论文信息 - Accelerating High-Throughput Computing through OpenCL

Accelerating High-Throughput Computing through OpenCL

This paper presents the implementation of a HTCondor pool with GPU execution capabilities through OpenCL. Implementation is discussed from both the system setup and the software design standpoint. The GPU landscape is investigated and the efficiency of the system is evaluated via Fast- Fourier Transform computations. Experimental results show that HTCondor GPU performance matches a dedicated GPU cluster.

Violeta Holmes | Joshua Higgins | Andrei Dafinoiu

[1] Alan R. Jones,et al. Fast Fourier Transform , 1970, SIGP.

[2] Christopher Robert Cullinan,et al. Computing Performance Benchmarks among CPU, GPU, and FPGA , 2012 .

[3] David Gubb. Implementation of A Condor Pool At The UniversityOf Huddersfield That Conforms To A Green ITPolicy , 2013 .

[4] Miron Livny,et al. Mechanisms for High Throughput Computing , 1997 .

[5] Imre Szeberényi,et al. Harnessing Wasted Computing Power for Scientific Computing , 2013, LSSC.

[6] John E. Stone,et al. OpenCL: A Parallel Programming Standard for Heterogeneous Computing Systems , 2010, Computing in Science & Engineering.

[7] William J. Dally,et al. GPUs and the Future of Parallel Computing , 2011, IEEE Micro.