Spring Cloud Gateway - Redis Rate Limiter

The Spring Cloud project is part of the Spring ecosystem. It's a way for developers to easily implement services from patterns that were already implemented in a cloud-native mindset that's scalable, resilient and fault-tolerant.

In this article, we're going to go over the implementation of the AbstractRateLimiter; RedisRateLimiter. We'll create an API that greets the user while rate limiting the greetings!

Let's start by creating a new project with three dependencies:

• spring-cloud-gateway
• spring-data-redis
• spring-security

Create a Default user:

This bean creates a User with default credentials that will be used for rate limiting.

Create a New Route:

This route will match whatever URI that has a path "/hi" and will forward the request to "localhost:8080/hello"

Now that we have the RouteLocator handling the request we can try it out.

The 404 is because of the request being routed to /hello that doesn't exist.

Create the Rest Controller:

If we run the curl command again we get a successful response now that the request can be routed to the destination.

Add the Rate Limiter:

The rate limiter takes two values:

• replenishRate: The number of requests allowed by a user every second.
• burstCapacity: The maximum number of requests allowed by a user every second.

To better understand these values check the docs and the algorithm behind the rate limiting.

Update The Route To Include The Rate Limiter:

Now when we try the same curl command again we notice three new headers:

• X-RateLimit-Remaining (Used by clients to throttle their requests) 
• X-RateLimit-Burst-Capacity (Defined in the rate limiter above)
• X-RateLimit-Replenish-Rate (Defined in the rate limiter above)

Test The Rate Limiter:

The code for this is hosted on GitHub: JadCham/spring-redis-ratelimit-demo

If you're interested to learn more about rate limiting take a look at stripe's article on rate limiters.