A new approach to embedded computer benchmarking

[摘要] ENGLISH ABSTRACT: Years of experience gained in the field while designing and building new digital hardware forclients in both the private and public sectors have shown that choosing the correct processorfor the job beforehand is difficult.The reason is the type of situation where the final speed of the application cannot be knownbefore it is tested on the actual hardware, but that the hardware cannot be built before it isknown which CPU will be fast enough to run the application.Designers have traditionally tried to reduce the risk by referring to various benchmarkprograms to compare processors with each other, and then by over designing. In this regardvarious attempts have been made to describe the performance of CPU's, but these aregenerally very application-specific and the accuracy depends on who is doing the measuring.Manufactures have been known to quote figures that will portray their hardware in the bestpossible light.The purpose of this thesis will be to develop a robust, simple and quick way to determinewhat performance a CPU will achieve in a given practical application. The theory will then betested on several CPU platforms.The development of such a method has very practical application in the engineering industry,as the type and number of CPU's in a design have very real cost implications. The method willalso have spin-offs in the System-On-a Chip (SOIC) and FPGA environment when themethod is used to analyse the requirements of a given application. The results can then beused to influence the data flow paths and CPU architecture of such a design.The thesis first does a literature survey of current benchmarking methods. This will then beused to influence the postulation of a theory of how the problem of benchmarking should betackled. The theory will then be tested on several platforms, as stated above.

[发布机构] Stellenbosch University