Analyzing the energy-efficiency of sparse matrix multiplication on heterogeneous systems: A comparative study of GPU, Xeon Phi and FPGA

ISPASS 2016 - International Symposium on Performance Analysis of Systems and Software

Volumn , Issue , 2016, Pages 46-56

  Giefers, Heiner ^a   Staar, Peter ^a   Bekas, Costas ^a   Hagleitner, Christoph ^a  

^a IBM RESEARCH ZURICH   (Switzerland)

Author keywords

[No Author keywords available]

Indexed keywords

ACCELERATION; BIG DATA; COMPUTER HARDWARE; DATA MINING; DATA TRANSFER; FIELD PROGRAMMABLE GATE ARRAYS (FPGA); HARDWARE; MATRIX ALGEBRA; RECONFIGURABLE HARDWARE; TELECOMMUNICATION NETWORKS;

COMPARATIVE STUDIES; COMPUTE-INTENSIVE TASKS; HARDWARE ACCELERATORS; HETEROGENEOUS SYSTEMS; HIGH PERFORMANCE COMPUTING; MODERN COMPUTER SYSTEMS; OPTIMAL SOLUTIONS; SYSTEM'S PERFORMANCE;

ENERGY EFFICIENCY;

EID: 84978634383     PISSN: None     EISSN: None     Source Type: Conference Proceeding    
DOI: 10.1109/ISPASS.2016.7482073     Document Type: Conference Paper

References (70)

