SPARC/Solaris and workunit processing time.
12 Feb 2006 17:29:12 UTC
Topic 190763
(moderation:
Can you compare workunit processing time between workstations with any meaning or do they vary greatly. I seem to be averaging about 60000 per workunit. Another user seems to be about 82000. I am on a Sunblade 100 500Mhz Ultrasparc IIe 2Gb ram and he is on a Sunfire V250 1.2 Ghz Ultrasparc IIIi. I would expect I would have the longer processing times.
SPARC/Solaris and workunit processing time.
)
The length of WU differs. On the same machine I had some which took 1.400 seconds and now I get some using 14.000 seconds. You can compare boxes only, when they are working on the same WU.
yes, but if i always get only
)
yes, but if i always get only the half of my "claimed credit" doesn't it show that my box is to slow ??? :-)
16925481 4315576 4 Feb 2006 20:58:40 UTC 5 Feb 2006 23:34:10 UTC Over Success Done 81,918.52 81.31 46.56
16925475 4315574 4 Feb 2006 20:57:38 UTC 5 Feb 2006 23:35:12 UTC Over Success Done 82,051.10 81.44 45.73
16846560 4290769 3 Feb 2006 19:24:29 UTC 4 Feb 2006 20:58:40 UTC Over Success Done 82,097.69 81.48 45.79
16846555 4290768 2 Feb 2006 19:12:40 UTC 4 Feb 2006 19:20:34 UTC Over Success Done 82,226.80 81.61 47.09
How credit is given is
)
How credit is given is explained here
http://boinc.berkeley.edu/credit.php
yes.. thank you... you are
)
yes.. thank you... you are everywhere to help :-)
but if my "claimed credit" is so high, it is certainly because of something wrong on my system, isn't it ?
RE: yes.. thank you... you
)
[AF>...,
Your claim could be high for at least a couple of reasons.
1) The benchmarks code just happens to favor your processor/machine configuration, or..
2) Your machine has now a process running that was not in-process when the benchmarking was done, and so could not be accounted for in benching, and that is processor-intensive enough to slow your machine considerably, for example a screensaver.
Michael
microcraft
"The arc of history is long, but it bends toward justice" - MLK
RE: RE: yes.. thank
)
now, i think, it's due to the computation-way from einstein...
because if i compare with the 2 other projects i run on this box... einstein is the only one where i claim the double....
an example with sztaki
1655586 7028 9 Mar 2006 7:16:26 UTC 9 Mar 2006 9:01:59 UTC Over Success Done 1,086.50 1.58 1.58
1655587 5615 9 Mar 2006 7:15:08 UTC 9 Mar 2006 7:50:56 UTC Over Success Done 887.56 2.14 1.58
1655588 8426 9 Mar 2006 7:14:24 UTC 9 Mar 2006 9:59:02 UTC Over Success Done 1,192.25 2.16 1.58
--> 1655589 11445 9 Mar 2006 7:15:04 UTC 9 Mar 2006 7:44:58 UTC Over Success Done 816.84 0.83 1.58
and another with seti
--> 253990747 2187475 8 Mar 2006 9:54:20 UTC 10 Mar 2006 14:09:58 UTC Over Success Done 11,223.52 11.41 31.45
253990748 1858354 8 Mar 2006 9:54:20 UTC 8 Mar 2006 21:44:37 UTC Over Success Done 42,246.00 46.68 31.45
253990749 1909603 8 Mar 2006 9:54:20 UTC 8 Mar 2006 17:04:15 UTC Over Success Done 11,997.53 19.86 31.45
253990750 1791368 8 Mar 2006 9:54:20 UTC 9 Mar 2006 7:09:08 UTC Over Success Done 19,493.59 31.45 31.45