Advertisement

Dynamic Reconfiguration of Computer Platforms at the Hardware Device Level for High Performance Computing Infrastructure as a Service

  • Akihiro Misawa
  • Susumu Date
  • Keichi Takahashi
  • Takashi Yoshikawa
  • Masahiko Takahashi
  • Masaki Kan
  • Yasuhiro Watashiba
  • Yoshiyuki Kido
  • Chonho Lee
  • Shinji Shimojo
Conference paper
Part of the Communications in Computer and Information Science book series (CCIS, volume 864)

Abstract

Users’ needs and requirements for high performance computing (HPC) has become increasingly diversified. As user needs become increasingly diverse, it becomes increasingly difficult to own high-performance computing platforms themselves and the HPC platform provider are required to provide computing platforms to execute diverse applications. In this paper, we propose a computer architecture for providing HPC infrastructure dynamically and promptly as a cloud computing service in response to users’ request for computing platforms. In order to gain flexibility to accommodate various HPC jobs with application specific computing platforms, the proposed system reconfigures a software and hardware platform by utilizing the synergy of Open Grid Scheduler/Grid Engine and OpenStack. The experimental system developed in this research shows the high flexibility of hardware platform reconfiguration and the high performance of Spark’s benchmark application. In addition, our simulation evaluation shows that dynamic reconfigurable hardware cluster system can improve hardware resource utility rate, and also eliminating the worst case of resource congestion in the real-world operational record of our university’s computer center during the first half of 2016.

Keywords

Cloud computing Disaggregation Resource pool GPU/FPGA accelerator Hetero computer Distributed storage Job scheduling Resource management PCI Express OpenStack Software defined system 

References

  1. 1.
    Linux Accelerated Computing Instances of Amazon Web Service. http://docs.aws.amazon.com/AWSEC2/latest/UserGuide/accelerated-computing-instances.html. Accessed 31 Aug 2017
  2. 2.
  3. 3.
    A Set of Scripts to Create Simplest MPI Cluster on SoftLayer. https://github.com/irifed/softlayer-mpicluster. Accessed 31 Aug 2017
  4. 4.
    Apache Hadoop Homepage. http://hadoop.apache.org/. Accessed 31 Aug 2017
  5. 5.
    Apche Spark Homepage. http://spark.apache.org/. Accessed 31 Aug 2017
  6. 6.
    Intel Rack Scale Architecture Overview. http://goo.gl/ATtRR5. Accessed 31 Aug 2017
  7. 7.
    Open Compute Project. http://www.opencompute.org/. Accessed 31 Aug 2017
  8. 8.
    Han, S., Egi, N., Panda, A., et al.: Network support for resource disaggregation in next-generation datacenters. In: 12th International Proceedings on ACM Workshop on Hot Topics in Networks (HotNets), pp. 10:1–10:7. ACM, New York (2013)Google Scholar
  9. 9.
    Suzuki, J., Hidaka, Y., Higuchi, J., et al.: Expressether - Ethernet-based virtualization technology for reconfigurable hardware platform. In: 14th International Proceedings on IEEE Symposium on High-Performance Interconnects, Stanford, CA, USA, pp. 45–51. IEEE (2006)Google Scholar
  10. 10.
    Yoshikawa, T., Suzuki, J., Hidaka, Y., et al.: Bridge chip composing a PCIe switch over Ethernet to make a seamless disaggregated computer in data-center scale. In: 26th International Proceedings on IEEE Hot Chips 26 Symposium (HC26), Cupertino, CA, USA, p. 1. IEEE (2014)Google Scholar
  11. 11.
    OpenStack framework Homepage. https://www.openstack.org/software/. Accessed 31 Aug 2017
  12. 12.
    Open Grid Scheduler Homepage. http://gridscheduler.sourceforge.net/. Accessed 31 Aug 2017
  13. 13.
    Nomura, S., Mitsuishi, T., Suzuki, J., et al.: Performance analysis of the multi-GPU system with ExpEther. In: ACM SIGARCH Computer Architecture News - HEART 2014, vol. 42, issue 4, pp. 9–14. ACM, New York (2014)CrossRefGoogle Scholar
  14. 14.
    Mitsuishi, T., Suzuki, J., Hayashi, Y., et al.: Breadth first search on cost-efficient multi-GPU systems. In: ACM SIGARCH Computer Architecture News - HEART 2015, vol. 43, issue 4, pp. 58–63. ACM, New York (2015)CrossRefGoogle Scholar
  15. 15.
    Cybermedia Center. http://www.hpc.cmc.osaka-u.ac.jp/en/. Accessed 31 Aug 2017
  16. 16.
    Klusáček, D., Rudová, H.: Alea 2 - job scheduling simulator. In: 3rd Proceedings on ICST Conference on Simulation Tools and Techniques, Brussels, Belgium, Belgium, pp. 61:1–61:10. ICST (2010)Google Scholar
  17. 17.
    Misawa, A., Date, S., Takahashi, K., et al.: Highly reconfigurable computing platform for high performance computing infrastructure as a service: Hi-IaaS. In 7th International Proceedings on Cloud Computing and Services Science (CLOSER 2017), Setúbal, Portugal, pp. 135–146. Science and Technology Publications, Lda (SciTePress) (2017)Google Scholar
  18. 18.
    Suzuki, J., Hidaka, Y., Higuchi, J., et al.: Disaggregation and sharing of I/O devices in cloud data centers. IEEE Trans. Comput. 65(10), 3013–3026 (2016)MathSciNetCrossRefGoogle Scholar
  19. 19.
    Sefraoui, O., Aissaoui, M., Ekeuldj, M.: Dynamic reconfigurable component for cloud computing resources. Int. J. Comput. Appl. 88(7), 1–5 (2014)Google Scholar
  20. 20.
    Xu, F., Liu, F., Jin, H., et al.: Managing performance overhead of virtual machines in cloud computing: a survey, state of the art, and future directions. Proc. IEEE 102(1), 11–31 (2014)CrossRefGoogle Scholar
  21. 21.
    Katrinis, K., Syrivelis, D., Pnevmatikatos, D., et al.: Rack-scale disaggregated cloud data centers: the dReDBox project vision. In: 20th International Proceedings on Design, Automation and Test in Europe Conference and Exhibition (DATE), Cupertino, CA, USA, pp. 690–695. IEEE (2016)Google Scholar
  22. 22.
    Lee, G., Chun, B., Katz, R.H.: Heterogeneity-aware resource allocation and scheduling in the cloud. In: 3rd International Proceedings on USENIX conference on Hot topics in cloud computing (HotCloud 2011), p. 4. USENIX Association, Berkley (2011)Google Scholar
  23. 23.
    Wheeler, M.F., Pencheva, G., Tavakoli, R., et al.: Enabling high-performance computing as a service. Computer 45, 72–80 (2012)Google Scholar

Copyright information

© Springer International Publishing AG, part of Springer Nature 2018

Authors and Affiliations

  • Akihiro Misawa
    • 1
  • Susumu Date
    • 1
  • Keichi Takahashi
    • 1
  • Takashi Yoshikawa
    • 1
    • 2
  • Masahiko Takahashi
    • 2
  • Masaki Kan
    • 2
  • Yasuhiro Watashiba
    • 1
    • 3
  • Yoshiyuki Kido
    • 1
  • Chonho Lee
    • 1
  • Shinji Shimojo
    • 1
  1. 1.Cybermedia CenterOsaka UniversityIbarakiJapan
  2. 2.System Platform Research LaboratoriesNECKawasakiJapan
  3. 3.Information of ScienceNara Institute of Science and TechnologyIkomaJapan

Personalised recommendations