Nikhil Kumar (SyncSort) on Converting SQL to MapReduce

According to Gartner – 70% of all data warehouses are performance and capacity constrained, and often ELT workloads can drive up to 80% of database capacity.

What happened? The original vision was the Enterprise Data Warehouse (EDW) is the single, consistent version of the truth - but the modern day reality is an EDW architecture that consists of thousands of lines of SQL code and ELT “pushdown” data transformations happening primarily in the EDW and staging layers. Organizations have created constrained Data Warehouses with spaghetti like architectures resulting in the inability to get value from one of their biggest investments. With constant upgrades to the EDW just to keep the lights on, companies are now under immense pressure to alleviate the resulting budget and performance problems.

The good news is that for the first time, there’s a completely new approach that is not only scalable but also economically feasible. When comparing the costs to manage data, Hadoop is orders of magnitude cheaper than the EDW, and that’s why it’s become a very disruptive force in the data landscape.

But where do you start? Nikhil Kumar will present SyncSort’s EDW offload framework and SILQ, a unique first of its kind utility to make that transition between legacy SQL and Hadoop as smooth as possible.

Speaker: Nikhil Kumar, Technical Product Manager, SyncSort