Skip to main content
SpringerLink
Log in
Book cover

Smart Infrastructure and Applications pp 409–426Cite as

Performance Characteristics for Sparse Matrix-Vector Multiplication on GPUs

  • Chapter
  • First Online:

Part of the book series: EAI/Springer Innovations in Communication and Computing ((EAISICC))

Abstract

The massive parallelism provided by the graphics processing units (GPUs) offers tremendous performance in many high-performance computing applications. One such application is Sparse Matrix-Vector (SpMV) multiplication, which is an essential building block for numerous scientific and engineering applications. Researchers who propose new storage techniques for sparse matrix-vector multiplication focus mainly on a single evaluation metrics or performance characteristics which is usually the throughput performance of sparse matrix-vector multiplication in FLOPS. However, such an evaluation does not provide a deeper insight nor allow to compare new SpMV techniques with their competitors directly. In this chapter, we explain the notable performance characteristics of the GPU architectures and SpMV computations. We discuss various strategies to improve the performance of SpMV on GPUs. We also discuss a few performance criteria that are usually overlooked by the researchers during the evaluation process. We also analyze various well-known schemes such as COO, CSR, ELL, DIA, HYB, and CSR5 using the discussed performance characteristics.

This is a preview of subscription content, log in via an institution.

Chapter
USD   29.95
Price excludes VAT (USA)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever
eBook
USD   119.00
Price excludes VAT (USA)
  • Available as EPUB and PDF
  • Read on any device
  • Instant download
  • Own it forever
Softcover Book
USD   159.99
Price excludes VAT (USA)
  • Compact, lightweight edition
  • Dispatched in 3 to 5 business days
  • Free shipping worldwide - see info
Hardcover Book
USD   219.99
Price excludes VAT (USA)
  • Durable hardcover edition
  • Dispatched in 3 to 5 business days
  • Free shipping worldwide - see info

Tax calculation will be finalised at checkout

Purchases are for personal use only

Learn about institutional subscriptions

References

  1. Yang, W., Li, K., Li, K.: A hybrid computing method of SpMV on CPU–GPU heterogeneous computing systems. J. Parallel Distrib. Comput. 104, 49–60 (2017)

    Article  Google Scholar 

  2. Mehmood, R., Lu, J.A.: Computational Markovian analysis of large systems. J. Manuf. Technol. Manag. 22, 804–817 (2011)

    Article  Google Scholar 

  3. Mehmood, R., Meriton, R., Graham, G., Hennelly, P., Kumar, M.: Exploring the influence of big data on city transport operations: a Markovian approach. Int. J. Oper. Prod. Manag. 37, 75–104 (2017)

    Article  Google Scholar 

  4. Mehmood, R., Alturki, R., Zeadally, S.: Multimedia applications over metropolitan area networks (MANs). J. Netw. Comput. Appl. 34, 1518–1529 (2011)

    Article  Google Scholar 

  5. Mehmood, R., Graham, G.: Big data logistics: a health-care transport capacity sharing model. Proc. Comput. Sci. 64, 1107–1114 (2015)

    Article  Google Scholar 

  6. Altowaijri, S., Mehmood, R., Williams, J.: A quantitative model of grid systems performance in healthcare organisations. ISMS 2010—UKSim/AMSS 1st International Conference on Intelligent Systems. Model. Simul. 431–436 (2010)

    Google Scholar 

  7. Huan, G., Qian, Z.: A new method of sparse matrix-vector multiplication on GPU. In: International Conference on Computer Science and Network Technology, pp. 954–958 (2012)

    Google Scholar 

  8. Hassani, R., Fazely, A., Choudhury, R.-U.-A., Luksch, P.: Analysis of sparse matrix-vector multiplication using iterative method in CUDA. In: 2013 IEEE Eighth International Conference on Networking, Architecture and Storage, pp. 262–266 (2013)

    Chapter  Google Scholar 

  9. Cheik Ahamed, A.-K., Magoulès, F.: Efficient implementation of Jacobi iterative method for large sparse linear systems on graphic processing units. J. Supercomput. 73, 3411–3432 (2017)

    Article  Google Scholar 

  10. Adhianto, L., Banerjee, S., Fagan, M., Krentel, M., Marin, G., Mellor-Crummey, J., Tallent, N.R.: HPCTOOLKIT: tools for performance analysis of optimized parallel programs. Concurr. Comput. Pract. Exp. 22, 685–701 (2010). http://hpctoolkit.org

    Google Scholar 

  11. Brahme, D., Mishra, B.R., Barve, A.: Parallel sparse matrix vector multiplication using greedy extraction of boxes. In: 2010 International Conference on High Performance Computing, pp. 1–10 (2010)

    Google Scholar 

  12. Ahamed, A.-K.C., Magoules, F.: Fast sparse matrix-vector multiplication on graphics processing unit for finite element analysis. In: 2012 IEEE 14th International Conference on High Performance Computing and Communication & 2012 IEEE 9th International Conference on Embedded Software and Systems, pp. 1307–1314 (2012)

