Using Graphics Processing Unit to Accelerate Database Query Execution
Institute of Telecommunications, The Faculty of Electronics and Information Technology, Warsaw University of Technology, ul. Nowowiejska 15/19, 00-665 Warszawa
Information Systems Architecture and Technology, 2013
@article{ksikazek2014using,
title={USING GRAPHICS PROCESSING UNIT TO ACCELERATE DATABASE QUERY EXECUTION},
author={KSI{k{A}}{.Z}EK, Kamil and SAPIECHA, Piotr},
journal={Information Systems Architecture and Technology},
pages={135},
year={2014}
}
One of the major problems in database management systems is handling large amounts of data while providing short response time. Problem is not only proper manner of storing records but also efficient way of processing them. In the meantime GPUs developed computational power many times greater than that offered by comparable CPUs. In our research we investigated benefits that using GPU in database queries execution can give. Using offered in PostgreSQL database User-Defined Aggregate extension mechanism, we implemented our versions of 3 standard aggregates: sum, average and population standard deviation that were executed on GPU by using OpenCL. We found that, while in simple aggregates (sum and average) there was no gain, in more complex aggregates (population standard deviation) we were able to achieve more than 2 times shorter execution time than in standard build in database aggregate. We conclude that processing database queries on GPU can be very effective, provided that proper method of dispatching operations between CPU and GPU is used.
October 6, 2014 by hgpu
