Toward GPUs being mainstream in analytic processing: An initial argument using simple scan-aggregate queries

Paper on ACM DL Local Download Presentation (PDF)

Abstract

There have been a number of research proposals to use discrete graphics processing units (GPUs) to accelerate database operations. Although many of these works show up to an order of magnitude performance improvement, discrete GPUs are not commonly used in modern database systems. However, there is now a proliferation of integrated GPUs which are on the same silicon die as the conventional CPU. With the advent of new programming models like heterogeneous system architecture, these integrated GPUs are considered first-class compute units, with transparent access to CPU virtual addresses and very low overhead for computation offloading. We show that integrated GPUs significantly reduce the overheads of using GPUs in a database environment. Specifically, an integrated GPU is 3x faster than a discrete GPU even though the discrete GPU has 4x the computational capability. Therefore, we develop high performance scan and aggregate algorithms for the integrated GPU. We show that the integrated GPU can outperform a four-core CPU with SIMD extensions by an average of 30% (up to 3:2x) and provides an average of 45% reduction in energy on 16 TPC-H queries.

Jason Power, Yinan Li, Mark D. Hill, Jignesh M. Patel, and David A. Wood. 2015. Toward GPUs being mainstream in analytic processing: An initial argument using simple scan-aggregate queries. In Proceedings of the 11th International Workshop on Data Management on New Hardware (DaMoN’15), Ippokratis Pandis and Martin Kersten (Eds.). ACM, New York, NY, USA, , Article 11 , 8 pages. DOI=http://dx.doi.org/10.1145/2771937.2771941

@inproceedings{gpudb:Power:2015,
    author = {Power, Jason and Li, Yinan and Hill, Mark D. and Patel, Jignesh M. and Wood, David A.},
    title = {Toward GPUs Being Mainstream in Analytic Processing: An Initial Argument Using Simple Scan-aggregate Queries},
    booktitle = {Proceedings of the 11th International Workshop on Data Management on New Hardware},
    series = {DaMoN'15},
    year = {2015},
    isbn = {978-1-4503-3638-3},
    location = {Melbourne, VIC, Australia},
    pages = {11:1--11:8},
    articleno = {11},
    numpages = {8},
    url = {http://doi.acm.org/10.1145/2771937.2771941},
    doi = {10.1145/2771937.2771941},
    acmid = {2771941},
    publisher = {ACM},
    address = {New York, NY, USA},
} 

Updated: