Getting Started with AWS: Intro to Serverless Computing

Source: AWS Cloud Practitioner Essentials

You may have heard the term Serverless Computing but what is it actually?

Serverless computing is an execution model for the cloud in which the developer writes and runs an application without having to take care of the underlying server. The cloud provider does provisioning, scaling, server maintenance, etc. to meet demand.

In the cloud platform Amazon Web Services, serverless architecture has emerged as a key pattern for developing event-driven, low-maintenance applications that are scalable.

But keep in mind, serverless is not equal to “no servers.” It means:

• Managing servers is not your responsibility
• Infrastructure is abstracted away by AWS
• Billed only when you use it (execution time and resources used)

This model is perfect for applications that have varying traffic, event-based workflows, and microservices architectures.

Core AWS Serverless Services

AWS Lambda

1. AWS Lambda

Event-Driven, AWS Lambda executes your code when events happen in Amazon S3, Amazon API Gateway, your database such as Amazon DynamoDB, etc. This allows it to be very well applied to create responsive and decoupled systems.

It has some features that are very useful, one being the automatic scaling. AWS Lambda automatically scales the number of executions based on the number of requests or events and can process from a few requests to thousands of requests per second.

Lambda charges for each execution, making it a pay-per-execution service. This means that you only pay for the time your code actually uses, not for server capacity that you don’t use. This makes it especially cost-effective for workloads that have varying or unpredictable traffic.

AWS Lambda fits in well into many situations, including file processing workflows where files are uploaded and automatically processed, backend API development, where it powers serverless REST APIs and data transformation pipelines, where it processes and transforms data between services in real-time.

Amazon API Gateway

2. Amazon API Gateway

The Amazon API Gateway is the “front door” for all client requests to serverless applications on AWS. It’s tasked with exposing backend services in a secure, scalable and managed manner, particularly when creating the modern serverless architectures.

One of its main features is the ability to develop and handle REST APIs and WebSocket APIs. This enables applications to enable both traditional request–response communication as well as real-time, two-way communication for applications such as chat systems or live update systems.

API Gateway can directly connect with other AWS services such as AWS Lambda, which allows requests to be directed to a serverless function without requiring any server infrastructure. This seamless integration streamlines application design and minimizes operational overhead.

API Gateway not only routes the traffic, but it also takes care of the cross-cutting issues including authentication, request throttling, and traffic management. These properties contribute to the security, performance, and resilience of APIs in different scenarios.

Some typical applications include creating web APIs for modern web applications that are scalable and low maintenance, especially for mobile app backends.

Amazon DynamoDB

3. Amazon DynamoDB

Amazon DynamoDB is a fully managed, serverless NoSQL database that provides high performance and seamless scalability, with no need to manage server hardware. It manages the provisioning, scaling and maintenance of infrastructure without any involvement from application developers; they develop applications without worrying about database operations.

One of the main advantages of it is automatic scaling. DynamoDB scales its capacity in response to the demand for capacity, so that it can maintain consistent performance whether the application is running a few requests or millions of operations per second. This is very good for applications of changing or rapidly changing traffic times.

DynamoDB’s performance is optimized for extremely low latency at any scale. It is designed to deliver quick and consistent response times, even under heavy loads or in distributed environments like applications with data stored in different locations worldwide.

One of the other big pros is it is completely serverless, so that there is no requirement to worry about servers, patching systems or infrastructure setups. AWS takes care of all underlying hardware and operational maintenance.

DynamoDB is frequently employed for session data in a Web application, for powering real-time applications like gaming or messaging, or for storing IoT data for large volumes of continuously streaming data that need to be processed efficiently.

4. Amazon S3 (Storage Layer)

Amazon S3 is a highly durable, fully managed object storage service that is often the storage backplane in cloud-native, serverless applications. It’s designed to provide strong durability guarantees, store and retrieve virtually unlimited amounts of data, and is a core service in AWS.

A major aspect of it is high durability object storage, which stores data as objects in buckets that are automatically replicated across several systems for durability and long-term storage. This will allow it to be used for critical data storage where loss cannot be tolerated.

In serverless workflows, Amazon S3 also serves as an important component, reacting to service calls with event notifications, such as those that are generated by AWS Lambda. For instance, a file can be automatically uploaded to or deleted from a bucket, which may trigger a function to process the event, resulting in an event-driven architecture.

One of the other key features is S3’s ability to scale automatically. Storage provisioning does not need to be done in advance, since the service can automatically scale up to accommodate increasing amounts of data to ensure the system’s performance without manual operation.

Typical examples of applications include file uploads for web and mobile applications, hosting of static websites with HTML, CSS and JavaScript files served directly from S3, and storing data lakes with large volumes of structured and unstructured data for analytics and machine learning applications.

Now let us dive into how the Serverless Architecture Works

Source: Multi Tier Archtecture

A typical AWS serverless workflow is as follows

1. A user requests an API.
2. A user makes a request of an API.
3. The request is sent to API Gateway.
4. The logic is executed in AWS Lambda.
5. The data is saved in DynamoDB or S3.
6. Response is returned to the user

This is an event driven approach, meaning that resources are only consumed when they are needed.

The advantages of Serverless on AWS

1. No Server Management (There are no servers to provision, patch or maintain.)

2. Automatic Scaling (Applications scale up on demand.)

3. Cost Efficiency (You only pay for computing time, not for idle computing capacity.)

4. High Availability (Fault tolerance and redundancy are automatically managed by AWS.)

Conclusion

With AWS serverless computing, developers can concentrate on application logic only, and AWS takes care of infrastructure complexity. You can use AWS Lambda, AWS API Gateway, AWS DynamoDB, AWS S3, and more to create highly scalable, efficient, and resilient systems with very little operational overhead.

When developing modern, cloud-native applications, serverless is one of the most effective architectures for you to use.