Message boards : Number crunching : Running 2 WU's simultaneously with one processor after benchmark..
Message board moderation

To post messages, you must log in.

AuthorMessage
Profile Fuzzy Hollynoodles
Avatar

Send message
Joined: 13 Jul 05
Posts: 51
Credit: 10,626
RAC: 0
Message 8681 - Posted: 19 Jul 2005, 0:33:50 UTC
Last modified: 19 Jul 2005, 0:36:51 UTC

I was checking the Work tab in my BOINC Manager and saw that two WU's, an Einstein and a LHC, were running at the same time. I don't have a HT processor!

18-07-2005 15:58:29||Starting BOINC client version 4.45 for windows_intelx86
18-07-2005 15:58:29||Data directory: C:\Programmer\BOINC
18-07-2005 15:58:30|Einstein@Home|Computer ID: 56003; location: home; project prefs: home
18-07-2005 15:58:30|LHC@home|Computer ID: 38657; location: home; project prefs: home
18-07-2005 15:58:30|orbit@home|Computer ID: 245; location: home; project prefs: default
18-07-2005 15:58:30|SETI@home|Computer ID: 1138836; location: home; project prefs: home
18-07-2005 15:58:30|SETI@home Beta Test|Computer ID: 470; location: home; project prefs: default
18-07-2005 15:58:30||General prefs: from SETI@home (last modified 2005-07-04 10:43:55)
18-07-2005 15:58:30||General prefs: using separate prefs for home
18-07-2005 15:58:32||Remote control not allowed; using loopback address
18-07-2005 15:58:32|Einstein@Home|Deferring computation for result l1_0672.0__0672.2_0.1_T10_S4lA_2
18-07-2005 15:58:32|LHC@home|Resuming computation for result wjun1B_v6s4hvnom_mqx__15__64.289_59.299__6_8__6__25_1_sixvf_boinc21307_3 using sixtrack version 4.67
18-07-2005 15:58:34|orbit@home|Deferring communication with project for 28 minutes and 10 seconds
18-07-2005 15:58:40|SETI@home|Started upload of 07fe05aa.532.5536.828400.7_3_0

....

18-07-2005 23:50:08|Einstein@Home|Pausing result l1_0672.0__0672.2_0.1_T10_S4lA_2 (removed from memory)
18-07-2005 23:50:08|LHC@home|Restarting result wjun1B_v6s4hvnom_mqx__17__64.271_59.281__6_8__6__65_1_sixvf_boinc22947_1 using sixtrack version 4.67
18-07-2005 23:50:09||request_reschedule_cpus: process exited
19-07-2005 00:52:52||Using earliest-deadline-first scheduling because computer is overcommitted.
19-07-2005 01:23:55||Suspending computation and network activity - running CPU benchmarks
19-07-2005 01:23:55|LHC@home|Pausing result wjun1B_v6s4hvnom_mqx__17__64.271_59.281__6_8__6__65_1_sixvf_boinc22947_1 (removed from memory)
19-07-2005 01:23:55||request_reschedule_cpus: process exited
19-07-2005 01:23:57||Running CPU benchmarks
19-07-2005 01:24:54||Benchmark results:
19-07-2005 01:24:54|| Number of CPUs: 1
19-07-2005 01:24:54|| 1387 double precision MIPS (Whetstone) per CPU
19-07-2005 01:24:54|| 2782 integer MIPS (Dhrystone) per CPU
19-07-2005 01:24:54||Finished CPU benchmarks
19-07-2005 01:24:54||Resuming computation and network activity
19-07-2005 01:24:54||request_reschedule_cpus: Resuming activities
19-07-2005 01:24:54|Einstein@Home|Restarting result l1_0672.0__0672.2_0.1_T10_S4lA_2 using einstein version 4.79
19-07-2005 01:24:54|LHC@home|Restarting result wjun1B_v6s4hvnom_mqx__17__64.271_59.281__6_8__6__65_1_sixvf_boinc22947_1 using sixtrack version 4.67
19-07-2005 02:23:33|SETI@home|Started upload of 07fe05aa.532.5536.828400.7_3_0
19-07-2005 02:24:24|SETI@home|Temporarily failed upload of 07fe05aa.532.5536.828400.7_3_0: -106
19-07-2005 02:24:24|SETI@home|Backing off 1 hours, 13 minutes, and 40 seconds on upload of file 07fe05aa.532.5536.828400.7_3_0
19-07-2005 02:24:54|LHC@home|Pausing result wjun1B_v6s4hvnom_mqx__17__64.271_59.281__6_8__6__65_1_sixvf_boinc22947_1 (removed from memory)
19-07-2005 02:24:56||request_reschedule_cpus: process exited

Can someone tell me if they are both corrupted now?
"I'm trying to maintain a shred of dignity in this world" - Me
ID: 8681 · Report as offensive     Reply Quote
Heffed

Send message
Joined: 2 Sep 04
Posts: 71
Credit: 8,657
RAC: 0
Message 8683 - Posted: 19 Jul 2005, 0:53:35 UTC - in response to Message 8681.  

