On 3/22/07, Ph. Marek <philipp.marek@bmlv.gv.at> wrote:
> I don't really know VBscript, but there are some things I notice:
My goal was to recreate what Malcolm had done based solely on his
description above. But some of your ideas were interesting although they
don't appear match what Malcolm has done.
- You open the file that you just created *and still have opened*.
> Maybe the lookups should be done in a second loop and/or another script.
- IIRC Win32 still uses timers (with HZ=150?), so you'll get time readings
> only with 7ms granularity.
> So it might be better to do 1000 lookups and measure their cumulative
> time,
> with different numbers of entries in the directories.
>
> > Does that change the conclusion at all?
> Could you please create the files, and try to access some (in *random*
> order)
> in another script?
OK, I did that. The summary: no change. NTFS appears to be just as fast for
this scenario as any other modern filesystem.
Across 1024 file opens in a directory of 2^20 files, an average time
(seconds) of 0.025 for not-sharded and 0.026 for sharded. See the attached
scripts and log files for the details. I ran the read script three times,
recording the 3rd iteration's time to minimize the impact of my slower
laptop.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Thu Mar 22 18:09:31 2007