May: Overview of how Cassandra provides a distributed, fault tolerant and scalable database cluster
Apache Cassandra is scalable, distributed and fault tolerant cluster database originally created by Facebook and now a top level Apache project http://cassandra.apache.org/. It borrows the consistency model from Amazon Dynamo and the data model from Google Big Table. Twitter, Netflix, Reddit and digg are amongst the large scale sites that rely on and contribute to the technology.
In this talk I'll discuss
- how data is replicated around the cluster;
- the Write and Read path through the server;
- what Eventual Consistency means and how Read Repair, Hinted Handoff and Anti-Entrophy work;
- the problem with Distributed Deletes;
- Distributed Counters and the Column Orientated data model.
Hopefully by the end of the talk you'll have a good idea about how and why Cassandra is different to a traditional RDBMS and have some ideas of how to build applications with it while avoiding some of the common mistakes.
Speaker: Aaron Morton
Aaron Morton recently left a role at Weta Digital to spend more time contributing to the Cassandra project and helping people deliver scalable and reliable applications built on the platform. He is on the twitters at @aaronmorton.
When
Wednesday, May 18: 5:00 for 5:15 - 6:30pm
Where
This meeting is hosted by:
Equinox
Level 5
Equinox House
111 The Terrace (or 222 Lambton Quay through Amazon)
Wellington
The Java User Group thanks our host!