On Mon, Sep 28, 2009 at 4:34 PM, <Casper.Dik at sun.com> wrote: > > >>For a test file with 24884706 bytes in /tmp (=tmpfs/ramdisk): >>- GNU "cksum" currently takes 181 seconds for 1000 iterations >>- AST "cksum" (called as external program) currently takes 244 seconds >>for 1000 iterations (partially caused by dragging more shared libraries >>around, startup time issue with libast-based applications and some other >>things) >>- AST "cksum" called as ksh93 builtin takes 216 seconds for 1000 >>iterations (e.g. ~~24 seconds are saved compared to the external >>application) >> > > On the same file? > > I would suggest that that is not a proper benchmark for cksum.
Why? Roland said Sun uses the same kind of loop to evaluate the performance of cksum and other tools. > > Why can't you run it once? The risk of statistical poisoning is too high. A statistic made from one sample is junk, not a statistic. -- Jennifer Pioch, Uni Frankfurt