1. "The Green500 List-June 2015," http://www.green500.org/lists/ green201506, accessed: 2015-09-04.
2. B. Subramaniam, W. Saunders, T. Scogland, and w.-c. Feng, "Trends in energy-efficient computing: A perspective from the Green500," in Green Computing Coriference (IGCC), 2013.
3. J. Fowers, G. Brown, P. Cooke, and G. Stitt, " A Performance and Energy Comparison of FPGAs, GPUs, and Multicores for Slidingwindow Applications," in Int. Symp. on Field-programmable Gate Arrays (FPGA), 2012.
4. S. Kestur, J. Davis, and O. Williams, "BLAS Comparison on FPGA, CPU and GPU," in Annual Symposium on VLSI (ISVLSI), 2010.
5. C. W. F1etcher, I. A. Lebedev, N. B. Asadi, D. R. Burke, and J. Wawrzy nek, "Bridging the GPGPU-FPGA Efficiency Gap," in Field Programmable Gate Arrays (FPGA), 2011.
6. B. Cope, P. Cheung, W. Luk, and L. Howes, "Performance Comparison of Graphics Processors to Reconfigurable Logic: A Case Study," IEEE Trans. on Computers, vol. 59, no. 4, 2010.
7. A. Putnam, A. Caulfield, E. Chung, D. Chiou, K. Constantinides, J. Demme, H. Esmaeilzadeh, J. Fowers, G. Gopal, J. Gray, M. Haselman, S. Hauck, S. Heil, A. Hormati, J.-Y. Kim, S. Lanka, J. Larus, E. Peterson, S. Pope, A. Smith, J. Thong, P. Xiao, and D. Burger, "A Reconfigurable Fabric for Accelerating Large-Scale Datacenter Services," in Int. Symp. on Computer Architecture (ISCA), 2014.
8. J. Ouyang, S. Lin, W. Qi, Y. Wang, B. Yu, and S. Jiang, "SDA: Software-Defined Accelerator for Large-Scale DNN Systems," in HotChips26, 2014.
9. P. K. Gupta, "Intel Xeon+FPGA Platform for the Data Center," in Field Programmable Logic and Applications (FPL), Workshop on Recorifigurable Computing for the Masses, 2014.
10. J. Stuecheli, B. Blaner, C. Johns, and M. Siegel, "CAPI: A Coherent Accelerator Processor Interface," IBM Journal of Research and Development, vol. 59, no. 1, 2015.
11. K. Asanovic, R. Bodik, B. C. Catanzaro, J. J. Gebis, P. Husbands, K. Keutzer, D. A. Patterson, W. L. Plishker, I. Shalf, S. W. Williams, and K. A. Yelick, "The Landscape of Parallel Computing Research: A View from Berkeley," EECS Department, University of California, Berkeley, Tech. Rep., 2006.
12. A. Danalis, G. Marin, C. McCurdy, J. Meredith, P. Roth, K. Spafford, V. Tipparaju, and J. Vetter, "The Scalable HeterOgeneous Computing (SHOC) Benchmark Suite," in 3rd Workshop on General-Purpose Computation on Graphics Processors (GPGPU), 2010.
13. S. Che, M. Boyer, J. Meng, D. Tarjan, J. Sheaffer, S.-H. Lee, and K. Skadron, "Rodinia: A benchmark suite for heterogeneous computing," in Int. Symp. on Workload Characterization (llSWC), 2009.
14. W-c. Feng, H. Lin, T. Scogland, and J. Zhang, "OpenCL and the 13 Dwarfs: A Work in Progress," in Int. Conf. on Peiformance Engineering (ICPE), 2012.
15. A. Guerre, J.-T. Acquaviva, and Y. Lhuillier, "A unified methodology for a fast benchmarking of parallel architecture," in Design, Automation and Test in Europe (DATE), 2014.
16. H. Giefers, R. Polig, and C. Hagleitner, "Analyzing the energyefficiency of dense linear algebra kerneis by power-profiling a hybrid CPUIFPGA system," in Application-specijic Systems, Architectures and Processors (ASAP), 2014.
17. B. Li, H.-C. Chang, S. L. Song, c.-Y. Su, T. Meyer, J. Mooring, and K. Cameron, "The Power-Performance Tradeoffs of the Intel Xeon Phi on HPC Applications," in Int. Workshop on Large Scale Parallel Processing (LSPP), 2014.
18. J. Sim, A. Dasgupta, H. Kim, and R. Vuduc, "A performance analysis framework for identifying potential benefits in GPGPU applications," in Principles and Practice of Parallel Programming (PPoPP), 2012.
19. M. S.-A. H. Wong, M.-M. Papadopoulou and A. Moshovos, "Demystifying GPU microarchitecture through microbenchmarking," in Peiformance Analysis of Systems Software (ISPASS), 2010.
20. M. Naumov, L. S. Chien, P. Vandermersch, and U. Kapasi, "CUSPARSE Library: A Set of Basic Linear AlgebraSubroutines for Sparse Matrices," in GPU Technology Coriference, 2010.
21. T. A. Davis and Y. Hu, "The University of Florida Sparse Matrix Collection," ACM Trans. Math. Softw., vol. 38, no. 1, 2011.
22. X. Liu, M. Smelyanskiy, E. Chow, and P. Dubey, "Efficient Sparse Matrix-vector Multiplication on x86-based Many-core Processors," in Int. Con! on Supercomputing (ISC), 2013.
23. E. Saule, K. Kaya, and Ü. VataIyürek, "Performance Evaluation of Sparse Matrix Multiplication Kerneis on Intel Xeon Phi," in Parallel Processing and Applied Mathematics (PPAM), 2013.
24. Y. Zhang, Y. Shalabi, R. Jain, K. Nagar, and J. Bakos, "FPGA vs. GPU for sparse matrix vector multiply," in Field-Programmable Technology (FPT), 2009.
25. T. Brewer, "Instruction Set Innovations for the Convey HC-l Computer," Micro, IEEE, vol. 30, no. 2, 2010.
26. K. Nagar and J. Bakos, " A Sparse Matrix Personality for the Convey HC-l," in Field-Programmable Custom Computing Machines (FCCM), 2011.
27. S. K. Amd, John D. Davis and E. S. Chung, "Towards a Universal FPGA Matrix-Vector Multiplication Architecture," in Field-Programmable Custom Computing Machines (FCCM), 2012.
28. NVML API REFERENCE MANUAL, Version 4.304.55 ed., NVIDIA Corp., 2012.
29. M. Burtscher, I. Zecena, and Z. Zong, "Measuring GPU Power with the K20 Built-in Sensor," in Proceedings ofWorkshop on General Purpose Processing Using GPUs, ser. GPGPU-7, 2014.
30. H. David, E. Gorbatov, U. R. Hanebutte, R. Khanna, and C. Le, "RAPL: Memory Power Estimation and Capping," in Int. Symp. on Low Power Electronics and Design (ISLPED), 2010.
