Resources for API304: Building rate-limited solutions on AWS
Have you ever built a wonderfully scalable service only to have it overwhelm an existing backend service that can't scale? Or maybe you have to integrate with a third party API that limits you to a few transactions per second? If so, this session is for you. We will discuss the best ways to build rate limiting into your systems to improve their overall reliability. After a quick overview of the AWS tools to build in rate limiting we will dive into the details of three different implementations.
Using Load Shedding to Avoid Overload
Amazon Builders Library
This Amazon Builders Library paper looks at the principals of system overload and how to use load shedding to avoid it.
Leaky Bucket Algorithm
Wikipedia
Wikipedia article on the leaky bucket algorithm
Token Bucket Algorithm
Wikipedia
Wikipedia article on the token bucket algorithm
Controlling Concurrency in Distributed Systems Using AWS Step Functions
Blog
A blog along with code samples to control concurrency using Step Functions
The three most important AWS WAF rate based rules
Blog
A blog that details important AWS WAF ratbe based rules.
Throttling multi-tenant Rest APIs with API Gateway
Blog
A two part blog that explores the details rate limiting with API Gateway
ServerlessLand
Code Examples
Guides, blogs, code samples, and more!
Presented by:
Fairness in Multi-Tenant Systems
Amazon Builders Library
This Amazon Builders Library paper looks at a few approaches Amazon has taken to manage API requests to its systems to avoid overload by implementing API rate limiting (also referred to as throttling or admission control). Without these kinds of protection, a system becomes overloaded when more traffic flows into the system than it is scaled to handle at that time.