    Chapter  Google Scholar 

  13. Guo, P., Wang, L., Chen, P.: A performance modeling and optimization analysis tool for sparse matrix-vector multiplication on GPUs. IEEE Trans. Parallel Distrib. Syst. 25, 1112–1123 (2014)

    Article  Google Scholar 

  14. Guo, P., Wang, L.: Auto-tuning CUDA parameters for sparse matrix-vector multiplication on GPUs. In: Proceedings—2010 International Conference on Computational and Information Sciences, ICCIS 2010, pp. 1154–1157 (2010)

    Chapter  Google Scholar 

  15. Merrill, D., Garland, M.: Merge-based parallel sparse matrix-vector multiplication. In: International Conference for High Performance Computing, Networking, Storage and Analysis, SC 16, pp. 678–689 (2016)

    Chapter  Google Scholar 

  16. Hou, K., Feng, W.C., Che, S.: Auto-tuning strategies for parallelizing sparse matrix-vector (SpMV) multiplication on multi- and many-core processors. In: Proceedings—2017 IEEE 31st International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2017, pp. 713–722 (2017)

    Google Scholar 

  17. Mehmood, R., Crowcroft, J.: Parallel Iterative Solution Method for Large Sparse Linear Equation Systems. UCAM-CL-TR-650. University of Cambridge, Computer Laboratory (2005)

    Google Scholar 

  18. Mehmood, R.: Disk-Based Techniques for Efficient Solution of Large Markov Chains. Computer Science, University of Birmingham (2004)

    Google Scholar 

  19. Barrett, R., Berry, M., Chan, T.F., Demmel, J., Donato, J., Dongarra, J., Eijkhout, V., Pozo, R., Romine, C., van der Vorst, H.: Templates for the Solution of Linear Systems: Building Blocks for Iterative Methods. Society for Industrial and Applied Mathematics, Philadelphia (1994)

    Book  Google Scholar 

  20. Eleliemy, A., Fayez, M., Mehmood, R., Katib, I., Aljohani, N.: Loadbalancing on parallel heterogeneous architectures: spin-image algorithm on CPU and MIC. In: 9th EUROSIM Congress on Modelling and Simulation. EUROSIM (2016)

    Google Scholar 

  21. Kwiatkowska, M., Mehmood, R.: Out-of-Core solution of large linear Systems of Equations Arising from stochastic modelling. In: Process Algebra and Probabilistic Methods: Performance Modeling and Verification, pp. 135–151. Springer, Berlin (2002)

    Chapter  Google Scholar 

  22. Kwiatkowska, M., Mehmood, R., Norman, G., Parker, D.: A symbolic out-of-core solution method for Markov models. Electr. Notes Theor. Comput. Sci. 68, 589–604 (2002)

    Article  Google Scholar 

  23. Mehmood, R.: A Survey of Out-of-Core Analysis Techniques in Stochastic Modelling. University of Birmingham, UK (2003)

    Google Scholar 

  24. Mehmood, R.: Serial disk-based analysis of large stochastic models. In: Baier, C., Haverkort, B.R., Hermanns, H., Katoen, J.-P., Siegle, M. (eds.) Validation of Stochastic Systems: A Guide to Current Research, pp. 230–255. Springer, Berlin (2004)

    Chapter  Google Scholar 

  25. Mehmood, R., Crowcroft, J., Elmirghani, J.M.H.: A parallel implicit method for the steady-state solution of CTMCs. In: 14th IEEE International Symposium on Modeling, Analysis, and Simulation, pp. 293–302 (2006)

    Chapter  Google Scholar 

  26. Mehmood, R., Parker, D., Kwiatkowska, M.: An Efficient BDD-Based Implementation of Gauss-Seidel for CTMC Analysis. University of Birmingham, UK (2003)

    Google Scholar 

  27. Mehmood, R., Parker, D., Kwiatkowska, M.: An Efficient Symbolic Out-of-Core Solution Method for Markov Models., University of Birmingham, UK (2003)

    Google Scholar 

  28. Magoulès, F., Ahamed, A.-K.C.: Alinea: an advanced linear algebra library for massively parallel computations on graphics processing units. Int. J. High Perform. Comput. Appl. 29, 284–310 (2015)

    Article  Google Scholar 

  29. Muhammed, T., Mehmood, R., Albeshri, A., Katib, I.: UbeHealth: a personalized ubiquitous cloud and edge-enabled networked healthcare system for smart cities. IEEE Access. 6, 32258–32285 (2018)

    Article  Google Scholar 

  30. Owens, J.D., Houston, M., Luebke, D., Green, S., Stone, J.E., Phillips, J.C.: GPU computing. Proc. IEEE. 879–899 (2008)

    Google Scholar 

  31. Fevgas, A., Daloukas, K., Tsompanopoulou, P., Bozanis, P.: Efficient solution of large sparse linear systems in modern hardware. In: 2015 6th International Conference on Information, Intelligence, Systems and Applications (IISA), pp. 1–6 (2015)