> Can someone tell me if they are both corrupted now?

They shouldn't be. This is an elusive BOINC issue that crops up from time to time. The only harm done should be slower processing due to the science apps fighting for CPU cycles.
ID: 8683 · Report as offensive     Reply Quote
Profile Fuzzy Hollynoodles
Avatar

Send message
Joined: 13 Jul 05
Posts: 51
Credit: 10,626
RAC: 0
Message 8686 - Posted: 19 Jul 2005, 1:16:34 UTC - in response to Message 8683.  

> > Can someone tell me if they are both corrupted now?
>
> They shouldn't be. This is an elusive BOINC issue that crops up from time to
> time. The only harm done should be slower processing due to the science apps
> fighting for CPU cycles.
>
>

Thanks! Phew!!! :-)
"I'm trying to maintain a shred of dignity in this world" - Me
ID: 8686 · Report as offensive     Reply Quote
Hefto99

Send message
Joined: 23 Oct 04
Posts: 8
Credit: 573,022
RAC: 0
Message 8687 - Posted: 19 Jul 2005, 1:30:54 UTC

It happened also to me. Results were OK, CPU time was just divided for both applications, so it took longer time to complete WUs. I restarted BOINC and
everything was normal again afterwards.
ID: 8687 · Report as offensive     Reply Quote
Profile Jim Baize
Avatar

Send message
Joined: 17 Sep 04
Posts: 103
Credit: 38,543
RAC: 0
Message 8692 - Posted: 19 Jul 2005, 2:09:58 UTC - in response to Message 8687.  

Is this a self-correcting issue? Meaning, after the 60 minutes of crunching both WU's does it then go back to crunching only 1 or does it continue to cruch 2 at a time?

Jim

> It happened also to me. Results were OK, CPU time was just divided for both
> applications, so it took longer time to complete WUs. I restarted BOINC and
> everything was normal again afterwards.
>
ID: 8692 · Report as offensive     Reply Quote
Profile Fuzzy Hollynoodles
Avatar

Send message
Joined: 13 Jul 05
Posts: 51
Credit: 10,626
RAC: 0
Message 8693 - Posted: 19 Jul 2005, 2:21:54 UTC - in response to Message 8692.  
Last modified: 19 Jul 2005, 2:26:56 UTC

> Is this a self-correcting issue? Meaning, after the 60 minutes of crunching
> both WU's does it then go back to crunching only 1 or does it continue to
> cruch 2 at a time?
>
> Jim
>
> > It happened also to me. Results were OK, CPU time was just divided for
> both
> > applications, so it took longer time to complete WUs. I restarted BOINC
> and
> > everything was normal again afterwards.
> >
>

Yes, it seemed to correct itself:

