Abstract: The inability of common benchmarks to gracefully scale to evaluate systems of various size and configuration contributes greatly to the incapacity usually shown in comparing varied systems under a common metric. This paper looks at an older, self-scaling I/O benchmark whose results can be normalized using techniques of predicted performance for comparative evaluation of a diverse range of stand-alone computer systems. We assert that there is no reason these comparisons need stop at local disk workstations and show the comparison to produce a valid evaluation of network file-system enabled computers using SUN's NFS protocol.

Available as postscript or PDF.   (I would really recommend PDF)

Additional Resources:
         Source for Peter Chen and David Patterson's Benchmark is available here.
         The source for the modifications to the benchmark adding RPC calls to flush remote cache is available here.
         Our experimental results are all available here.