Sliding Window

The Sliding Window strategy tracks individual request timestamps and counts requests within a rolling time window. This eliminates the boundary burst problem of Fixed Window.

How It Works

Time ──────────────────────────────────────────────>
     ←────── window (60s) ──────→
     •    •  •    •  •  •  •  •  •  •  (10 requests)
     ↑                           ↑
   oldest                      now

Instead of resetting at fixed intervals, the window "slides" with time. Only requests within the last windowMs milliseconds are counted.

Configuration

Options for the Sliding Window strategy:

Usage

import { slidingWindow } from '@ratelock/local'

const limiter = await slidingWindow({
    limit: 100,
    windowMs: 60_000,
})

const result = await limiter.check('user:123')
// { allowed: true, remaining: 99, reset: 1716000060000, windowStart: 1716000000000, windowEnd: 1716000060000 }

Result Fields

The check result returned by the Sliding Window limiter:

Pros and Cons

Pros:

• No boundary burst problem
• More accurate rate limiting
• Smooth request distribution

Cons:

• Higher memory usage (stores individual timestamps)
• Slightly slower than Fixed Window
• Requires cleanup of expired entries

When to Use

• When you need precise rate limiting
• When users exploit Fixed Window boundaries
• APIs with strict rate limit requirements
• When accuracy matters more than memory usage