How Cassandra 4.X Help in Cutting Cost Mission
summary
Self-hosted cluster shrinks from 42 nodes to a footprint that drops Cassandra from the top bill to 20th position, achieving about 69% TCO reduction. Savings come from moving to Graviton3 with Java 11, switching EBS from gp2 to gp3 + Zstandard compression (≈25% disk savings, EBS cost cut ~2x on cold tables), dropping Spring Data ORM due to prepare-statement bugs, enabling inter-node/client compression, and using S3 incremental backups with Gateway endpoints.
AI-written summary · about
chapters (13)
- 0:00Cost-saving with Cassandra 4
- 1:49Focus on AWS and self-hosted
- 2:53Java 11 and Graviton processors
- 5:47Storage cost savings with EBS
- 9:16Linux read-ahead for EBS
- 11:11Network traffic costs
- 12:39Compression between nodes
- 14:52Virtual tables for tracking
- 15:58Spring Data optimizations
- 17:26Vault integration consistency
- 19:35End of life risks
- 21:05Final cost-cutting strategies
- 22:59Summary and Q&A
related talks
35mReplacing Cassandra: A Digital Transformation for the World’s Largest Digital Identity Network
17mCloud Compute Cost Showdown
39mMaking Cassandra Faster in Cloud-native architecture
24mStaying Ahead of the Curve with Spring and Cassandra 4
23mArchitecting for Performance, Scalability and Real-time Data with AWS
40mLatency, Throughput & Fault Tolerance: Building a Modern Database
52mHow Netflix Provisions Optimal Cloud Deployments of Cassandra
01h 05mRoundtable: Cloud Databases as a Service