Skip to main content
SpringerLink
Log in

A Scheduling Theory Framework for GPU Tasks Efficient Execution

  • Conference paper
  • First Online:
High Performance Computing for Computational Science – VECPAR 2018 (VECPAR 2018)

Part of the book series: Lecture Notes in Computer Science ((LNTCS,volume 11333))

Included in the following conference series:

Abstract

Concurrent execution of tasks in GPUs can reduce the computation time of a workload by overlapping data transfer and execution commands. However it is difficult to implement an efficient runtime scheduler that minimizes the workload makespan as many execution orderings should be evaluated. In this paper, we employ scheduling theory to build a model that takes into account the device capabilities, workload characteristics, constraints and objective functions. In our model, GPU tasks scheduling is reformulated as a flow shop scheduling problem, which allow us to apply and compare well known methods already developed in the operations research field. In addition we develop a new heuristic, specifically focused on executing GPU commands, that achieves better scheduling results than previous techniques. Finally, a comprehensive evaluation, showing the suitability and robustness of this new approach, is conducted in three different NVIDIA architectures (Kepler, Maxwell and Pascal).

This work has been supported by the Ministry of Education of Spain (TIN2016-80920-R) and the Junta de Andalucía of Spain (TIC-1692).

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

Access this chapter

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 39.99
Price excludes VAT (USA)
  • Available as EPUB and PDF
  • Read on any device
  • Instant download
  • Own it forever
Softcover Book
USD 54.99
Price excludes VAT (USA)
  • Compact, lightweight 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

Institutional subscriptions

Notes

  1. 1.

    HyperQ is a feature of modern GPUs since Kepler architecture, where several hardware managed queues can schedule different kernel commands, but can also change the execution order of these commands.

References

  1. Allahverdi, A., Ng, C., Cheng, T., Kovalyov, M.Y.: A survey of scheduling problems with setup times or costs. Eur. J. Oper. Res. 187(3), 985–1032 (2008). https://doi.org/10.1016/j.ejor.2006.06.060

    Article  MathSciNet  MATH  Google Scholar 

  2. Basaran, C., Kang, K.D.: Supporting preemptive task executions and memory copies in GPGPUs. In: 2012 24th Euromicro Conference on Real-Time Systems, pp. 287–296, July 2012. https://doi.org/10.1109/ECRTS.2012.15

  3. Chen, G., Zhao, Y., Shen, X., Zhou, H.: EffiSha: a software framework for enabling effficient preemptive scheduling of GPU. In: Proceedings of the 22nd ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming. PPoPP 2017, pp. 3–16. ACM, New York (2017). https://doi.org/10.1145/3018743.3018748

  4. Framinan, J., Gupta, J., Leisten, R.: A review and classification of heuristics for permutation flow-shop scheduling with makespan objective. J. Oper. Res. Soc. 55, 1243–1255 (2004). https://doi.org/10.1057/palgrave.jors.2601784

    Article  MATH  Google Scholar 

  5. Garey, M.R., Johnson, D.S., Sethi, R.: The complexity of flowshop and jobshop scheduling. Math. Oper. Res. 1(2), 117–129 (1976)

    Article  MathSciNet  Google Scholar 

  6. Graham, R., Lawler, E., Lenstra, J., Kan, A.: Optimization and approximation in deterministic sequencing and scheduling: a survey. Ann. Discrete Math. 5(C), 287–326 (1979). https://doi.org/10.1016/S0167-5060(08)70356-X

    Article  MathSciNet  MATH  Google Scholar 

  7. Khronos Group: OpenCL 2.0 API Specification, October 2014. https://www.khronos.org/registry/cl/specs/opencl-2.0.pdf

  8. Lázaro-Muñoz, A.J., González-Linares, J., Gómez-Luna, J., Guil, N.: A tasks reordering model to reduce transfers overhead on GPUs. J. Parallel Distrib. Comput. 109, 258–271 (2017). https://doi.org/10.1016/j.jpdc.2017.06.015

    Article  Google Scholar 

  9. Lee, H., Al Faruque, M.A.: GPU-EvR: run-time event based real-time scheduling framework on GPGPU platform. In: Proceedings of the Conference on Design, Automation and Test in Europe. DATE 2014, pp. 220:1–220:6. European Design and Automation Association, Leuven (2014)

    Google Scholar 

  10. Nawaz, M., Enscore, E.E., Ham, I.: A heuristic algorithm for the m-machine, n-job flow-shop sequencing problem. Omega 11(1), 91–95 (1983). https://doi.org/10.1016/0305-0483(83)90088-9

    Article  Google Scholar 

  11. NVIDIA: CUDA Multi-process Service, March 2015. https://docs.nvidia.com/deploy/pdf/CUDAMultiProcessServiceOverview.pdf

  12. NVIDIA: CUDA Programming Guide, September 2015. http://docs.nvidia.com/cuda/cuda-c-programming-guide/index.html

  13. NVIDIA: CUDA Samples, September 2015. http://docs.nvidia.com/cuda/cuda-samples/index.html

  14. Palmer, D.S.: Sequencing jobs through a multi-stage process in the minimum total time–a quick method of obtaining a near optimum. J. Oper. Res. Soc. 16(1), 101–107 (1965). https://doi.org/10.1057/jors.1965.8

