Pig Team is happy to announce Pig 0.4.0 release!
Pig is a Hadoop subproject that provides high-level data-flow language and an execution framework for parallel computation on a Hadoop cluster. More details about Pig can be found at http://hadoop.apache.org/pig/. This release introduces two new types of join. The skewed join improves join performance for the data with large skew in the join key. The merge join improves performance for the case where both inputs are sorted on the join key. The release also includes support for outer join. The details of the release can be found at http://hadoop.apache.org/pig/releases.html The publishing of this release has been delayed due to problems with Apache infrastructure that prevented us from publishing the updated site. Olga