Running Cassandra

Java: Server
Location: A105
Average rating: ***..
(3.00, 3 ratings)

Apache Cassandra is a powerful new distributed database system that, when used correctly, provides a simple framework for managing large, rapidly changing, and/or high value datasets. But Cassandra is a bit rough around the edges. In particular, the system has a reputation for being unforgiving when misconfigured or burdened with unusual work loads. This talk will cover the peculiarities of the Cassandra API, what the system is good at and what it’s not, and how Cassandra is developing into a more friendly general purpose database. I’ll share tips about how we’ve successfully run Cassandra in production at SimpleGeo for almost two years, along with the mistakes we’ve made and success we’ve enjoyed along the way. I’ll also cover the core concepts underlying Cassandra’s architecture, why they’re important, and how they’ll impact applications that use the system. Hopefully you’ll come away with a deeper understanding of what I think is an very exciting, hackable, and extremely powerful new technology.

Photo of Mike Malone

Mike Malone


Mike Malone is lead architect at SimpleGeo, where he works on building and integrating scalable systems that power the company’s location platform. Since joining SimpleGeo, Mike has been working to ensure operational continuity in the face of rapid growth, partial system failures, and traffic bursts. Before joining SimpleGeo, Mike helped build the microblogging web site Pownce, where he learned a lot about the technical and social difficulties of scaling an online community. After Pownce’s acquisition by Six Apart in 2008, Mike worked on the TypePad platform team, where he gained a great deal of experience building RESTful web services. In his spare time Mike enjoys tinkering with new technologies. When he’s not on the computer, you can probably find him hanging out with his girlfriend, Katie, and their friends at a good bar.