Embedded instrumentation for evaluating task sharing performance in a distributed computing system

Current trends in performance measurement on distributed computers are briefly reviewed. An example of application level performance measurement on a distributed computing system using software based instrumentation techniques is considered. Embedded software instrumentation for measuring the performance of a decentralized adaptive task sharing algorithm in a distributed computing system is described. For the application described, simplicity and flexibility provided by software-based instrumentation are demonstrated, and the limitations are discussed. The instrumentation subsystem, tradeoffs involved, design choices, and quantitative results obtained from the implementation on an experimental distributed computing system consisting of VAX stations connected by a local area network are discussed. >

[1]  A. Mink,et al.  Performance measurement of a shared-memory multiprocessor using hardware instrumentation , 1989, [1989] Proceedings of the Twenty-Second Annual Hawaii International Conference on System Sciences. Volume 1: Architecture Track.

[2]  Alan Mink,et al.  Multiprocessor performance-measurement instrumentation , 1990, Computer.

[3]  Alan Mink,et al.  A Hardware Instrumentation Approach for Performance Measurement of a Shared-Memory Multiprocessor , 1989 .

[4]  Alan Mink,et al.  Hardware-Assisted Multiprocessor Performance Measurement , 1987, Performance.

[5]  H. G. Rotithor,et al.  Decentralized decision-making in adaptive task sharing , 1990, Proceedings of the Second IEEE Symposium on Parallel and Distributed Processing 1990.

[6]  Barton P. Miller,et al.  IPS-2: The Second Generation of a Parallel Program Measurement System , 1990, IEEE Trans. Parallel Distributed Syst..