rueidislimiter
This module provides an interface for token bucket rate limiting with precise control over limits and time windows. Inspired by GitHub's approach to scaling their API with a sharded, replicated rate limiter in Redis (github.blog).
Features
- Token Bucket Algorithm: Implements a token bucket algorithm to control the number of actions (e.g., API requests) a user can perform within a specified time window.
- Customizable Limits: Allows configuration of request limits and time windows to suit various application requirements.
- Distributed Rate Limiting: Leverages Redis to maintain rate limit counters, ensuring consistency across distributed environments.
- Reset Information: Provides
ResetAtMs
timestamps to inform clients when they can retry requests.
Installation
To install the rueidislimiter
module, run:
go get github.com/redis/rueidis/rueidislimiter
Usage
Basic Rate Limiting Example
The following example demonstrates how to initialize a rate limiter with a custom request limit and time window, and how to check and allow requests based on an identifier (e.g., a user ID or IP address):
package main
import (
"context"
"fmt"
"time"
"github.com/redis/rueidis"
"github.com/redis/rueidis/rueidislimiter"
)
func main() {
client, err := rueidis.NewClient(rueidis.ClientOption{
InitAddress: []string{"localhost:6379"},
})
if err != nil {
panic(err)
}
// Initialize a new rate limiter with a limit of 5 requests per minute
limiter, err := rueidislimiter.NewRateLimiter(rueidislimiter.RateLimiterOption{
ClientOption: rueidis.ClientOption{InitAddress: []string{"localhost:6379"}},
KeyPrefix: "api_rate_limit",
Limit: 5,
Window: time.Minute,
})
if err != nil {
panic(err)
}
identifier := "user_123"
// Check if a request is allowed
result, err := limiter.Check(context.Background(), identifier)
if err != nil {
panic(err)
}
fmt.Printf("Allowed: %v, Remaining: %d, RetryAfter: %v\n", result.Allowed, result.Remaining, result.RetryAfter)
// Allow a request
result, err = limiter.Allow(context.Background(), identifier)
if err != nil {
panic(err)
}
fmt.Printf("Allowed: %v, Remaining: %d, RetryAfter: %v\n", result.Allowed, result.Remaining, result.RetryAfter)
// Allow multiple requests
result, err = limiter.AllowN(context.Background(), identifier, 3)
if err != nil {
panic(err)
}
fmt.Printf("Allowed: %v, Remaining: %d, RetryAfter: %v\n", result.Allowed, result.Remaining, result.RetryAfter)
}
API
NewRateLimiter
Creates a new rate limiter with the specified options:
ClientOption
: Options to connect to Redis.
KeyPrefix
: Prefix for Redis keys used by this limiter.
Limit
: Maximum number of allowed requests per window.
Window
: Time window duration for rate limiting. Must be greater than 1 millisecond.
limiter, err := rueidislimiter.NewRateLimiter(rueidislimiter.RateLimiterOption{
ClientOption: rueidis.ClientOption{InitAddress: []string{"localhost:6379"}},
KeyPrefix: "api_rate_limit",
Limit: 5,
Window: time.Second,
})
Check
Checks if a request is allowed under the rate limit without incrementing the count.
result, err := limiter.Check(ctx, "user_identifier")
Returns a Result
struct:
Allowed
: Whether the request is allowed.
Remaining
: Number of remaining requests in the current window.
ResetAtMs
: Unix timestamp in milliseconds at which the rate limit will reset.
Allow
Allows a single request, incrementing the counter if allowed.
result, err := limiter.Allow(ctx, "user_identifier")
AllowN
Allows n
requests, incrementing the counter accordingly if allowed.
result, err := limiter.AllowN(ctx, "user_identifier", 3)
n
: The number of requests to allow.
Implementation Details
The rueidislimiter
module employs Lua scripts executed within Redis to ensure atomic operations for checking and updating rate limits. This approach minimizes race conditions and maintains consistency across distributed systems.
By utilizing Redis's expiration capabilities, the module automatically resets rate limits after the specified time window, ensuring efficient memory usage and accurate rate limiting behavior.
For more information on the design and implementation of Redis-based rate limiters, refer to GitHub's detailed account of scaling their API with a sharded, replicated rate limiter in Redis (github.blog).