Global Shared Memory Design for Multi-GPU Graphics Cards on Personal Supercomputer

Abstract:

When programming CUDA or OpenCL on multi-GPU systems, the programmers usually expect the GPUs on the same system can communicate fast with each other. For instance, they hope a device memory copy from GPU1s memory to GPU2s memory can be done inside the graphics card, and needn’t to employ the PCIE, which is in relative low speed. In this paper, we propose an idea to add a multi-channel memory to the multi-GPU board, and this memory is only for transferring data between different GPUs. This multi-channel memory should have multiple interfaces, including one common interface shared by different GPUs, which is connected with a FPGA arbitration circuit and several other interfaces connected with dedicated GPUs frame buffer independently. To distinguish the shared memory of a stream multiprocessor, we call this memory Global Shared Memory. We analyze the performance improvement expectation with this global shared memory, with the case of accelerating computer tomography algebraic reconstruction on multi-GPU.

• Periodical: Applied Mechanics and Materials (Volumes 263-266)
• Pages: 1236-1241
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.263-266.1236
• Online since: December 2012
• Authors: Sen Guo*, San Feng Chen, Yong Sheng Liang
• Keywords: CUDA, Global Shared Memory, GPGPU, Multi-GPU

References

[1] England. Proceedings of the 16th annual conference on Computer graphics and interactive techniques (1989).

[2] Potmesil.M, Hofert E.M. Proceedings of the 16th annual conference on Computer graphics and interactive techniques (1989).

[3] Feldkamp, L. A., Davis, L. C, Kress, J. W. Journal of the Optical Society of America A: Optics, Image Science, and Vision. Vol 1(1984).

[4] Information on: http://gpgpu.org/oldsite/data/history.shtml.

[5] Information on: http://www.khronos.org/opencl/.

[6] Benjamin Block, Peter Virnau, Tobias Preis. Computer Physics Communications. Vol 181(2010)

[7] Enos, J.J. , Guochun Shi, Showerman, M.T. Cluster Computing and Workshops(2009).

[8] R.Gordon, R.Bender and G.T. Herman. Theoretical Biology, vol.29(2010).

[9] A.H. Andersen and A.C. Kak. Ultrason. Img., Vol. 6(1984).

[10] A.C. Kak and M. IEEE Press, 1988.

[11] Y.Lu, W. Wang, S. Chen, Y. Xie, J. Qin, W. Pang, P.A. Heng. Proceedings of Sixth International Conference on Computer Graphics, Imaging and Visualization,(2009).

DOI: 10.1109/cgiv.2009.18

[12] Benjamin Kecka, Hannes Hofmanna. Medical Imaging 2009: Physics of Medical Imaging(2009).

[13] B. Keck, H. Hofmann, et al. Proceedings of SPIE(2009).