Cassandra - an introduction
summary
Cassandra is a distributed, Java‑based partitioned row store (open sourced by Facebook in 2008) that shards data on a hash ring using partition keys and a Murmur3 partitioner, nodes coordinate via gossip protocol, and you scale by adding machines while ranges and replicas move automatically. Writes are fast because of commit log + memtable -> SSTable and background compaction; durability uses replication factor per keyspace (common RF=3) and tunable consistency level per query (ONE, QUORUM, ALL). Tooling: ccm for local clusters, works with Spark/Kafka/Solr; watch out for GC, repair jobs and disk/SSD needs.
AI-written summary · about
chapters (9)
- 0:00Introduction and motivation
- 2:14Relational DB limitations
- 7:26Cassandra history
- 9:35Architecture and data
- 13:25Replication and availability
- 16:01Consistency levels
- 19:27Read and write mechanics
- 23:55Data modeling strategies
- 26:33Tools, complaints, and tips
related talks
53mApache Cassandra. Inception - all you need to know
45mCassandra Architecture: Understanding, Scaling and Optimizing
32mCassandra at Yammer #bbuzz
50mNoSQL Databases You Have to Know
31mCassandra for giant 3D tables
35mThe What and Why of Cassandra Datamodelling at Scale
27mSponsored Session: How is Amazon Keyspaces (for Apache Cassandra) Architected to Deliver Performance at Any Scale?
30mApache Cassandra: What’s New and What’s Next