Apache Cassandra 1.2 introduces a number of changes for performance and usability. None will impact developers as much as the new CQL3 query language and it's underlying transport. This talk will detail the new architecture in such a way as to give current users a deeper understanding, while exposing new users to the fundamentals of Cassandra's API, storage architecture and clustering internals.
Apache Cassandra 1.2 introduces a number of changes for performance and usability. None will impact developers as much as the new CQL3 query language and it's underlying transport.
Given the scope of the changes, this talk will take a detailed look at the client and transport components of Cassandra through comparing the current Thrift-based API and transport with the new Netty-based transport and CQL3 language.
In doing so, attendees will be exposed to the fundamentals of Apache Cassandra's internal and external storage APIs, the mechanics of clustering via the Gossip protocol used for inter-node communication, and how these components interact (or dont) with the existing clients.
This talk will give attendees experienced with Cassandra the knowledge necessary to make the best choice on which protocol to use for their own development projects given their requirements. Attendees new to Apache Cassandra will come away with an understanding of the storage model, distribution characteristics and query architecture.