Replicated Imprecise Computations for Fault-Tolerant Real-Time Systems

  • Albert C. Yu
  • Kwei-Jay Lin
Part of the The Springer International Series in Engineering and Computer Science book series (SECS, volume 318)


Many real-time systems must be fault-tolerant and must handle failures in a timely manner. An approach incorporating the redundancy-masking technique and the imprecise computation model is presented in this paper. The redundancy-masking technique is preferred for time-critical applications than the rollback-and-retry techniques because of the fast error detection and the potential for forward error recovery. The imprecise computation model provides flexible functionality by trading off the quality of the result produced by a task with the amount of processing time required to produce it. It therefore permits the performance of a real-time system to degrade gracefully. Several algorithms for scheduling replicated periodic tasks on a real-time multiprocessor system have been studied and evaluated by stochastic analysis and Monte Carlo simulations. The results show that the algorithms are resilient under hardware failures.


Schedule Algorithm Task Allocation Processor Utilization Task System Early Deadline First 
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.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    J. Bannister and K. Trivedi. Task allocation in fault-tolerant distributed systems. Acta Information, 20:261–281, 1983.MATHCrossRefMathSciNetGoogle Scholar
  2. [2]
    Jen-Yao Chung, J. W.-S. Liu, and K. J. Lin. Scheduling periodic jobs that allow imprecise results. IEEE Transactions on Computers, pages 1156–1174, September 1990.Google Scholar
  3. [3]
    S.K. Dhall and C.L. Liu. On a real-time scheduling problem. Operations Research, 26:127–140, 1978.MATHCrossRefMathSciNetGoogle Scholar
  4. [4]
    N. Karmarkar. A new polynomial-time algorithm for linear programming. Combinatorica, 4:373–395, 1984.MATHCrossRefMathSciNetGoogle Scholar
  5. [5]
    K.H. Kim and H.O. Welch. Distributed execution of recovery block. IEEE Trans. Computers, C-38(5):626–636, May 1989.CrossRefGoogle Scholar
  6. [6]
    H. Kopetz, et al. Distributed fault-tolerant real-time systems: The mars approach. IEEE Micro, February 1989.Google Scholar
  7. [7]
    C.M. Krishna and K.G. Shin. On scheduling tasks with a quick recovery from failure. IEEE Trans. Computers, C-35(5):448–454, May 1986.Google Scholar
  8. [8]
    K. J. Lin, S. Natarajan, and J. W.-S. Liu. Imprecise results: Utilizing partial computations in real-time systems. In Proceedings of the Eighth Real-Time Systems Symposium, pages 210–217, San Jose, Calif., December 1987.Google Scholar
  9. [9]
    C. L. Liu and J. W. Layland. Scheduling algorithms for multiprogramming in a hard-real-time environment. JACM, 20(1):46–61, 1973.MATHCrossRefMathSciNetGoogle Scholar
  10. [10]
    R. Strong, D. Dolev, and F. Cristian. New latency bounds for atomic broadcasts. In Proc. of 11th Real-Time Sys. Symp., pages 156–165, 1990.Google Scholar
  11. [11]
    A. C. Yu and K.J. Lin. Scheduling parallelizable imprecise computations on multiprocessors. In Proceedings of 5th IEEE Int. Parallel Processing Symposium, pages 531–536, April 1991.Google Scholar
  12. [12]
    A.C. Yu. Scheduling parallel real-time tasks that allow imprecise results. Report UIUCDCS-R-92-1738, University of Illinois Department of Computer Science, Urbana, Illinois, March 1992.Google Scholar

Copyright information

© Kluwer Academic Publishers 1995

Authors and Affiliations

  • Albert C. Yu
    • 1
  • Kwei-Jay Lin
    • 2
  1. 1.Systems Division, RE/R8/N541Hughes Aircraft CompanyEl Segundo
  2. 2.Department of ECEUniversity of CaliforniaIrvine

Personalised recommendations