serverless may not be the best fit for all
applications or
scenarios.
You need to evaluate application characteristics and ensure that the application is
aligned
to serverless architecture patterns.
Applications that qualify for a serverless architecture include some of the
following
characteristics: Short-running stateless functions (seconds
or minutes).
Seasonal workloads with varying off-peak and peaks.
Production volumetric data that shows too much idle time.
Event-based processing or asynchronous request processing for implementing use
cases.
Microservices that can be built as functions that are stateless.
Serverless architectures are well-suited for use cases around data and event
processing,
IoT, microservices, and mobile backends.
Given its inherent and automatic scaling, rapid provisioning, and a pricing model
that
does not charge for idle time, supporting microservices architecture has become one
of the most common use cases of serverless computing today.
Serverless is well-suited to working with structured text, audio, image, and video
data
around tasks such as data enrichment, transformation, validation and cleansing, PDF
processing,
audio normalization, thumbnail generation, and video transcoding.
Parallel tasks such as data search and processing, and genome processing, are also
well-suited
to be run on a serverless runtime.
Serverless is also well-suited for working with all sorts of data stream
ingestions,
including business data streams, IoT sensor data, log data, and financial market
data.
And finally, let’s look at some challenges worth considering about serverless.
Serverless workloads are designed to scale up and down in response to workload, but
for
workloads characterized by long-running processes managing a traditional server
environment
might be simpler and more cost-effective.
The serverless application architecture can be vendor dependent, and so there is a
potential
for vendor lock-in, particularly involving platform capabilities such as
authentication,
scaling, monitoring, or configuration management.
Because serverless architectures scale up and down in response to workload, they
also
sometimes need to start up from zero to serve a new request.
For certain applications, this delay isn’t much of an impact, but for something
like
a low-latency financial application, this delay wouldn’t be acceptable.

You might also like