Cassandra Architecture: Understanding, Scaling and Optimizing
summary
Clear, concrete overview that explains how Cassandra achieves single-digit millisecond latency, linear scaling by consistent hashing and token ranges, and multi‑DC replication with replication factor 3 examples and quorum math. It walks through the actual write/read pipeline — commit log -> memtable -> SSTable and compaction — and notes operational realities like relaxed fsync defaults and Windows support limits. Watch if you want mechanical details and tradeoffs; skip if you only need high‑level marketing.
AI-written summary · about
chapters (25)
- 0:00Introduction
- 2:20Why Cassandra
- 5:50The Cap Theorem
- 10:40Architecture
- 14:28Strong Consistency
- 18:45Splitting Data
- 22:10Address Space
- 22:40Consistent Hash
- 23:40Linear Scalability
- 24:53Replication Factor
- 25:52Cassandra QA
- 34:08MultiDC Replication
- 34:39Consistency Levels
- 35:38ReadWrite Path
- 37:05Commit Log
- 39:47SSD Table
- 40:21Super Duper
- 41:28Compaction
- 42:06Read
- 42:45Partition
- 43:17Keycase
- 43:28Rowcase
- 43:39Bloom Filter
- 43:54Summary
- 44:04Key takeaways
related talks
29mCassandra - an introduction
53mApache Cassandra. Inception - all you need to know
40mLatency, Throughput & Fault Tolerance: Building a Modern Database
23mDeciphering Cassandra’s Prophecy: Preventing the Fall of Your Application
35mThe What and Why of Cassandra Datamodelling at Scale
45mFrom SQL to Cassandra
30mApache Cassandra: What’s New and What’s Next
27mSponsored Session: How is Amazon Keyspaces (for Apache Cassandra) Architected to Deliver Performance at Any Scale?