Is it me or has OLAP finally (re)emerged from the shadow of the transactional systems?
I was pleasantly surprised by AtScale‘s subdued emergence at June’s Hadoop Summit. (Refer to my previous summary for more.) This week, Socrata’s Evan Chan presented “Interactive OLAP Queries using Cassandra and Spark” at the Seattle Spark Meetup. Evan began work using Spark with Cassandra to do rollup and other traditional slice-and-dice queries. But he started prior to the emergence of Shark so he wasn’t using SQL. But no matter, he recognized what he was doing and called it out as OLAP.
It’s necessary for Big Data to address transactional systems first since there’s pressing need there to handle the volume and variety of data being generated. And data discovery is really an evolution of classical Data Warehousing’s ETL/ELT. So it’s no wonder that model-based analytics need to wait its turn. But it looks like OLAP’s moment in the wilderness is over. The game is on.