An Alternative Implementation for Accelerating Some Functions of Operating System

János Végh, Ádám Kicsák, Zsolt Bagoly, Péter Molnár

Abstract

Processes running under an operating system are independent and autonomous entities. However, they need to share resources, communicate, use OS services, etc. The operating system’s services can be reached through system calls, which contribute – sometimes excessive – overhead activity. In some cases the payload activity, used in the system call, is much shorter than that needed for implementing the Exceptional Control Flow, implementing the system call frame. In certain cases, the OS service in question can be implemented in an alternative way, practically without overhead. The paper presents such a case, using an easy to understand simple example, an alternative implementation of a simple binary semaphore. The semaphore has been implemented and tested in a prototyping environment, using an operating system running on a soft processor equipped with custom instruction. For implementing the semaphores, a reconfigurable device was used.

References

  1. Adapteva (2014). Parallella Board. http://www.parallella.org/.
  2. Akesson, J. F. (2001). Interprocess Communication Utilising Special Purpose Hardware. PhD thesis, Uppsala University.
  3. Altera (2011). Nios II Processor Reference. http://www.altera.com/literature/hb/nios2/ n2cpu nii5v1.pdf.
  4. Altera (2013). DE2 board. https://www.terasic.com.tw/cgibin/page/archive.pl?No=30.
  5. Bryant, R. E. and O'Hallaron, D. R. (2014). Computer Systems; A Programmer's Perspective. Pearson.
  6. C. A. Thekkath and H. M. Levy (1994). Hardware and software support for efficient exception handling. In ASPLOS-VI Proceedings. DOI: 10.1145/195470.195515.
  7. Ferreira, C. M. and Oliveira, A. S. R. (2009). RTOS Hardware Coprocessor Implementation in VHDL, page 6. Intellectual Property / Embedded Systems Conference (IP/ESC).
  8. Li, Q. and Yao, C. (2003). Real-Time Concepts for Embedded Systems. CMP Books.
  9. Li, Y., Gu, P.-P., and Wang, X.-X. (2011). The implementation of Semaphore Management in Hardware Real Time Operating System. Information Technology Journal, 10(1):158-163.
  10. Micrium (2012). µC/OS-II. http://micrium.com/rtos/ucosii/overview/.
  11. Ousterhout, J. K. (1990). Why aren't operating systems getting faster as fast as hardware? In USENIX Summer Conference.
  12. So, H. K.-H. (2007). BORPH: An Operating System for FPGA-Based Reconfigurable Computers. PhD thesis, University of California, Berkeley.
  13. Xilinx (2012). Zedboard. http://zedboard.org/.
Download


Paper Citation


in Harvard Style

Végh J., Kicsák Á., Bagoly Z. and Molnár P. (2014). An Alternative Implementation for Accelerating Some Functions of Operating System . In Proceedings of the 9th International Conference on Software Engineering and Applications - Volume 1: ICSOFT-EA, (ICSOFT 2014) ISBN 978-989-758-036-9, pages 494-499. DOI: 10.5220/0005104704940499


in Bibtex Style

@conference{icsoft-ea14,
author={János Végh and Ádám Kicsák and Zsolt Bagoly and Péter Molnár},
title={An Alternative Implementation for Accelerating Some Functions of Operating System},
booktitle={Proceedings of the 9th International Conference on Software Engineering and Applications - Volume 1: ICSOFT-EA, (ICSOFT 2014)},
year={2014},
pages={494-499},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0005104704940499},
isbn={978-989-758-036-9},
}


in EndNote Style

TY - CONF
JO - Proceedings of the 9th International Conference on Software Engineering and Applications - Volume 1: ICSOFT-EA, (ICSOFT 2014)
TI - An Alternative Implementation for Accelerating Some Functions of Operating System
SN - 978-989-758-036-9
AU - Végh J.
AU - Kicsák Á.
AU - Bagoly Z.
AU - Molnár P.
PY - 2014
SP - 494
EP - 499
DO - 10.5220/0005104704940499