Cassandra Connection Pool

The Cassandra Connection Pool is derived from the jdbc-pool project developed by Filip Hanik as part of Tomcat 7. Its purpose is to provide a flexible and scalable method for pooling connections to a Cassandra cluster. It provides the following features:

  • possibility to define the maximum number of active connections, maximum and minimum number of idle connections
  • connection validation on create, borrow, release and periodically on idle connections
  • host failover policy (fail fast, single host, all hosts)
  • automatic retrieval of the addresses of the hosts belonging to a Cassandra cluster
  • JMX bean for monitoring and configuring via management consoles
  • configurable via IoC containers (e.g. Spring)