Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
![]() |
World Community Grid Forums
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
No member browsing this thread |
Thread Status: Active Total posts in this thread: 113
|
![]() |
Author |
|
hchc
Veteran Cruncher USA Joined: Aug 15, 2006 Post Count: 798 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
https://www.worldcommunitygrid.org/contribution/workunit/659931562
----------------------------------------ARP1_0033793_117_0 Linux Debian No Reply 2025-02-05 03:46:59 UTC 2025-02-06 15:46:59 UTC ARP1_0033793_117_1 Linux Ubuntu No Reply 2025-02-05 03:46:20 UTC 2025-02-06 15:46:20 UTC ARP1_0033793_117_2 Linux Ubuntu No Reply 2025-02-05 03:46:22 UTC 2025-02-06 15:46:22 UTC ARP1_0033793_117_3 Linux Debian In Progress 2025-02-06 15:46:28 UTC 2025-02-08 03:46:28 UTC ARP1_0033793_117_4 Linux Mint In Progress 2025-02-06 15:46:35 UTC 2025-02-08 03:46:35 UTC ARP1_0033793_117_5 Linux Ubuntu In Progress 2025-02-06 15:47:03 UTC 2025-02-08 03:47:03 UTC Got one, and the initial 3 wingmen all No Reply. The machine running it normally finishes in 9-10 hours, but it's taking 8 hours per 25% now, so perhaps it'll be done in 32 hours? Doozy.
[Edit 1 times, last edit by hchc at Feb 6, 2025 11:20:26 PM] |
||
|
Mike.Gibson
Ace Cruncher England Joined: Aug 23, 2007 Post Count: 12359 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
hchc
That sounds like the TimeStep has changed. Normally, a 36 sec TimeStep would have the progress% increasing by 0.020 or 0.021. If you are taking about 4 x as long, the progress would increase by 0.005 (occasionally 0.006). Changing the TimeStep to 9 secs results in 4 x as many calculations need to be done, so 4 x as long (less an amount to allow for only the same number of checkpoints). Mike |
||
|
Mike.Gibson
Ace Cruncher England Joined: Aug 23, 2007 Post Count: 12359 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
Al
It appears to be by design, or maybe they have heeded our comments. They restarted this time with the low generations but not the 3 ultras. Then as they expanded they released some higher generations and currently all but the top 5 generations.. However, 8 units seem to have stuck in generations 104 - 110 so atleast 11 stuck including the ultras. Mike |
||
|
hchc
Veteran Cruncher USA Joined: Aug 15, 2006 Post Count: 798 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
Mike Gibson said:
----------------------------------------hchc That sounds like the TimeStep has changed. Normally, a 36 sec TimeStep would have the progress% increasing by 0.020 or 0.021. If you are taking about 4 x as long, the progress would increase by 0.005 (occasionally 0.006). Changing the TimeStep to 9 secs results in 4 x as many calculations need to be done, so 4 x as long (less an amount to allow for only the same number of checkpoints). Mike Interesting. I checked the % increments, and it's jumping by 0.014% every few seconds. Currently 29% done after 9 hours.
|
||
|
Mike.Gibson
Ace Cruncher England Joined: Aug 23, 2007 Post Count: 12359 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
hchc
That corresponds to a TimeStep of 24 secs which should take only 50% extra time. Mike |
||
|
alanb1951
Veteran Cruncher Joined: Jan 20, 2006 Post Count: 952 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
hchc I wonder if it is running the 32-bit application?That corresponds to a TimeStep of 24 secs which should take only 50% extra time. Mike Cheers - Al. |
||
|
alanb1951
Veteran Cruncher Joined: Jan 20, 2006 Post Count: 952 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
Adri,
----------------------------------------Here's the next generation of one you posted about yesterday... ARP1_0033792_115_0 Ubuntu Valid 2025-02-06T05:47:06 2025-02-06T14:36:02 8.35/8.38 Cheers - Al. [Edited to acknowledge that wingman 2 finally validated] [Edit 1 times, last edit by alanb1951 at Feb 7, 2025 7:58:16 PM] |
||
|
hchc
Veteran Cruncher USA Joined: Aug 15, 2006 Post Count: 798 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
Mike Gibson said:
----------------------------------------That corresponds to a TimeStep of 24 secs which should take only 50% extra time. Interesting. 3.240%/hour is what BOINC Manager says. alanb1951 said: I wonder if it is running the 32-bit application? Unless there's another way to tell, it says x86_64 for that ARP/WRF process, so I assume it's 64-bit. Probably just a busy work unit. So far none of the 6 wingmen have returned it. It'd be cool to manually kick off the ultras or somehow script the feeder to prioritize old work units.
|
||
|
MJH333
Senior Cruncher England Joined: Apr 3, 2021 Post Count: 266 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() |
Here's an old friend:
workunit 659920306 ARP1_0034173_120_0 Windows 11 No Reply 2025-02-05 03:16:28 UTC 2025-02-06 15:16:28 UTC ARP1_0034173_120_1 Windows 10 No Reply 2025-02-05 03:16:22 UTC 2025-02-06 15:16:22 UTC ARP1_0034173_120_2 Windows 10 Valid 2025-02-05 03:16:27 UTC 2025-02-05 18:02:13 UTC 13.96 / 14.36 602.2 / 616.3 ARP1_0034173_120_3 Windows 11 Valid 2025-02-06 15:16:31 UTC 2025-02-07 10:04:11 UTC 18.63 / 18.63 630.4 / 616.3 ARP1_0034173_120_4 Windows 10 Invalid 2025-02-06 15:16:34 UTC 2025-02-07 02:39:37 UTC 11.28 / 11.34 523.8 / 523.8 I picked 34173 up on 15 February 2022 when it was at generation 98. It had been stuck and had just got going again after having its timestep reduced. My result _2 went to pending verification status against _4 but was validated by _3. I wonder whether the workunit is on the cusp of becoming stuck again. (It is currently on time_step 36.) Cheers, Mark |
||
|
Mike.Gibson
Ace Cruncher England Joined: Aug 23, 2007 Post Count: 12359 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
113 is now empty.
There are 82 extreme units missing from listings which would indicate that they have not moved since the restart so could be considered as stuck. This would include the 11 already identified. Mike |
||
|
|
![]() |