Etherpad Scale Calculator

This page is designed to help you understand your hardware requirements for Deploying Etherpad at scale

The performance here is a rough guideline, we hope(not tested) over the coming months (by 2021) to have a roughly 5x improvement in performance.

We haven't tested Database performance with load across multiple nodes as of yet - We hope to have this research completed over the coming months



An author is someone who writes at ~40wpm on a pad. This is a very active author. Most authors do not hit this average but we must account for it to handle edge cases. User productivity tends to drop with more than 3 concurrent authors on a pad


A lurker is someone who is recieving pad updates but doesn't contribute


More info

Stability can be improved by increasing memory heap size availability if memory availability runs low

Each etherpad instance/node can currently handle ~10k msgs/sec