31. Watts up? and Watts up? PRO Operators Manual, https://www. wattsupmeters.com, Electronic Educational Devices, accessed: 2015-09-08.
32. Power and Peiformance Benchmark Methodology, V2.1 ed., Standard Performance Evaluation Corporation (SPEC), SPEC Power and Performance Committee, 2012.
33. V G. Oklobdzija, The Computer Engineering Handbook, 2001.
34. D. Brooks, P. Bose, S. Schuster, H. Jacobson, P. Kudva, A. Buyuktosunoglu, J.-D. Wellman, V Zyuban, M. Gupta, and P. Cook, "PowerAware Microarchitecture: Design and Modeling Challenges for NextGeneration Microprocessors," IEEE Micro, vol. 20, no. 6, 2000.
35. R. Gonzalez and M. Horowitz, "Energy dissipation in general purpose microprocessors," Solid-State Circuits, vol. 31, no. 9, 1996.
36. M. Horowitz, T. Indermaur, and R. Gonzalez, "Low-Power Digital Design," in IEEE Symp. Low Power Electronics, 1994.
37. Intel Xeon Phi Coprocessor. Datasheet, Intel, 2014.
38. Tesla K20 GPU Accelerator, BD-06455-00Lv07, NVIDIA, 2013.
39. PCle-385N Product Brief, Nallatech, 2014.
40. Altera SDK for OpenCL. Programming Guide, 13th ed., Altera Corp., Dec. 2013.
41. Intel Xeon Processor E5-2600 v2. Product Brief, Intel, 2014.
42. Stratix V Device Handbook, Altera, Jul. 2014.
43. "NVIDIA CUDA Sparse Matrix library," https://developer.nvidia.coml cusparse, accessed: 2015-09-04. 56
44. B. Dally, P. Hanrahan, and R. Fedkiw, " A Streaming Supercomputer. Stanford Computer Systems Laboratory White Paper," 2001.
45. G. Goumas, K. Kourtis, N. Anastopoulos, V Karakasis, and N. Koziris, "Performance evaluation of the sparse matrix-vector multiplication on modern architectures," The Journal of Supercomputing, vol. 50, no. 1, 2009.
46. R. W Vuduc, "Automatie Performance Tuning of Sparse Matrix Kerneis," Ph.D. dissertation, University of California, Berkeley, CA, USA, 2004.
47. Y. Saad, "SPARSKIT: a basic tool kit for sparse matrix computations," Tech. Rep., 1994, version 2.
48. S. Yan, C. Li, Y. Zhang, and H. Zhou, "yaSpMV: Yet Another SpMV Framework on GPUs," in Principles and Practice of Parallel Programming (PPoPP), 2014.
49. B.-Y. Su and K. Keutzer, "clSpMV: A Cross-Platform OpenCL SpMV Framework on GPUs," in Supercomputing (ISC), 2012.
50. M. Kreutzer, G. Hager, G. Wellein, H. Fehske, and A. R. Bishop, "A unified sparse matrix data format for efficient general sparse matrixvector multiply on modern processors with wide SIMD units," SIAM Journal on Scientific Computing, vol. 36, no. 5, 2014.
51. J. W Choi, A. Singh, and R. W Vuduc, "Model-driven Autotuning of Sparse Matrix-vector Multiply on GPUs," in Principles and Practice of Parallel Programming (PPoPP), 2010.
52. R. Pozo and K. Remington, "SparseLib++ vl.5-Sparse Matrix Class Library," Tech. Rep., 1996, reference Guide.
53. T. A. Davis, "SuiteSparse: A Suite of Sparse Matrix Software," http: //www.suitesparse.com. accessed: 2015-09-16.
54. J. L. Greathouse and M. Daga, "Efficient sparse matrix-vector multiplication on gpus using the csr storage format," in High Performance Computing, Networking, Storage and Analysis (SC), 2014.
55. P. W J. Staar, P. K. Barkoutsos, R. Istrate, A. C. I. Malossi, I. Tavernelli, N. Moll, H. Giefers, C. Hagleitner, C. Bekas, and A. Curioni, "Stochastic Matrix-Function Estimators: Scalable BigData Kerneis with High Performance," in Parallel and Distributed Processing Symposium (IPDPS), 2016.
56. M. Seltzer, D. Krinsky, K. Smith, and X. Zhang, "The Case for Application-Specific Benchmarking," in Hot Topics in Operating Systems (HOTOS), 1999.
57. C. Gregg and K. Hazelwood, "Where is the Data? Why you Cannot Debate CPU vS. GPU Performance Without the Answer," in Peiformance Analysis of Systems and Software (ISPASS), 201l.
58. Altera Corp., "IP Compiler for PCI Express," 2013.
59. -, "External Memory Interface Handbook," 2015.
60. -, "Floating-Point Megafunctions," User Guide, 2013.
61. Xilinx, Inc., "Floating-Point Operator v7.l," LogiCORE IP Product Guide, 2015.
62. R. Nane, V-Mo Sima, C. Pilato, J. Choi, B. Fort, A. Canis, Y. Chen, H. Hsiao, S. Brown, F. Ferrandi, J. Anderson, and K. Bertels, "A survey and evaluation of fpga high-level synthesis tools," ComputerAided Design of Integrated Circuits and Systems, IEEE Transactions on, 2016, Preprint.
63. Xilinx, Inc., "SDAccel Development Environment," User Guide, 2015.
64. K. Hili, S. Craciun, A. George, and H. Lam, "Comparative analysis of OpenCL vs. HDL with image-processing kerneis on Stratix-V FPGA," in Application-specijic Systems, Architectures and Processors (ASAP), 2015.
65. L. A. Barroso and U. Hölzle, "The Case for Energy-Proportional Computing," Computer, vol. 40, no. 12, 2007.
66. S. Srikantaiah, A. Kansal, and F. Zhao, "Energy Aware Consolidation for Cloud Computing," in HotPower, 2008.
67. A.-C. Orgerie, M. D. d. Assuncao, and L. Lefevre, " A Survey on Techniques for Improving the Energy Efficiency of Large-scale Distributed Systems," ACM Comput. Surv., vol. 46, no. 4, 2014.
68. HP Moonshot System Family Guide, Hewlett-Packard, 2014.
69. R. Luijten, D. Pham, R. Clauberg, M. Cossale, H. Nguyen, and M. Pandya, "Energy-Efficient Microserver Based on a 12-Core l.8GHz 188K-CoreMark 28nm Bulk CMOS 64b SoC for Big-Data Applications with 159GB/slL Memory Bandwidth System Density," in SolidState Circuits Conference (ISSCC), 2015.
70. G. Wang, H. Lam, A. George, and G. Edwards, "Performance and Productivity Evaluation of Hybrid-Threading HLS versus HDLs," in High Peiformance Extreme Computing Coriference (HPEC), 2015.