Hi All, Hazelcast now also have a MapReduce implementation. May be we can change hive so it can also submit jobs to Hazelcast in addition to MapReduce. Then users can load lot of data to hazelcast and process them with HiveSQL.
1. It will enable users to also do in memory computing using Hive SQL 2. It will enable ad-hoc queries that return fast results over a large dataset using HiveSQL IMO it will be a pretty tool In-memory computing scenario. WDYT? --Srinath p.s. One concern is when we support Spark, it will also have much similar support via .cache() in RDDs. So this might be a redundant feature at the time. -- ============================ Director, Research, WSO2 Inc. Visiting Faculty, University of Moratuwa Member, Apache Software Foundation Research Scientist, Lanka Software Foundation Blog: http://srinathsview.blogspot.com twitter:@srinath_perera Site: http://people.apache.org/~hemapani/ Photos: http://www.flickr.com/photos/hemapani/ Phone: 0772360902
_______________________________________________ Architecture mailing list [email protected] https://mail.wso2.org/cgi-bin/mailman/listinfo/architecture
