Hadoop

There are those out there who say that massively parallel processing (MPP) appliances aren't really related to the big data concept. Others, meanwhile, say that MPP appliances are indeed true big data appliances. Those who say that MPP and big data go hand in hand, however, got a little extra proof for their side of the issue thanks to Microsoft's (News - Alert) recent announcement of PolyBase at the PASS Summit last week.

The announcement at the PASS Summit, commonly regarded as the primary SQL Server conference, detailed the story behind PolyBase. PolyBase is an integration of Microsoft's MPP product with both SQL Server Parallel Data Warehouse and Hadoop. Set for release by June 2013, Polybase will not only allow data that's in Hadoop to be queried with Structured Query Language (SQL), it will even integrate the data in the Distributed File System of Hadoop with the engine on the relational database.

While this is some pretty good news right here for those looking for that coming together of SQL and Hadoop, Microsoft's Polybase isn't actually the first such attempt to bring SQL with Hadoop or MapReduce. Programs like Cloudera, Rainstor and Teradata (News - Alert) Aster all brought in that combination of SQL and Hadoop. Yet Polybase--while not a hundred percent original--did bring in some new and exciting ways to use the program. For instance, the first phase of PolyBase's rollout will focus on SQL Query over HDFS, though the second phase will put onus on MapReduce instead of SQL. Moreover, PolyBase will only make its appearances is SQL Server PDW. So keeping an eye on this development will be a smart move indeed.

There's actually quite a bit at stake here, as Microsoft's trying to bring in a reliable big data solution. The growing importance of big data in businesses is a clear impetus to bring a solution that can not only handle such a job today, but can be poised to take on the issues of tomorrow as well. That's the sign of the best kind of software; the ability to be ready for tomorrow's problems, instead of just calling it a day when the issues faced today are handled. Seeing Microsoft involved in Hadoop not only makes for a great solution today, it also builds on big data and improves on it for tomorrow.