Implementing the Future of PostgreSQL Clustering with Tungsten
|From PostgreSQL Conference West Speakers|
Users have traditionally used database clusters to solve database availability and performance requirements. However, clustering requirements are changing as hardware improvements make performance concerns obsolete for many users. In this talk I will discuss how the Tungsten project uses master/slave replication, group communications, and rules processing to develop easy-to-manage database clusters that solve database availability, protect data, and address hardware utilization. Our implementation is based on existing PostgreSQL capabilities like Londiste and WAL shipping, which we eventually plan to replace with our own log-based replication. Come see the future of database clustering with Tungsten!