Why Serverless!!

This application uses a mix of Managed and Serverless AWS services. This is not a perfect implementation, but I think it is still a better implementation considering the use case.

Serverless allows programmers (and the operations teams they work with) not to spend time worrying about setting up, tuning and scaling applications to work in a certain way - that’s all looked after at the backend by the cloud provider.

Let the undifferentiated heavy lifting manage by the AWS and we provide the logic only where needed.

This is one of those use cases that fit into the Serverless architecture pretty well. Most internal file management applications have limited and predictable traffic patterns.

The application itself will be used by a predictable number of users and the kind of files stored inside will also vary. Some of the common considerations I have taken are:

• Total number of users

• Total number of files vs total number of files per user

• User and File security

• The sensitivity of the documents

This application is not built to serve millions of users or documents. It is rather meant to prove the concept of implementing a moderately complex use case with serverless and managed services together. Although this can be further extended to scale appropriately.

If we look at the weightage of the application, heavy features are around File Storage and Security. We are using S3 to manage the file storage, website hosting, and Amazon Cognito for user management and role-based security.

There are a few points at which we need to run custom logic and that is where we will be using Lambda functions. The reason we can integrate these services together is due to their native integration.