Performance based multiple server support for distributed real time applications

ABSTRACT

In real time systems, tasks must be performed correctly within time constraints. The system throughput is also a key performance measure. One of the difficulties, which the system designer faces, is to control these performance factors. A heavily requested computation, or a subtask that requires long service time, may reduce the system performance. This paper presents two different solutions for the above problem using multiple servers to support the execution of such a subtask. The first approach uses a pipeline subtask decomposition while the second approach uses a set of parallel servers. A queuing model, that takes into consideration the communication delay, is used for the analysis. In addition, the proposed approaches consider the operating environment, both system and user constraints, and the required resource cost.

REFERENCES

• 1: R.A. Ammax, H. A. Sholl. B. Qin, L. Lipsky, ,and J. Munoz; "An Architecture Assessn~ent Environment for the Design and the Analysis of Massively PandlelComputers" iEEE International Conference on System, Man, and Cybernetics, Nov. 1989.

INDEX TERMS

Primary Classification:
  G. Mathematics of Computing
  G.m MISCELLANEOUS
      Subjects: Queueing theory**

Additional Classification:
  C. Computer Systems Organization
  C.3 SPECIAL-PURPOSE AND APPLICATION-BASED SYSTEMS
      Subjects: Real-time and embedded systems
  C.4 PERFORMANCE OF SYSTEMS
      Subjects: Performance attributes

  D. Software
  D.4 OPERATING SYSTEMS
      D.4.8 Performance
          Subjects: Queueing theory

  F. Theory of Computation
  F.2 ANALYSIS OF ALGORITHMS AND PROBLEM COMPLEXITY
      F.2.2 Nonnumerical Algorithms and Problems
          Subjects: Computations on discrete structures

General Terms:
Algorithms, Performance, Theory