New ustat algorithm and bulk load integration This is the initial phase of the Update Statistics change to use counting Bloom filters and a Hive backing sample table created during bulk load and amended whenever a new HFile for a table is created. All changes are currently disabled pending some needed fixes and testing.
changes to improve plan quality of OSS queries. b1 and b2: NJ was not being explored because RHS of NJ was another join and didn't have index information to check for keyless NJs. This change helped OSS b1 and b2 queries to get better NJ plans.
b6, b6, and b8: Optimizer chose Hj + MDAM plan where as NJ + regular scan runs better. The costing code didn't account for probes being done to get beginKey for missing _SALT_ column. This change will include probing cost.
QA: Regressions completed performance workloads including HPIT showed no negatvie impact OSS queries showed significant improvement.