    Google Scholar 

  32. Kirk, D.B., Hwu, W.M.W.: Programming Massively Parallel Processors: A Hands-on Approach (2013)

    Google Scholar 

  33. Cheng, J., Grossman, M., McKercher, T.: Professional CUDA C Programming. Wiley, New York (2014)

    Google Scholar 

  34. NVIDIA: Pascal GPU Architecture | NVIDIA. https://www.nvidia.com/en-us/data-center/pascal-gpu-architecture

  35. NVIDIA: NVIDIA Tesla P100 Whitepaper (2016)

    Google Scholar 

  36. NVIDIA: NVIDIA Tesla V100. https://www.nvidia.com/en-us/data-center/tesla-v100/?ncid=van-tesla-v100

  37. NVIDIA: History of NVIDIA – From Graphics Cards to Mobile Processors. http://www.nvidia.co.uk/object/corporate-timeline-uk.html

  38. Nvidia: Nvidia CUDA C Programming Guide Version 4.2 (2012)

    Google Scholar 

  39. Bell, N., Garland, M.: Efficient Sparse Matrix-Vector Multiplication on CUDA (2008)

    Google Scholar 

  40. Nvidia: Profiler User’s Guide

    Google Scholar 

  41. Saad, Y.: SPARSKIT: A Basic Tool Kit for Sparse Matrix Computations Version 2 (1994)

    Google Scholar 

  42. Bell, N., Garland, M.: Implementing sparse matrix-vector multiplication on throughput-oriented processors. In: Proceedings of the Conference on High Performance Computing Networking, Storage and Analysis—SC ’09, p. 1 (2009)

    Google Scholar 

  43. Liu, W., Vinter, B.: CSR5: An Efficient Storage Format for Cross-Platform Sparse Matrix-Vector Multiplication. Arxiv, Ithaca, NY (2015)

    Book  Google Scholar 

  44. Guo, P., Lee, C.W.: A performance prediction and analysis integrated framework for SpMV on GPUs. In: Procedia Computer Science, pp. 178–189. The Author(s), (2016)

    Article  Google Scholar 

  45. Fujita, M., McGeer, P.C., Yang, J.C.-Y.: Multi-terminal binary decision diagrams: an efficient data structure for matrix representation. Formal Meth. Syst. Design. 10, 149–169 (1997)

    Article  Google Scholar 

  46. Maggioni, M., Berger-Wolf, T.: Optimization techniques for sparse matrix-vector multiplication on GPUs. J. Parallel Distrib. Comput. 93-94, 66–86 (2016)

    Article  Google Scholar 

  47. Filippone, S., Cardellini, V., Barbieri, D., Fanfarillo, A.: Sparse matrix-vector multiplication on GPGPUs. ACM Trans. Math. Softw. 43, 1–49 (2017)

    Article  MathSciNet  Google Scholar 

  48. Williams, S., Oliker, L., Vuduc, R., Shalf, J., Yelick, K., Demmel, J.: Optimization of sparse matrix-vector multiplication on emerging multicore platforms—long version. Parallel Comput. 35, 178–194 (2009)

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Department of Computer Science, King Abdulaziz University, Jeddah, Saudi Arabia

    Sarah AlAhmadi, Thaha Muhammed & Aiiad Albeshri

  2. High Performance Computing Center, King Abdulaziz University, Jeddah, Saudi Arabia

    Rashid Mehmood

Authors
  1. Sarah AlAhmadi
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Thaha Muhammed
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Rashid Mehmood
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Aiiad Albeshri
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Sarah AlAhmadi .

Editor information

Editors and Affiliations

  1. High Performance Computing Center, King Abdulaziz University, Jeddah, Saudi Arabia

    Rashid Mehmood

  2. Nvidia AI Technology Center, Singapore, Singapore

    Simon See

  3. Faculty of Computing and Information Technology (FCIT), King Abdulaziz University, Jeddah, Saudi Arabia

    Iyad Katib

  4. European Alliance for Innovation, Gent, Belgium

    Imrich Chlamtac

Rights and permissions

Reprints and permissions

Copyright information

© 2020 Springer Nature Switzerland AG

About this chapter

Check for updates. Verify currency and authenticity via CrossMark

Cite this chapter

AlAhmadi, S., Muhammed, T., Mehmood, R., Albeshri, A. (2020). Performance Characteristics for Sparse Matrix-Vector Multiplication on GPUs. In: Mehmood, R., See, S., Katib, I., Chlamtac, I. (eds) Smart Infrastructure and Applications. EAI/Springer Innovations in Communication and Computing. Springer, Cham. https://doi.org/10.1007/978-3-030-13705-2_17

Download citation

  • DOI: https://doi.org/10.1007/978-3-030-13705-2_17

  • Published:

  • Publisher Name: Springer, Cham

  • Print ISBN: 978-3-030-13704-5

  • Online ISBN: 978-3-030-13705-2

  • eBook Packages: EngineeringEngineering (R0)

Publish with us

Policies and ethics

Search

Navigation