Hardware Acceleration for Query Processing: Leveraging FPGAs, CPUs, and Memory

Database management systems have become an indispensable tool for industry, government, and academia, and form a significant component of modern datacenters. They can be used in a multitude of scenarios, including online analytical processing, data mining, e-commerce, and scientific analysis. Given the exponential growth in new data produced each year, there is a pressure on software and hardware developers to create datacenters that can cope with increasing requirements. The authors look at the organization of a modern relational database management system and propose optimizations and redesign for the storage access, memory, and CPU. The research leading to these results has received funding from the EU’s Seventh Framework Programme (FP7/2007- 2013) under the AXLE project (grant 318633). Gorker Alp Malazgirt undertook this work with a Severo Ochoa grant at the Barcelona Supercomputing Center.