Scaling Stories from Discord
The Engineering & developers at Discord moved from MongoDB to Cassandra once to process billions of messages, now they recently published about migrating from Cassandra to ScyllaDB to process trillions of messages.
"177 Cassandra nodes to 72 ScyllaDB nodes
Latency to fetch went down from 40–125ms to 15 ms
Message insert went from 5–70ms to consistent 5 ms."
I'm interested in checking all these related materials:
- Blog post from the Discord team: How Discord Stores Trillions Of Messages (Online discussions)
- An expected video from Hussein Nasser about it.
- Benchmark (from ScyllaDB): ScyllaDB vs Apache Cassandra.
Also consider
- The Discord blog (RSS) is a gem.
- I'm always fascinated by the technical details of this kind of engineering migrations. So, I check from time to time materials like this List of Tech Migrations maintained by Kiswono Prayogo.
Originally published on Bear.