19-07-2005 01:23:55|LHC@home|Pausing result wjun1B_v6s4hvnom_mqx__17__64.271_59.281__6_8__6__65_1_sixvf_boinc22947_1 (removed from memory)
19-07-2005 01:23:55||request_reschedule_cpus: process exited
19-07-2005 01:23:57||Running CPU benchmarks
19-07-2005 01:24:54||Benchmark results:
19-07-2005 01:24:54|| Number of CPUs: 1
19-07-2005 01:24:54|| 1387 double precision MIPS (Whetstone) per CPU
19-07-2005 01:24:54|| 2782 integer MIPS (Dhrystone) per CPU
19-07-2005 01:24:54||Finished CPU benchmarks
19-07-2005 01:24:54||Resuming computation and network activity
19-07-2005 01:24:54||request_reschedule_cpus: Resuming activities
19-07-2005 01:24:54|Einstein@Home|Restarting result l1_0672.0__0672.2_0.1_T10_S4lA_2 using einstein version 4.79
19-07-2005 01:24:54|LHC@home|Restarting result wjun1B_v6s4hvnom_mqx__17__64.271_59.281__6_8__6__65_1_sixvf_boinc22947_1 using sixtrack version 4.67
19-07-2005 02:23:33|SETI@home|Started upload of 07fe05aa.532.5536.828400.7_3_0
19-07-2005 02:24:24|SETI@home|Temporarily failed upload of 07fe05aa.532.5536.828400.7_3_0: -106
19-07-2005 02:24:24|SETI@home|Backing off 1 hours, 13 minutes, and 40 seconds on upload of file 07fe05aa.532.5536.828400.7_3_0
19-07-2005 02:24:54|LHC@home|Pausing result wjun1B_v6s4hvnom_mqx__17__64.271_59.281__6_8__6__65_1_sixvf_boinc22947_1 (removed from memory)
19-07-2005 02:24:56||request_reschedule_cpus: process exited
19-07-2005 03:26:22||request_reschedule_cpus: process exited
19-07-2005 03:26:22|Einstein@Home|Computation for result l1_0672.0__0672.2_0.1_T10_S4lA_2 finished
19-07-2005 03:26:22||Resuming round-robin CPU scheduling.
19-07-2005 03:26:23|LHC@home|Restarting result wjun1B_v6s4hvnom_mqx__17__64.271_59.281__6_8__6__65_1_sixvf_boinc22947_1 using sixtrack version 4.67
19-07-2005 03:26:23|Einstein@Home|Started upload of l1_0672.0__0672.2_0.1_T10_S4lA_2_0
19-07-2005 03:26:42|Einstein@Home|Finished upload of l1_0672.0__0672.2_0.1_T10_S4lA_2_0
19-07-2005 03:26:42|Einstein@Home|Throughput 7362 bytes/sec
19-07-2005 03:26:43|Einstein@Home|Sending scheduler request to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi
19-07-2005 03:26:43|Einstein@Home|Requesting 0 seconds of work, returning 1 results
19-07-2005 03:26:45|Einstein@Home|Scheduler request to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi succeeded
19-07-2005 03:38:05|SETI@home|Started upload of 07fe05aa.532.5536.828400.7_3_0
19-07-2005 03:38:11|SETI@home|Temporarily failed upload of 07fe05aa.532.5536.828400.7_3_0: -106
19-07-2005 03:38:11|SETI@home|Backing off 2 hours, 38 minutes, and 2 seconds on upload of file 07fe05aa.532.5536.828400.7_3_0
19-07-2005 03:42:49||request_reschedule_cpus: process exited
19-07-2005 03:42:49|LHC@home|Computation for result wjun1B_v6s4hvnom_mqx__17__64.271_59.281__6_8__6__65_1_sixvf_boinc22947_1 finished
19-07-2005 03:42:49|LHC@home|Starting result wjun1B_v6s4hvnom_mqx__17__64.274_59.284__6_8__6__70_1_sixvf_boinc23152_3 using sixtrack version 4.67
19-07-2005 03:42:49|LHC@home|Started upload of wjun1B_v6s4hvnom_mqx__17__64.271_59.281__6_8__6__65_1_sixvf_boinc22947_1_0
19-07-2005 03:42:56|LHC@home|Finished upload of wjun1B_v6s4hvnom_mqx__17__64.271_59.281__6_8__6__65_1_sixvf_boinc22947_1_0
19-07-2005 03:42:56|LHC@home|Throughput 7551 bytes/sec
19-07-2005 03:42:57|LHC@home|Sending scheduler request to http://lhcathome-sched1.cern.ch/scheduler/cgi
19-07-2005 03:42:57|LHC@home|Requesting 0 seconds of work, returning 1 results
19-07-2005 03:42:58|LHC@home|Scheduler request to http://lhcathome-sched1.cern.ch/scheduler/cgi succeeded

It seemed that the failed upload to Seti got it back to normal!

But funny, when I marked the LHC WU, which was stated as preempted, the Show Graphic button was still highlighted!


"I'm trying to maintain a shred of dignity in this world" - Me
ID: 8693 · Report as offensive     Reply Quote
Udo

Send message
Joined: 14 Jul 05
Posts: 3
Credit: 562,479
RAC: 0
Message 8735 - Posted: 20 Jul 2005, 11:20:53 UTC
Last modified: 20 Jul 2005, 11:21:24 UTC

This is the 'normal' behaviour if you have more than one project.
At Switch time all activity is suspended and the next project gets the processor.
If you have 'even weighted' projects (i.e. each project with '100') then Boinc will switch the projects every hour.
Udo

ID: 8735 · Report as offensive     Reply Quote

Message boards : Number crunching : Running 2 WU's simultaneously with one processor after benchmark..


©2024 CERN