August 2015 Summaries

DataStax Community Edition 2.1.9 has been released as a maintenance update for Apache Cassandra 2.1 branch. It is now available for download from PlanetCassandra and includes bug fixes listed in the CHANGES.txt file. The upgrade notes in NEWS.txt highlight changes such as cqlsh displaying UTC times, commit log files no longer being recycled, and deprecation of set/getCompactionStrategyClass JMX methods.

DevCenter version 1.4.1 has been released and can be downloaded. This update introduces various enhancements and bug fixes, which are detailed in the improvements and bug fixes sections.

The C/C++ driver for Apache Cassandra and DataStax Enterprise has released version 2.1 GA, which includes support for client-side timestamps, full range of stream IDs, retry policies, idle connection heartbeats, raw paging state token exposure, ability to disable schema metadata, and several internal improvements. The driver now allows timestamps to be generated client-side, supports various retry policies, enables TCP_NODELAY by default, and provides the option to disable schema metadata retrieval and maintenance. Future releases will focus on supporting Apache Cassandra 2.2 and 3.0.

Apache Cassandra 2.2 introduced User Defined Functions (UDFs) and User Defined Aggregates (UDAs), which allow certain types of computation to occur server-side, directly on the Cassandra cluster. UDAs are aggregate functions that can be run directly on Cassandra and are composed of two parts: a UDF (called a 'state function' when in the context of UDAs) and the UDA itself, which calls the UDF for each row returned from the query. These features help to both conserve network bandwidth and cut down on client-side computations by performing functions and aggregations directly at the data source.

Apache Cassandra 2.2 introduced User Defined Functions (UDFs) and User Defined Aggregates (UDAs), which allow certain types of computation to occur server-side, directly on the Cassandra cluster. This can reduce network traffic to the client and reduce client-side resource utilization. UDFs are identified by their signature: the combination of its keyspace, function name and arguments. They can be used in scenarios such as filtering items based on the number of available states or retrieving specific information from User Defined Types (UDTs). By supporting JSR 223 scripting languages, UDFs allows you to write server-side executable scripts in your favorite language among many supported languages.