Skip to content
BENIBIX Learn

Sliding Window

Data⚙️ ProcessingEvent Time and Windows

Also known as Hopping Window, Moving Window

Windows of one size that overlap. Each step gives a new result.

Think of it like

You look at the last ten minutes of a race. A minute later, you look at the last ten again.

Example

A site alerts when errors in the last 10 minutes pass 100. The check runs every minute on a sliding window.

Why it matters

It smooths out spikes and still updates often. Most live alerts and moving averages use it.

Definition

A sliding window has a fixed size, but a new one starts at a shorter step. So the windows overlap. A 10-minute window that slides every minute gives a fresh 10-minute result each minute. One event falls into several windows. It is used for moving averages and for alerts on the last few minutes. Some tools call it a hopping window.
The same events, grouped three ways: tumbling, sliding and session windows.

Common mistake

Adding the windows up for a total. Events sit in several windows, so the total is too high.

Key takeaways

  1. 1
    Size and step. Such as 10 minutes every 1.
  2. 2
    Overlap. One event, many windows.
  3. 3
    For moving views. Averages and alerts.

Compare