Skip to content

AWS Planning Document

Jump to bottom
Ryan Connell edited this page Mar 14, 2018 · 1 revision

AWS vs Data Abstraction Layer

There are two possible routes we can take when designing a Data Abstraction Layer for S3 (Simple Storage Service) on AWS

  1. There is no official Rust SDK so instead we could use an unofficial one to make the process of authenticating and dealing with S3 easier.
  2. We can interact with S3 using a REST API.

While both options are doable I think it would be better to use an unofficial Rust SDK so that we don’t need to worry about implementation details. A library such as https://github.com/rusoto/rusoto might be useful for this.

Deploying Containers to AWS

We can deploy containers running the master and workers to EC2. This can also be done in two different ways:

  1. We could write code in rust using the aforementioned library to deploy and control our containers.
  2. We can interact with EC2 using the Amazon EC2 command line interface.
Clone this wiki locally