Cassandra for Metadata Storage: Successes and Failures
summary
Describes using Apache Cassandra to store metadata for an exabyte-scale S3-compatible object storage. Covers challenges of metadata (concurrent writes/deletes, versioning) and complications introduced by active‑active, eventual‑consistency operation across two datacenters, and how Cassandra was applied to address those issues.
AI-written summary · about
chapters (9)
- 0:00Introduction
- 8:12First Cassandra experience
- 10:43Data loss history
- 18:14Zombie data problem
- 23:12Partitioning issues
- 37:56Slow performance diagnosis
- 44:23Corrupted Cassandra data
- 49:36Cassandra unpredictability
- 50:20Q&A
related talks
01h 01mOK S3
29mCassandra - an introduction
53mApache Cassandra. Inception - all you need to know
31mCassandra for giant 3D tables
46mJourney of metadata
33mUsing Apache Cassandra as your Document Database
46mApache Cassandra: 8 years in production
27mSponsored Session: How is Amazon Keyspaces (for Apache Cassandra) Architected to Deliver Performance at Any Scale?