Thread 'CPDN re-doing work?'

Author	Message
old_user9662 Send message Joined: 2 Sep 04 Posts: 29 Credit: 13,100 RAC: 0	Message 5630 - Posted: 25 Oct 2004, 11:25:56 UTC Anyone else see this? .. -- .. 1wv6_100110161 - PH 1 TS 192786 - 27/01/1822 09:00 - H:M:S=0447:37:43 AVG= 8.36 DLT= 3.41 1wv6_100110161 - PH 1 TS 192787 - 27/01/1822 09:30 - H:M:S=0447:37:47 AVG= 8.36 DLT= 3.97 2004-10-24 22:11:57 [---] Running CPU benchmarks 2004-10-24 22:11:57 [---] Suspending computation and network activity - running CPU benchmarks CPDN Monitor got quit request... Detaching shared memory... 2004-10-24 22:12:58 [---] Benchmark results: 2004-10-24 22:12:58 [---] Number of CPUs: 1 2004-10-24 22:12:58 [---] 480 double precision MIPS (Whetstone) per CPU 2004-10-24 22:12:58 [---] 1352 integer MIPS (Dhrystone) per CPU 2004-10-24 22:12:58 [---] Finished CPU benchmarks 2004-10-24 22:12:59 [---] Resuming computation and network activity Starting model in /Users/cfreaner/boinc/projects/climateprediction.net... Created shared memory region key = 24515 Env Used=DYLD_LIBRARY_PATH=/Users/cfreaner/boinc/projects/climateprediction.net:../ Starting model ID 1wv6_100110161 Phase 1 Stack size=48.00 MB Waiting for model startup, this may take a minute... 1wv6_100110161 - PH 1 TS 192673 - 25/01/1822 00:30 - H:M:S=0447:22:53 AVG= 8.36 DLT= 0.00 1wv6_100110161 - PH 1 TS 192674 - 25/01/1822 01:00 - H:M:S=0447:23:26 AVG= 8.36 DLT=32.56 .. -- .. Notice the last time step before benchmarking and the first one after; CPDN backed up 114 timesteps - around 15 minutes work on my computer. I see the same thing happening when it switches with S@H every hour. I\'ve tried resetting my switch times from 60 min to 66 min, but it doesn\'t seem to have taken effect yet. C <img> ID: 5630 · Reply Quote

geophi Volunteer moderator Send message Joined: 7 Aug 04 Posts: 2187 Credit: 64,822,615 RAC: 5,275	Message 5633 - Posted: 25 Oct 2004, 12:50:06 UTC - in response to Message 5630. > 1wv6_100110161 - PH 1 TS 192787 - 27/01/1822 09:30 - H:M:S=0447:37:47 AVG= > 8.36 DLT= 3.97 > 2004-10-24 22:11:57 [---] Running CPU benchmarks > 2004-10-24 22:11:57 [---] Suspending computation and network activity - > running CPU benchmarks > CPDN Monitor got quit request... > Detaching shared memory... > 2004-10-24 22:12:58 [---] Benchmark results: > 2004-10-24 22:12:58 [---] Number of CPUs: 1 > 2004-10-24 22:12:58 [---] 480 double precision MIPS (Whetstone) per CPU > 2004-10-24 22:12:58 [---] 1352 integer MIPS (Dhrystone) per CPU > 2004-10-24 22:12:58 [---] Finished CPU benchmarks > 2004-10-24 22:12:59 [---] Resuming computation and network activity > Starting model in /Users/cfreaner/boinc/projects/climateprediction.net... > Created shared memory region key = 24515 > Env > Used=DYLD_LIBRARY_PATH=/Users/cfreaner/boinc/projects/climateprediction.net:../ > Starting model ID 1wv6_100110161 Phase 1 > Stack size=48.00 MB > Waiting for model startup, this may take a minute... > 1wv6_100110161 - PH 1 TS 192673 - 25/01/1822 00:30 - H:M:S=0447:22:53 AVG= > 8.36 DLT= 0.00 > .. -- .. > Notice the last time step before benchmarking and the first one after; CPDN > backed up 114 timesteps - around 15 minutes work on my computer. I see the > same thing happening when it switches with S@H every hour. I've tried > resetting my switch times from 60 min to 66 min, but it doesn't seem to have > taken effect yet. > > C > CPDN saves its work every 144 timesteps (3 model days), no matter what. You can get by this by checking "Leave applications in memory while pre-empted" in your general preferences. ID: 5633 · Reply Quote

old_user3222 Send message Joined: 30 Aug 04 Posts: 20 Credit: 11,115 RAC: 0	Message 5641 - Posted: 25 Oct 2004, 15:29:48 UTC As far as I am concerned I have noticed that the credit situation is being granted every 10802 trickles or timesteps or however you call them. Out of the 259248 timesteps in each of the 3 parts of one unit; everytime the total adds to a number + 10802, credit is granted. I work with a Intel(R) Pentium(R) 4 CPU 2.66GHz Pentium. ID: 5641 · Reply Quote

tiker Send message Joined: 7 Aug 04 Posts: 27 Credit: 766,972 RAC: 0	Message 5672 - Posted: 26 Oct 2004, 17:40:17 UTC - in response to Message 5641. > As far as I am concerned I have noticed that the credit situation is being > granted every 10802 trickles or timesteps or however you call them. > > Out of the 259248 timesteps in each of the 3 parts of one unit; everytime the > total adds to a number + 10802, credit is granted. > > I work with a Intel(R) Pentium(R) 4 CPU 2.66GHz Pentium. > The trickle checkpoint is every 10802 timesteps. The write-to-disk (saving current progress) checkpoint is every 144 timesteps. --- <img src="http://www.funkymonkey.org/boinc/sig2.php?sah=169421&lhc=75&team=Boinc-IRC&pah=925&cpdn=360&pirates=8&name=tiker"> ID: 5672 · Reply Quote

old_user3222 Send message Joined: 30 Aug 04 Posts: 20 Credit: 11,115 RAC: 0	Message 5673 - Posted: 26 Oct 2004, 18:01:54 UTC - in response to Message 5672. Last modified: 26 Oct 2004, 18:02:27 UTC > The trickle checkpoint is every 10802 timesteps. The write-to-disk (saving > current progress) checkpoint is every 144 timesteps. > --- Thanks for info - that's new to me - will pass it on. <img src="http://150.214.190.154/BOINCStatistics/Signature/Signature.php?userName=Liberto&cache=1&auxiliar=1"> ID: 5673 · Reply Quote

old_user9662 Send message Joined: 2 Sep 04 Posts: 29 Credit: 13,100 RAC: 0	Message 5711 - Posted: 28 Oct 2004, 3:12:32 UTC - in response to Message 5633. > > > CPDN saves its work every 144 timesteps (3 model days), no matter what. You > can get by this by checking "Leave applications in memory while pre-empted" in > your general preferences. > Thanks, geophi! I went back and dug through the FAQ and and found the info about memory. When in doubt, read the manual C <img src="http://www.boincstats.com/stats/banner.php?cpid=a7211a7321f3fd3cdc8eda3361db9b94" border="0"> ID: 5711 · Reply Quote