[tahoe-dev] Tahoe-LAFS v1.8.0 potentially delayed by performance issue

[Zooko O'Whielacronx]

On Fri, Aug 13, 2010 at 3:51 PM, Brian Warner <warner at lothar.com> wrote:
>
> It is entirely possible that, given the unreliability/slowness of the
> grid, Zooko's observed performance problems are a function of the
> servers that his download happened to use, more than the version of the
> downloader. But perhaps not. So we're still investigating.
...
> Ideally we'd like to see a repeatable example of a download on the
> pubgrid that is fast/smooth in 1.7 and slow/choppy in 1.8 and uses the
> same servers in both: that would be the most-useful bit of evidence.

for what it is worth, my results are pretty consistent if I download
100 MB at once. For shorter files (I tried 1 MB and 100 KB) the
results vary a lot.

Here is a summary of the seconds-to-completion all of my 100 MB tests
on the Public Test Grid
(http://tahoe-lafs.org/trac/tahoe-lafs/wiki/TestGrid ) that I posted
to #1170:

run 101: http://tahoe-lafs.org/trac/tahoe-lafs/ticket/1170#comment:18

v1.8.0c2 - 18m20s

run 102: http://tahoe-lafs.org/trac/tahoe-lafs/ticket/1170#comment:19

v1.7.1   -  8m55s

run 103: http://tahoe-lafs.org/trac/tahoe-lafs/ticket/1170#comment:22

v1.7.1   -  7m51s

run 104: http://tahoe-lafs.org/trac/tahoe-lafs/ticket/1170#comment:23

v1.8.0rc2 - 20m23s

run 105 - http://tahoe-lafs.org/trac/tahoe-lafs/ticket/1170#comment:25

v1.7.1   -  7m41s

run 106: http://tahoe-lafs.org/trac/tahoe-lafs/ticket/1170#comment:26

v1.8.0c2 - 14m29s

FYI I didn't mark this ticket as a critical regression and send mail
to tahoe-dev about it until I had seen this pattern shown here.

All of those runs have complete flogs and download-status-pages (if
available) attached to the ticket and named after their run number for
your debugging pleasure. :-)

Regards,

Zooko