The new year is here, and so is Cassandra 1.2.0!

Key improvements include:

• Virtual nodes, which improve the granularity of capacity increases and dramatically improve repair and rebuild times in larger clusters. See also this post on upgrading an existing cluster to vnodes.
• CQL3 improvements, notably the addition of collection types, queryable system information, and a CQL-native protocol.
• Request tracing is available to both CQL and classic Thrift requests, and can also be managed programatically.
• Atomic batches address the possibility of mid-batch coordinator failure.
• Configurable policies for disk failure
• Last but not least, many performance improvements to memory usage, column indexes, compaction, streaming, startup time, and more.

2013 is going to be another good year to be a Cassandra user!

Edit: DataStax documentation for Cassandra 1.2 is now up.