Thanks Andrew -- there is a proliferation of interactive query systems for Hadoop these days. Hive is cool but Hive is slow. Cloudera has Impala Hortonworks has Stinger I'm a fan of Shark and the whole Spark ecosystem. Once we solve the ingest problem, we'll want to work on this class of problems. -Toby On Wed, Nov 6, 2013 at 10:54 AM, Andrew Otto <acotto(a)gmail.com> wrote: