Cassandra gets a clean up and speed up in release 2.1

Big Data users who swear by Cassandra should seriously consider upgrading to the new version.
Written by Steven Vaughan-Nichols, Senior Contributing Editor

The Apache Software Foundation (ASF) announced on September 11 at the Cassandra Summit, the release of Apache Cassandra v2.1, the open-source, Big Data distributed database.


Cassandra, which was created by Facebook, has proven to be very popular in the enterprise. Apache claims that "Cassandra powers hundreds of applications across dozens of industries that demand high performance at scale, and is in use at Adobe, Comcast, eBay, GE, GoDaddy, HP, IBM, Intuit, Netflix, Safeway, Sky, Sony, Spotify, Travelocity, and The Weather Channel."

The latest Cassandra only brings in a few new features. Duyhai Doan, a Cassandra expert at Orange Mobile, said in a press release, "For developers, the biggest game changer in 2.1 is the introduction of CQL3 tuple and user defined type (UDT) as both pave the way for new data model patterns and usage. In addition, the ease of use core value for Cassandra is fulfilled in this release." This is the first version of Cassandra which is suitable for production use on Windows.

The real difference between Cassandra 2.0 and 2.1, according to Brian O'Neill, CTO at Health Market Science, is that "The 2.1 release tightens the nuts and bolts and drops nitro into that engine to make those analytics blazing fast."

Apache agrees that Cassandra 2.1 is much faster. "Every release reinforces why Cassandra is the database of choice for growing enterprises," said Jonathan Ellis, Vice President of Apache Cassandra and CTO of DataStax in a statement. "With 2.1 delivering over 50% better performance over 2.0's already-strong numbers, Cassandra 2.1 lets our users continuously improve their engagement with their customers at the highest speeds to date."

It had better. DataStax has bet its future on Cassandra. With over a hundred million dollars in new investment and alliances with Accenture and Google DataStax appears to have made a safe bet.

Related Stories:

Editorial standards