George Chow, Simba | Hadoop Summit 2016 San Jose
summary
Impala’s 2012 launch re-centers SQL on Hadoop and sparks a wave of open-source MPP engines (Hive, Impala, Drill, Phoenix) aiming for fast SQL analytics; the claim is that open development accelerates optimizer work that once took a decade. Multiple projects attempt a true converged transactional+analytic workload but none has broad production traction yet; notable efforts include Spanner, Presto, Splice Machine and traditional vendors experimenting with scale-out transactional systems.
AI-written summary · about
