[ Tuesday, February 29, 2000 ] Christian Robottom Reis wrote:
> James, I've run a whole truckload of benchmarks on raid1 with varying
> chunksizes on three different kernels, and on a plain disk. I'm about to
> publish some of the stuff, but I'm wondering very hard why is it that the
> readbalancing test showed _awful_ numbers on tiotest 0.21 and great
> numbers on 0.24 - any idea? Just have a look:
I'm not going to actually wade through these numbers... just far too
many to really deal with :)
tiotest.c has only changed cosmetically
tiobench.pl's only non-cosmetic change was in stat calculation to allow
for multiple runs in an efficient and harmonic (literally) way. Unfortuantely
for this case, it becomes the same calculation as before (first number
divided by second in tiotest output) for num_runs == 1.
Tell ya what, pick out an isolated case which is heavily reproducible,
print out the tiobench output, then print out the tiotest output.
James
