The Vertica Analytic Database (Vertica) is a commercialization of the design outlined in the C-Store research project. Vertica has been purpose built from the ground up to handle analytic type queries on massive amounts of data. It is an interesting technology that dabbles in advanced storage, distributed systems and high performance computing. Though interesting in its own right, Vertica is remarkable when viewed through a historical lens. Several of its key architectural concepts are close derivations of innovations described in the C-Store paper. A few ideas from the C-Store paper were never implemented or implemented in a way that is almost unrecognizable from the original blueprint, and in some ways these are the most interesting.
In this talk we give a brief overview of C-Store and focus on its evolution to a commercial product with particular attention paid to the technological differences between blueprint and product.
Stephen Walkauskas is a Software Engineer at HP Vertica. He works on the Vertica Analytic Database in areas such as recovery, storage optimization, efficient re-distribution of data across the cluster, backup and restore, and Hadoop integration. Prior to joining Vertica he was one of the very first developers at Endeca which grew to over 500 employees over an eight year span and was eventually purchased by Oracle for over $1B. Stephen holds a B.S. in Physics from Boston College. However, he enjoyed his computational work and transitioned into software engineering. Stephen is originally from Pittsburgh, PA, where he's returned to open a Vertica Engineering office after nearly two decades in Boston.
jennsbl [atsymbol] cs ~replace-with-a-dot~ cmu ~replace-with-a-dot~ edu