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

[FreeStorm .]

Hi,

As discussed with Zooko on IRC:
For people who are sharing a server on Pubgrid, please do not change your
Tahoe-LAFS version until ticket 1170 [1] is done.
Because of performance tests

@+
Fred
[1] http://tahoe-lafs.org/trac/tahoe-lafs/ticket/1170


2010/8/13 Wayne Scott <wsc9tt at gmail.com>

> Does it matter what version the machine in the cluster are running?
>
> -Wayne
>
>
> On Fri, Aug 13, 2010 at 2:51 PM, Zooko O'Whielacronx <zooko at zooko.com>wrote:
>
>> Folks:
>>
>> It seems like Tahoe-LAFS v1.8.0c2 downloads this movie file:
>>
>>
>> http://127.0.0.1:3456/file/URI%3ACHK%3A4klgnafrwsm2nx3bqy24ygac5a%3Acrz7nhthi4bevzrug6xwgif2vhiacp7wk2cfmjutoz2ns3w45qza%3A3%3A10%3A1490710513/@@named=/bbb-360p24.i420.lossless.drc.ogg.fixed.ogg%2Bbbb-24fps.flac.via-ffmpeg.ogg
>>
>> About 1/3 as fast as Tahoe-LAFS v1.7.1 downloads the same file from
>> the same grid (the public Test Grid circa today).
>>
>> We're investigating (see ticket #1170 for our progress) and we might
>> delay the final release of Tahoe-LAFS v1.8.0 (previously planned for
>> August 15 -- this Sunday). This might turn out to require a patch
>> added on top of Tahoe-LAFS v1.8.0c2 to make it so it isn't such a
>> significant performance regression in this case. Or it might turn out
>> that this is a really peculiar situation which will almost never occur
>> for any other user and we'll decide to release v1.8.0c2 as v1.8.0
>> final.
>>
>> If you care about immutable download performance, please compare
>> v1.8.0c2 to v1.7.1 on your network with your files! And report your
>> results to this list. Nathan Eisenberg has done so [1], and shown that
>> v1.8.0c1 was substantially *faster* than v1.7.1 for his tests. (Note
>> that 1.8.0c1 and 1.8.0c2 are identical with regard to immutable file
>> download speed.) Earlier Kyle Markley did some benchmarking of v1.7.1
>> [2, 3, 4, 5, 6, 7, 8, 9], but nobody has yet replicated his experiment
>> using v1.8.0c2. I have done quite a bit of benchmarking this one
>> particular file from my office (SimpleGeo) and home, and my results
>> are all on #1170.
>>
>> Regards,
>>
>> Zooko
>>
>> [1] http://tahoe-lafs.org/pipermail/tahoe-dev/2010-August/004930.html
>> [2] http://tahoe-lafs.org/pipermail/tahoe-dev/2010-July/004776.html
>> [3] http://tahoe-lafs.org/pipermail/tahoe-dev/2010-July/004781.html
>> [4] http://tahoe-lafs.org/pipermail/tahoe-dev/2010-July/004782.html
>> [5] http://tahoe-lafs.org/pipermail/tahoe-dev/2010-July/004786.html
>> [6] http://tahoe-lafs.org/pipermail/tahoe-dev/2010-July/004811.html
>> [7] http://tahoe-lafs.org/pipermail/tahoe-dev/2010-July/004808.html
>> [8] http://tahoe-lafs.org/pipermail/tahoe-dev/2010-July/004838.html
>> [9] http://tahoe-lafs.org/pipermail/tahoe-dev/2010-August/004853.html
>>
>> http://tahoe-lafs.org/trac/tahoe-lafs/ticket/1170# does new-downloader
>> perform badly for certain situations (such as today's Test Grid)?
>> _______________________________________________
>> tahoe-dev mailing list
>> tahoe-dev at tahoe-lafs.org
>> http://tahoe-lafs.org/cgi-bin/mailman/listinfo/tahoe-dev
>>
>
>
> _______________________________________________
> tahoe-dev mailing list
> tahoe-dev at tahoe-lafs.org
> http://tahoe-lafs.org/cgi-bin/mailman/listinfo/tahoe-dev
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://tahoe-lafs.org/pipermail/tahoe-dev/attachments/20100813/ef340d41/attachment-0001.html>