In-memory grid files on graphics processors
HKUST, China
In DaMoN ’07: Proceedings of the 3rd international workshop on Data management on new hardware (2007), pp. 1-7
@conference{yang2007memory,
title={In-memory grid files on graphics processors},
author={Yang, K. and He, B. and Fang, R. and Lu, M. and Govindaraju, N. and Luo, Q. and Sander, P. and Shi, J.},
booktitle={Proceedings of the 3rd international workshop on Data management on new hardware},
pages={1–7},
year={2007},
organization={ACM}
}
Recently, graphics processing units, or GPUs, have become a viable alternative as commodity, parallel hardware for general-purpose computing, due to their massive data-parallelism, high memory bandwidth, and improved general-purpose programming interface. In this paper, we explore the use of GPU on the grid file, a traditional multidimensional access method. Considering the hardware characteristics of GPUs, we design a massively multi-threaded GPU-based grid file for static, memory-resident multidimensional point data. Moreover, we propose a hierarchical grid file variant to handle data skews efficiently. Our implementations on the NVIDIA G80 GTX graphics card are able to achieve two to eight times’ higher performance than their CPU counterparts on a single PC.
December 4, 2010 by hgpu
