AWS fundamentals: What is S3?
In this blog post series we’ll take a closer look at Amazon web services (AWS) and all their Serverless related services in detail.
Today we look at S3. What is it and how does the Serverless framework use it?
This chapter is a free sample taken from our “Learn Serverless” book.
S3 stands for “Simple Storage Service” and helps you store your data in the Amazon cloud. This data can be everything. Your photos, your code, your zipped music archive. AWS makes sure that the infrastructure is administrated so that you can always access your files without problems.
Furthermore you have “infinite” storage available. This means that you don’t have to think about the size before designing your infrastructure. Just upload your data and you are done.
Permission management, revisions and many more is also implemented in S3. You can define who can access what at every time. You have accidentally removed an object from your bucket? No problem. You can restore it easily (although you have to enable it when creating your bucket).
Apart from the web console you can use standardized protocols to upload and access your S3 data.
Buckets
Buckets give you a way to organize your files. Just create a bucket (e.g. „photos“) and upload everything related there. Permissions, Revisions and other settings can be defined on a bucket level. So you can e.g. share your photos online while keeping your uploaded code files (in your „code“ bucket) a secret.
How Serverless uses S3
Up until v0.4 Serverless uses S3 as a storage for your project related code (such as e.g. Lambda functions). Upon project creation Serverless creates an S3 bucket for your project where everything related and important is uploaded. The bucket also functions as a kind of backup. This way you can still access your Lambdas even if your development machine burns down to ashes.
Serverless has removed the S3 bucket dependency in v0.5 (The maximum storage space for Lambda functions was increased recently. Therefore no need for a separate S3 bucket which stores all the versioned Lambdas)).
But you could still use S3 buckets to e.g. host your fronted which talks to your Serverless backend.
