Question 1

How does Hammer compare to ExRated for Elixir rate limiting?

Accepted Answer

Hammer is more flexible with multiple backends and algorithms, while ExRated is simpler and uses ETS with a fixed window. Hammer is better for complex or distributed needs, but ExRated might be sufficient for basic, single-node rate limiting with less setup.

Question 2

What is the best backend for high-performance single-node rate limiting in Hammer?

Accepted Answer

For single-node performance, the Atomic or ETS backends are recommended. ETS is the default and offers good speed for most cases, while Atomic is optimized for high-throughput operations, as highlighted in the performance benchmarks and backend descriptions.

Question 3

How to set up Hammer with Redis for distributed rate limiting?

Accepted Answer

Add the hammer-backend-redis dependency to your mix.exs, configure the Redis connection in your application, and use the Hammer.Redis backend. Detailed setup instructions are available in the backend's GitHub repository, but it requires extra steps compared to the default ETS backend.

Question 4

Can I use Hammer with Phoenix to rate limit API requests?

Accepted Answer

Yes, but it requires custom integration since Hammer doesn't provide built-in Phoenix middleware. You'll need to write plugs or controllers that call Hammer's hit function, scoped by user IDs or IPs, as shown in the example usage in the README.

Question 5

What algorithm should I use for burst-tolerant rate limiting in Hammer?

Accepted Answer

Use the Token Bucket algorithm, which allows controlled bursts while maintaining an average rate. It's available for most backends like ETS, Atomic, and Redis, and is ideal for APIs or gaming systems that need to handle sudden traffic spikes, per the algorithm comparison guide.

Question 6

Is Hammer suitable for microservices with high concurrency?

Accepted Answer

Yes, especially with distributed backends like Redis, Hammer can handle high concurrency across nodes. However, you must configure it carefully and consider performance trade-offs, as some algorithms like sliding windows have higher overhead, which the README notes in its scalability section.

hammer

What is hammer?

Overview

Use Cases

Best For

Related Projects

Found a gem we're missing?

Not Ideal For

Pros & Cons

Pros

Cons

Frequently Asked Questions