    Article  Google Scholar 

  15. Park, J.J.K., Park, Y., Mahlke, S.: Chimera: collaborative preemption for multitasking on a shared GPU. In: Proceedings of the Twentieth International Conference on Architectural Support for Programming Languages and Operating Systems. ASPLOS 2015, pp. 593–606. ACM, New York (2015). https://doi.org/10.1145/2694344.2694346

  16. Pinedo, M.: Scheduling: Theory, Algorithms, and Systems. With CD-ROM, 3rd edn. Springer, New York (2008)

    MATH  Google Scholar 

  17. Ruiz, R., Maroto, C.: A comprehensive review and evaluation of permutation flowshop heuristics. Eur. J. Oper. Res. 165(2), 479–494 (2005). https://doi.org/10.1016/j.ejor.2004.04.017

    Article  MATH  Google Scholar 

  18. Taillard, E.: Some efficient heuristic methods for the flow shop sequencing problem. Eur. J. Oper. Res. 47, 65–74 (1990)

    Article  MathSciNet  Google Scholar 

  19. Tukey, J.W.: Exploratory Data Analysis. Addison-Wesley Publishing Company, Boston (1977)

    MATH  Google Scholar 

  20. Wang, Z., Yang, J., Melhem, R., Childers, B., Zhang, Y., Guo, M.: Simultaneous multikernel GPU: multi-tasking throughput processors via fine-grained sharing. In: 2016 IEEE International Symposium on High Performance Computer Architecture (HPCA), pp. 358–369, March 2016. https://doi.org/10.1109/HPCA.2016.7446078

  21. Xu, Q., Jeon, H., Kim, K., Ro, W.W., Annavaram, M.: Warped-Slicer: efficient intra-SM slicing through dynamic resource partitioning for GPU multiprogramming. In: Proceedings of the 43rd International Symposium on Computer Architecture. ISCA 2016, pp. 230–242. IEEE Press, Piscataway (2016). https://doi.org/10.1109/ISCA.2016.29

  22. Zhong, J., He, B.: Kernelet: high-throughput GPU kernel executions with dynamic slicing and scheduling. IEEE Trans. Parallel Distrib. Syst. 25(6), 1522–1532 (2014). https://doi.org/10.1109/TPDS.2013.257

    Article  MathSciNet  Google Scholar 

  23. Zhou, H., Tong, G., Liu, C.: GPES: a preemptive execution system for GPGPU computing. In: 21st IEEE Real-Time and Embedded Technology and Applications Symposium, pp. 87–97, April 2015. https://doi.org/10.1109/RTAS.2015.7108420

Download references

Author information

Authors and Affiliations

  1. Department of Computer Architecture, University of Málaga, Málaga, Spain

    Antonio-Jose Lázaro-Muñoz, Bernabé López-Albelda, Jose María González-Linares & Nicolás Guil

Authors
  1. Antonio-Jose Lázaro-Muñoz
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Bernabé López-Albelda
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Jose María González-Linares
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Nicolás Guil
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Jose María González-Linares .

Editor information

Editors and Affiliations

  1. Federal University of São Carlos, São Carlos, São Paulo, Brazil

    Hermes Senger

  2. Lawrence Berkeley National Laboratory, Berkeley, CA, USA

    Osni Marques

  3. Universidade Estadual Paulista Júlio de Mesquita Filho, Presidente Prudente, São Paulo, Brazil

    Rogerio Garcia

  4. Universidade Estadual Paulista Júlio de Mesquita Filho, São Paulo, São Paulo, Brazil

    Tatiana Pinheiro de Brito

  5. Universidade Estadual Paulista Júlio de Mesquita Filho, São Paulo, São Paulo, Brazil

    Rogério Iope

  6. Universidade Estadual Paulista Júlio de Mesquita Filho, São Paulo, São Paulo, Brazil

    Silvio Stanzani

  7. Universidad Nacional de San Luis, San Luis, Argentina

    Veronica Gil-Costa

Rights and permissions

Reprints and permissions

Copyright information

© 2019 Springer Nature Switzerland AG

About this paper

Check for updates. Verify currency and authenticity via CrossMark

Cite this paper

Lázaro-Muñoz, AJ., López-Albelda, B., González-Linares, J.M., Guil, N. (2019). A Scheduling Theory Framework for GPU Tasks Efficient Execution. In: Senger, H., et al. High Performance Computing for Computational Science – VECPAR 2018. VECPAR 2018. Lecture Notes in Computer Science(), vol 11333. Springer, Cham. https://doi.org/10.1007/978-3-030-15996-2_11

Download citation

  • DOI: https://doi.org/10.1007/978-3-030-15996-2_11

  • Published:

  • Publisher Name: Springer, Cham

  • Print ISBN: 978-3-030-15995-5

  • Online ISBN: 978-3-030-15996-2

  • eBook Packages: Computer ScienceComputer Science (R0)

Publish with us

Policies and ethics

Search

Navigation