Communication efficient work distributions in stencil operation based applications

Concurrency and Computation: Practice and Experience

Volumn 27, Issue 13, 2015, Pages 3262-3280

  Schneible, Joseph ^a   Riha, Lubomir ^a,c   Malik, Maria ^a   El Ghazawi, Tarek ^a   Alexandru, Andrei ^b  

^a Department of Engineering Management and Systems Engineering   (United States)

^b GEORGE WASHINGTON UNIVERSITY   (United States)

^c VSB TECHNICAL UNIVERSITY OF OSTRAVA   (Czech Republic)

Author keywords

GPU acceleration; heterogeneous computing; nearest neighbor; performance model; stencil operation

Indexed keywords

COMPUTER GRAPHICS; GRAPHICS PROCESSING UNIT; INTERCONNECTION NETWORKS (CIRCUIT SWITCHING); QUANTUM THEORY;

GPU ACCELERATIONS; HETEROGENEOUS COMPUTING; NEAREST NEIGHBORS; PERFORMANCE MODEL; STENCIL OPERATION;

PROGRAM PROCESSORS;

EID: 84939487242     PISSN: 15320626     EISSN: 15320634     Source Type: Journal    
DOI: 10.1002/cpe.3210     Document Type: Article

References (32)

1. Egri GI, et al,. Lattice QCD as a video game. Computer Physics Communications 2007; 177: 631-639.
2. Alexandru A, Lujan M, Pelissier C, Gamari B, Lee FX,. Efficient implementation of the overlap operator on multi-GPUs, [arXiv:1106.4964v1], 2011.
3. Clark MA, Babich R, Barros K, Brower RC, Rebbi C, Solving Lattice QCD systems of equations using mized precision solbers on GPUs, [arXiv: hep-lat/0911.3191v2], 2009.
4. Karypis G, Kumar V,. METIS-unstructured graph partitioning and sparse matrix ordering system, Version 2.0, 1995.
5. Devine K, Boman E, Heaphy R, Hendrickson B, Vaughan C,. Zoltan data management services for parallel dynamic applications, DOI: 10.1109/5992.988653.
6. Goehring T, Saad Y,. Heuristic algorithms for automatic graph partitioning, 1995.
7. Karypis G, Kumar V,. A coarse-grain parallel formulation of multilevel k-way graph partitioning algorithm, 1997.
8. InfiniBand architecture specification volume 1, Release 1.0, October 24, 2000, available from the InfiniBand Trade Association, http://www.infinibandta.org [January 2013].
9. Liu J, Mamidala A, Vishnu V, Panda DK,. Evaluating InfiniBand performance with PCI Express. IEEE Micro 2005; 25 (1): 20-29.
10. HPC Advisory Council. Interconnect analysis: 10GigE and InfiniBand in high performance computing, white paper, [Online] http://www.hpcadvisorycouncil.com/pdf/IB-and-10GigE-in-HPC.pdf [January 2013].
11. Koop MJ, Huang W, Gopalakrishnan K, Panda DK,. Performance analysis and evaluation of PCIe 2.0 and quad-data rate InfiniBand, High Performance Interconnects, 2008. HOTI '08. 16th IEEE Symposium on, 2008; 85-92.
12. Fan Z, Qiu F, Kaufman A, Yoakum-Stover S,. GPU cluster for high performance computing, Supercomputing, 2004. Proceedings of the ACM/IEEE SC2004 Conference, 2004; 47.
13. Kindratenko VV, Enos JJ, Shi G, Showerman MT, Arnold GW, Stone JE, Phillips JC, Hwu W-m,. GPU clusters for high-performance computing. Cluster Computing and Workshops, 2009. CLUSTER '09. IEEE International Conference on, 2009; 1-8.
14. Schaa D, Kaeli D,. Exploring the multiple-GPU design space. Parallel & Distributed Processing, 2009. IPDPS 2009. IEEE International Symposium on, 2009; 1-12.
15. Shimer G, Ayoub A, Lui P, Liu T, Kagan M, Troot CR, Scantlen G, Crozier PS,. The developmetn of Mellanox/NVIDIA GPUDirect over InfiniBand- a new model for GPU to GPU communications. Computer Science-Research and Development 2011; 26 (3-4): 267-273.
16. Wang H, Potluri S, Luo M, Singh AK, Sur S, Panda DK,. MVAPICH2-GPU: optimized GPU to GPU communication for InfiniBand clusters. Computer Science-Research and Development 2011; 26 (3-4): 1-10.
17. Munster G, Walzl M,. Lattice gauge theory a short primer, [arXiv: hep-lat/0012005v1], 2000.
18. Montvay I, Münster G,. Quantum fields on a lattice, cambridge monographs on mathematical physics, 1994.
19. Hess et al. JCTC "GROMACS 4: Algorithms for highly efficient, load-balanced, and scalable molecular simulation. J. Chem. Theory Comput., 2008
20. NVIDIA GPUDirect (nd). Nvidia.com. Retrieved from http://developer.nvidia.com/gpudirect, 2010 [January 2013].
21. Clos C,. A study of non-blocking switching networks. Bell System Technical Journal 1953; 32: 406-424.
22. Charles E,. Leiserson: fat-trees: universal networks for hardware-efficient supercomputing. IEEE Transactions on Computers 1985; 34: 892-901.
23. Shainer G, Ayoub A, Lui P, Liu T, Kagan M, Trott CR, Scantlen G, Crozier PS,. The development of Mellanox NVIDIA GPUDirect over Infiniband-a new model for GPU to GPU communications. Computer Science-Research and Development 2011; 26: 267-273. doi: 10.1007/s00450-011-0157-1.
24. NVIDIA GPUDirect technology-accelerating GPU-based systems, May 2010, in www.mellanox.com, Retrieved from http://www.mellanox.com/pdf/whitepapers/TB-GPU-Direct.pdf [January 2013].
25. Rackform iServ R350.v2 in siliconmechanics.com. Retrieved from http://www.siliconmechanics.com/i27076/GPU-1UServer.php [January 2013].
26. SRM-ASU-ESC4000 in avadirect.com. Retrieved from http://www.avadirect.com/2u-rackserver-configurator.asp?PRID=21061 [January 2013].
27. Mellanox grid switch 9024 DDR 24-port InfiniBand switch in colfaxdirect.com. Retrieved from http://www.colfaxdirect.com/store/pc/viewPrd.asp?idproduct=921&idcategory=7 [January 2013].
28. Mellanox InfiniScale IV IS5025 DDR 36-port InfiniBand switch in colfaxdirect.com. Retrieved from http://www.colfaxdirect.com/store/pc/viewPrd.asp?idproduct=483&idcategory=7 [January 2013].
29. Mellanox InfiniScale IV IS5022 QDR 8-port InfiniBand switch in colfaxdirect.com. Retrieved from http://www.colfaxdirect.com/store/pc/viewPrd.asp?idproduct=601&idcategory=7 [January 2013].
30. Mellanox InfiniScale IV IS5023 QDR 18-port InfiniBand switch in colfaxdirect.com. Retrieved from http://www.colfaxdirect.com/store/pc/viewPrd.asp?idproduct=873&idcategory=7 [January 2013].
31. Mellanox InfiniScale IV IS5025 QDR 36-port InfiniBand switch in colfaxdirect.com. Retrieved from http://www.colfaxdirect.com/store/pc/viewPrd.asp?idproduct=154&idcategory=7 [January 2013].
32. Gupta R,. General physics motivations for numerical simulations of quantum field theory, [arXiv: hep-lat/9905027v1] 1999.