Scale Your Application to Infinity
summary
IPS maps client XML schemas to a single Cassandra-backed model that services use for hundreds of Intuit apps, persisting attributes in Cassandra and large payloads in S3 while provisioning only 6 Cassandra tables across the product. The platform runs 10 production clusters active‑active in two regions, the largest at 117 nodes per region supporting ~220k PPS of CRUD+LIST. Indexes are binary exact‑match sparse, CDC publishes to Kafka with SQS backup, and automated node self‑heal plus daily repairs keep clusters healthy.
AI-written summary · about
chapters (11)
- 0:00Scaling apps with Cassandra
- 1:24IPS value for developers
- 4:12IPS NoSQL deep dive
- 8:52Logical schema in IPS NoSQL
- 13:34Relationship modeling in IPS
- 17:11Cassandra data modeling best practices
- 19:20Cassandra cluster setup on AWS
- 21:56Optimizing Cassandra for GC
- 24:51Cassandra operational best practices
- 26:39IPS NoSQL team and tools
- 27:02Q&A on Cassandra migration
related talks
46mCassandra Serving Netflix @ Scale
21mKeynote - Database Infrastructure at Instagram Scale
23mArchitecting for Performance, Scalability and Real-time Data with AWS
31mCassandra for giant 3D tables
35mReplacing Cassandra: A Digital Transformation for the World’s Largest Digital Identity Network
26mBuilding Financial Systems on Eventually Consistent DBs
53mApache Cassandra. Inception - all you need to know
21mBuilding Distributed Platforms at Extreme Scale on Fast Changing Data