Never ending WU's ?

Message boards : Number crunching : Never ending WU's ?
Message board moderation

To post messages, you must log in.

Previous · 1 · 2

AuthorMessage
Bryn Mawr

Send message
Joined: 16 Aug 21
Posts: 39
Credit: 14,533,126
RAC: 12,694
Message 1631 - Posted: 21 May 2022, 10:01:53 UTC - in response to Message 1630.  

Just aborted 62895949 after nearly 12h CPU runtime and showing 45% progress.
WU's come in showing a runtime of 5.5h. I would like to see the expected runtime somewhat more accurate if WU's run normal.
There seems to be quite a big spread in runtimes. Other projects are more predictable.


The expected run time appears to be the average of your last ‘n’ work units rather than tailored to the individual WU and, as you say, there is a big spread in the size of then with most around 5 hours but some, as in the one you aborted, running for 24 hours. The credits given do reflect the run time so, if that worries you, you don’t loose out.
ID: 1631 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
sam6861

Send message
Joined: 28 Dec 20
Posts: 13
Credit: 8,957,185
RAC: 0
Message 1632 - Posted: 22 May 2022, 2:45:27 UTC - in response to Message 1631.  

This topic is 1 year old. We now have checkpoints, and more accurate percent complete status.

Each tasks have 5 day deadline at the time of this reply, plenty of time for my slow intel atom N270 which mostly takes 1 to 3 days to complete 2 tasks.
ID: 1632 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
hoarfrost
Volunteer moderator
Project administrator
Project developer

Send message
Joined: 11 Oct 20
Posts: 324
Credit: 23,795,098
RAC: 11,087
Message 1633 - Posted: 22 May 2022, 7:24:37 UTC

Hello! :)

Tasks for target corona_NSP16_v1 much longer than for previous target and varies from ~5 hours up to ~20(!) hours for my Ryzen 3900X.
Each task (excluding last workunit of whole set) is a job for inspecting of reactions of 500 ligands with target protein, but if one ligand "looks" good, reaction with it is computed more times. If task have a subset of "good" ligands for current target, duration of its computation may be long and process should progress unevenly.
ID: 1633 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Bryn Mawr

Send message
Joined: 16 Aug 21
Posts: 39
Credit: 14,533,126
RAC: 12,694
Message 1634 - Posted: 22 May 2022, 9:27:41 UTC - in response to Message 1633.  

Hello! :)

Tasks for target corona_NSP16_v1 much longer than for previous target and varies from ~5 hours up to ~20(!) hours for my Ryzen 3900X.
Each task (excluding last workunit of whole set) is a job for inspecting of reactions of 500 ligands with target protein, but if one ligand "looks" good, reaction with it is computed more times. If task have a subset of "good" ligands for current target, duration of its computation may be long and process should progress unevenly.


Thanks for explaining the why as well as the what.
ID: 1634 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Weyland

Send message
Joined: 19 Feb 22
Posts: 2
Credit: 7,327,632
RAC: 6,932
Message 1635 - Posted: 25 May 2022, 17:38:47 UTC
Last modified: 25 May 2022, 17:39:23 UTC

I have tasks running between 10 and 20 hours, which expect to run for 5 to 35 days more.

Not sure what to do about them really, since at this rate they will invariably time out.
ID: 1635 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Weyland

Send message
Joined: 19 Feb 22
Posts: 2
Credit: 7,327,632
RAC: 6,932
Message 1636 - Posted: 31 May 2022, 17:23:30 UTC - in response to Message 1635.  

I have tasks running between 10 and 20 hours, which expect to run for 5 to 35 days more.

Not sure what to do about them really, since at this rate they will invariably time out.


It was a fluke, by the way, none of those tasks took over 48 hours in the end.
May have something to do with recently updated chipset drivers.
ID: 1636 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
rbpeake

Send message
Joined: 18 Jul 22
Posts: 2
Credit: 277,889
RAC: 0
Message 1667 - Posted: 20 Jul 2022, 14:05:35 UTC - in response to Message 1636.  

Also, BOINC needs time to adjust its estimated completion times for better accuracy for any new project/new work units it encounters.
ID: 1667 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Previous · 1 · 2

Message boards : Number crunching : Never ending WU's ?

©2024 SiDock@home Team