BTW, what current explanation of benchmark needing if all projects on host use FLOPS for crediting? Start value time estimate? Then benchmark should run only at first join to project or when new hardware detected by BOINC. Why we need repetitive benchmarks now, explain once more, please...
----- Original Message ----- From: "Richard Haselgrove" <r.haselgr...@btinternet.com> To: "Martin" <m_boinc...@ml1.co.uk>; <john.mcl...@sybase.com> Cc: "BOINC Developers Mailing List" <boinc_dev@ssl.berkeley.edu>; <boinc_dev-boun...@ssl.berkeley.edu> Sent: Tuesday, September 29, 2009 5:15 PM Subject: Re: [boinc_dev] [boinc_alpha] Card Gflops in BOINC 6.10 >> (the 5 minute variety that we have now) >> >> jm7 > > That would be 32 seconds: > > 24-Sep-2009 01:05:11 [---] Running CPU benchmarks > 24-Sep-2009 01:05:11 [---] Suspending computation - running CPU benchmarks > 24-Sep-2009 01:05:11 [climateprediction.net] [cpu_sched] Preempting > hadam3p_mpl7_1997_2_1006354325_2 (left in memory) > 24-Sep-2009 01:05:11 [s...@home] [cpu_sched] Preempting > 18au09ab.1624.23957.13.10.68_2 (left in memory) > 24-Sep-2009 01:05:11 [s...@home] [cpu_sched] Preempting > 21au09ad.7828.19704.11.10.138_1 (left in memory) > 24-Sep-2009 01:05:11 [s...@home] [cpu_sched] Preempting > 21au09ad.7828.20522.11.10.23_1 (left in memory) > 24-Sep-2009 01:05:11 [s...@home] [cpu_sched] Preempting > 23au09ab.2188.11928.4.10.199_0 (removed from memory) > 24-Sep-2009 01:05:42 [---] Benchmark results: > 24-Sep-2009 01:05:42 [---] Number of CPUs: 4 > 24-Sep-2009 01:05:42 [---] 2361 floating point MIPS (Whetstone) per CPU > 24-Sep-2009 01:05:42 [---] 5210 integer MIPS (Dhrystone) per CPU > 24-Sep-2009 01:05:43 [---] Resuming computation > 24-Sep-2009 01:05:43 [climateprediction.net] [cpu_sched] Resuming > hadam3p_mpl7_1997_2_1006354325_2 > 24-Sep-2009 01:05:43 [s...@home] [cpu_sched] Resuming > 18au09ab.1624.23957.13.10.68_2 > 24-Sep-2009 01:05:43 [s...@home] [cpu_sched] Resuming > 21au09ad.7828.19704.11.10.138_1 > 24-Sep-2009 01:05:43 [s...@home] [cpu_sched] Resuming > 21au09ad.7828.20522.11.10.23_1 > 24-Sep-2009 01:05:43 [s...@home] [cpu_sched] Starting > 23au09ab.2188.11928.4.10.199_0(resume) > > I still don't like that (removed from memory) and Starting ... (resume) > when there's no rescheduling and no change of task involved. > _______________________________________________ > boinc_dev mailing list > boinc_dev@ssl.berkeley.edu > http://lists.ssl.berkeley.edu/mailman/listinfo/boinc_dev > To unsubscribe, visit the above URL and > (near bottom of page) enter your email address. > _______________________________________________ boinc_dev mailing list boinc_dev@ssl.berkeley.edu http://lists.ssl.berkeley.edu/mailman/listinfo/boinc_dev To unsubscribe, visit the above URL and (near bottom of page) enter your email address.
