Journal of KIISE (정보과학회 논문지)

Korean Institute of Information Scientists and Engineers (한국정보과학회)
권호 표지
DOI QR코드

DOI QR Code

Sojourn Time Analysis Using SRPT Scheduling for Heterogeneous Multi-core Systems

Heterogeneous 멀티코어 시스템에서 SRPT 스케줄링을 사용한 체류 시간 분석

  • 양보미 (아주대학교 소프트웨어특성화학과) ;
  • 박현재 (아주대학교 컴퓨터공학과) ;
  • 최영준 (아주대학교 소프트웨어학과)
  • Received : 2015.12.31
  • Accepted : 2016.12.22
  • Published : 2017.03.15
⟨ Previous Next ⟩

Abstract

In this paper, we study the performance of recently popular multi-core systems in mobiles. Previous research on the multi-core performance usually focused on the desktop PC. However, there is enough scope to further analyze heterogeneous multi-core systems. Therefore, by extending homogeneous multi-core systems, we analyze the heterogeneous multi-core systems using Size Interval Task Allocation (SITA) for job allocation, and Shortest Remaining Processing Time (SRPT) scheduling, for each individual core. We propose a new computational method regarding the cutoff point, which is crucial in analyzing SITA, by calculating the sojourn time. This facilitate easy and accurate calculation of the sojourn time. We further confirm our analysis through the ESESC simulator that provides actual measurements.

본 논문에서는 최근 광범위하게 사용되고 있는 멀티 코어 환경에서의 모바일 장치의 성능에 대하여 연구하였다. 이전에 연구되어왔던 멀티 코어의 성능에 대한 분석은, 대부분 데스크톱 PC에서의 분석이었고, heterogeneous 멀티 코어에 대한 분석방법이 충분하지 않았다. 이러한 문제점을 보완하고자 homogeneous 멀티 코어의 분석 방법을 응용한 heterogeneous 멀티 코어 환경에서 성능을 분석하는 방법을 제안하였다. 본 연구에서는 이를 분석하는 데 있어서 작업의 할당에는 Size Interval Task Allocation (SITA) 기법을 사용하였고, 코어에서의 처리 방법은 Shortest Remaining Processing Time (SRPT) 기법을 사용하였다. 이 중 SITA 기법에서 가장 중요한 분석인 cutoff point에 대한 새로운 계산 방법을 제안하였고, 이를 체류 시간을 계산하는 데 사용함으로써 계산의 용이성과 정확성을 높였다. 또한, ESESC 시뮬레이터에서의 측정을 통해 분석값과 측정값에 큰 차이가 없음을 확인하였다.

Keywords

Acknowledgement

Grant : 디지털 헬스케어 소프트 웨어 시험평가센터 구축 과제

Supported by : 산업통상자원부

References

  1. NVIDIA, "the benefits of multiple CPU cores in mobile devices," White Paper, 2010. [Online] Available: ttps://www.nvidia.com/content/PDF/tegra_white_papers/Benefits-of-Multi-core-CPUs-in-Mobile-Devices_Ver1.2.pdf
  2. Scott Wasson (2005, August 31). IDF Fall 2005: Intel aims for performance per watt, [Online] available: http://techreport.com/review/8721/idf-fall-2005-intel-aims-for-performance-per-watt/3.
  3. Joel Hruska (2014, May 06). ARM's Race: An Attack Plan For Servers and Mobile [Online] Available: http://hothardware.com/reviews/arms-race-company-details-attack-plan-in-server-mobile-?page=3.
  4. Gepner, Pawel and Michal F. Kowalik, "Multi-Core Processors: New Way to Achieve High System Performance," Parallel Computing in Electrical Engineering, 2006. PARELEC 2006. International Symposium on. IEEE, 2006. pp. 9-13. 2006.
  5. Blake, G., Dreslinski, R. G., & Mudge, T, "A survey of multicore processors," IEEE Signal Processing Magazine, Vol. 26, No. 6, Nov. 2009.
  6. Harchol-Balter, Mor. Performance modeling and design of computer systems: queueing theory in action, Cambridge University Press, 2013.
  7. bachmat, Eitan and Hagit Sarfati, "Analysis of SITA policies," Performance Evaluation, Vol. 67, No. 2, pp. 102-120, 2010. https://doi.org/10.1016/j.peva.2009.09.007
  8. Harchol-Balter, Mor, Mark E. Crovella and Cristina D. Murta, "On choosing a task assignment policy for a distributed server system," Journal of Parallel and Distributed Computing, Vol. 59, No. 2, pp. 204-228, 1999. https://doi.org/10.1006/jpdc.1999.1577
  9. Harchol-Balter, Mor, "Task assignment with unknown duration," Distributed Computing Systems, 2000, Proc. 20th International Conference on. IEEE, 2000.
  10. Harchol-Balter, Mor. and Rein Vesilo, "To balance or unbalance load in size-interval task allocation," Probability in the Engineering and Informational Sciences, Vol. 24, No. 02, pp. 219-244, 2010. https://doi.org/10.1017/S0269964809990246
  11. Sharage, Linus E and Louis W. Miller, "The queue M/G/1 with the shortest remaining processing time discipline," Operations Research, Vol. 14, No. 4, pp. 670-684, 1966. https://doi.org/10.1287/opre.14.4.670
  12. Schrage, Lunes, "A Proof of the Optimality of the Shortest Remaining Processing Time Discipline," Operations Research, Vol. 16, No. 3, pp. 687-690, 1968. https://doi.org/10.1287/opre.16.3.687
  13. Smith, Donald R., "Technical Note-A New Proof of the Optimality of the Shortest Remaining Processing Time Discipline," Operations Research, Vol. 26, No. 1, pp. 197-199, 1978. https://doi.org/10.1287/opre.26.1.197
  14. Bansal, Nikhil and Mor Harchol-Balter, "Analysis of SRPT scheduling: Investigating unfairness," ACM, Vol. 29, No. 1, 2001.
  15. Bansal, Nikhil, "On the average sojourn time under M/M/1/SRPT," ACM SIGMETRICS Performance Evaluation Review, Vol. 31, No. 2, pp. 34-35, 2003. https://doi.org/10.1145/959143.959164
  16. Minghong Lin, Adam Wierman and Bert Zwart, "Heavy-traffic analysis of mean response time under Shortest Remaining," Performance Evaluation, Vol. 68, No. 10, pp. 955-966, 2011. https://doi.org/10.1016/j.peva.2011.06.001
  17. [Online] Available: https://en.wikipedia.org/wiki/Pareto_distribution.
  18. [Online] Available: https://en.wikipedia.org/wiki/Poisson_distribution.
  19. Hotovy, Steven, David Schneider and Timothy O'Don, "Analysis of the early workload on the Cornell Theory Center IBM SP2," ACM, Vol. 24, No. 1, 1996.
  20. ESESC simulator: Opne source program [Online] Available: http://masc.soe.ucsc.edu/esesc/.
  21. Ardestani, Ehsan K. and Jose Renau, "ESESC: A fast multicore simulator using time-based sampling," High Performance Computer Architecture (HPCA2013), 2013 IEEE 19th International Symposium on. IEEE, Feb. 2013.
  22. Milojicic, Dejan S., et al., "Process migration," ACM Computing Surveys, Vol. 32, Issue 3, Sept. 2000: pp. 241-299. https://doi.org/10.1145/367701.367728
  23. ARM, "big.LITTLE Technology: The Future of Mobile," White Paper, 2013 [Online] Available: https://www.arm.com/files/pdf/big_LITTLE_Technology_the_Futue_of_Mobile.pdf