Popular social networking service https://discord.com/company migrated its messages cluster from the open source https://thenewstack.io/4-common-questions-we-hear-about-apache-cassandra/ to distributed data store https://www.scylladb.com/?utm_content=inline-mention and reduced latencies from 200 milliseconds to 5 milliseconds. The https://www.scylladb.com/scylladb-summit-2023/presentations/ opened last Wednesday with a talk, “How Discord Migrated Trillions of Messages from Cassandra to ScyllaDB,” given by Discord Senior Software Engineer https://www.linkedin.com/in/bo-ingram-1a069275/.
In order to improve efficiency and reduce latencies, Discord engineers built a data service library to streamline queries going to and from the database.
Discord experienced cascading latencies caused by Cassandra’s hot partitions caused cascading latencies when high traffic was driven to one partition.
The core messaging database was the holdout because “we didn’t want to learn all our lessons with our messages database,” Ingram said.