Advertisement

Fine-Grained System-Call Scheduling in CEFOS on Commodity Processors

  • Shigeru Kusakabe
  • Kentaro Iio
  • Hideo Taniguchi
  • Makoto Amamiya
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3019)

Abstract

We are developing an operating system, CEFOS, based on a fine-grain multi-threading model. By consisting a process with fine-grain threads, we can employ flexible scheduling schemes as long as thread dependencies allow. This paper presents WSC mechanism in CEFOS, which buffers multiple system-calls until the number of system-call requests or the time-slice reaches some threshold and then transfers the control to the operating system kernel with a bucket of the buffered system-call requests. We have developed CEFOS based on Linux and measured the system-call performance on commodity Intel processors. While fine-grained system-call scheduling with WSC caused frequent cache misses, it reduced context switches and improved the total execution time thanks to the reduction of TLB misses.

Keywords

Elapse Time Total Execution Time User Process Context Switch Clock Speed 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Alverson, G., et al.: Tera Hardware-Software Cooperation. In: Proc. Supercomputing, Sam Jose (1997)Google Scholar
  2. 2.
    Introduction to Hyper-Threading Technology, Intel, Document Number 250008- 002 (2001)Google Scholar
  3. 3.
    Lidetke, J.: Latest Development in Operating System, Toward Real Microkernels. CACM 39(9), 70–77 (1996)Google Scholar
  4. 4.
    Li, L., Kusakabe, S., Amamiya, M.: Multithreading Language FUCE-C and Its Implementation. In: Hinokuni Symposium 2002, pp.459-466 (2002)Google Scholar
  5. 5.
  6. 6.
    Kusakabe, S., et al.: Parallel and Distributed Operating System CEFOS (Communication-Execution Fusion OS. IPSJ ISG Technical Notes 99(251), 25–32 (1999) (in Japanese)Google Scholar
  7. 7.
    Redstone, J., et al.: An Analysis of Operating System Behavior on a Simultaneous Multithreaded Architecture. ACM SIGPLAN Notices 35(11), 245–256 (2000)CrossRefGoogle Scholar
  8. 8.
    Taniguchi, H.: DRD: New Connection Mechanism between Internal Kernel and External Kernel. Transaction of IEICE J85-D-1(2) (2002) (in Japanese)Google Scholar
  9. 9.
    Thomas, E.A., et al.: Scheduler Activation: Effective kernel Support for the User- Level Management of Parallelism. In: Proc. of the 13th ACM Symp. on OS Principles, pp. 95–109 (1991)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2004

Authors and Affiliations

  • Shigeru Kusakabe
    • 1
  • Kentaro Iio
    • 1
  • Hideo Taniguchi
    • 1
  • Makoto Amamiya
    • 1
  1. 1.Grad. School of Information Science and Electrical EngineeringKyushu UniversityFukuokaJapan

Personalised recommendations