Hey all, In case you missed it, 3.0.0 is out!
http://www.cascading.org/2015/06/08/cascading-3-0-release/ <http://www.cascading.org/2015/06/08/cascading-3-0-release/> It took over a year and a half of work to get this release out. Though it feels anti-climatic at this point having been knee deep in it for so long, Cascading 3.0 is a major milestone. The Apache Tez support is important, but as importantly we have proven we can execute the Cascading intermediate model over multiple distinct computation execution models. This was achieved by isolating the API coupling and embody the semantic mapping from Cascading to the target execution model via rules in the query planner. Some of the thought behind the planner can be read here: http://www.cascading.org/2014/09/18/the-cascading-3-0-query-planner/ <http://www.cascading.org/2014/09/18/the-cascading-3-0-query-planner/> Still lacking is good documentation on leveraging the query planner for application specific optimizations. I plan to incrementally build this knowledge base up as we begin work on additional platform support, and work with other teams like Apache Flink and Hazelcast (https://hazelcast.atlassian.net/wiki/display/COM/HEP+6+-+Cascading+Integration <https://hazelcast.atlassian.net/wiki/display/COM/HEP+6+-+Cascading+Integration>), and obviously Apache Spark. Thanks every who has tested and/or offered guidance!! chris p.s. Do note that the current GA release of Driven does not support Cascading 3. If interested in being notified when Driven EAP can be used with Cascading 3 and Apache Tez, please sign up here: http://www.cascading.org/join/ <http://www.cascading.org/join/> — Chris K Wensel [email protected]
