Index  | Recent Threads  | Unanswered Threads  | Who's Active  | Guidelines  | Search
 

Quick Go ยป
No member browsing this thread
Thread Status: Active
Total posts in this thread: 8
[ Jump to Last Post ]
Post new Thread
Author
Previous Thread This topic has been viewed 1772 times and has 7 replies Next Thread
TheBizii
Advanced Cruncher
Slovenia
Joined: Jan 4, 2016
Post Count: 51
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Weird WU

I have come across a weird OPN WU. I fetched it yesterday and my computer just started working on it. After four minutes, the WU is halfway done.



However, I have also noticed a strange behavior (this is not the first time, so I guess something is either wrong with application code or my CPU is computing with otherworldly speeds). The progress got from 47% to 50% really fast and then reset itself back to 47%. That was done for like 10 times before the progress normalized. I recorded the same behavior here.
----------------------------------------
[Edit 3 times, last edit by TheBizii at Jun 9, 2020 2:29:23 PM]
[Jun 9, 2020 2:15:49 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Weird WU

Some ligands are really fast. As for the alternating just before checkpoint, that's normal. The program seeks the lowest / best docking energy. If it's not happy with the first pass, it retreats and tries again until it either finds a minimum threshold or gives up, then moves on to the next docking position.

PS, your image is way too wide for this (ancient) forum software, it can deal with 600 pixels wide. No auto shrink.
[Jun 9, 2020 2:30:41 PM]   Link   Report threatening or abusive post: please login first  Go to top 
TheBizii
Advanced Cruncher
Slovenia
Joined: Jan 4, 2016
Post Count: 51
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: Weird WU

Some ligands are really fast. As for the alternating just before checkpoint, that's normal. The program seeks the lowest / best docking energy. If it's not happy with the first pass, it retreats and tries again until it either finds a minimum threshold or gives up, then moves on to the next docking position.

PS, your image is way too wide for this (ancient) forum software, it can deal with 600 pixels wide. No auto shrink.


Roger that, thanks!
[Jun 9, 2020 2:49:30 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Falconet
Master Cruncher
Portugal
Joined: Mar 9, 2009
Post Count: 3295
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: Weird WU

I've seen several WUs that run to 50% in 10 minutes and then take a long time till 100%, ending up running somewhere between 3 and 3 and half hours, which is the normal runtime on my computer.

Seems like normal behaviour, work units are validated correctly.
----------------------------------------


AMD Ryzen 5 1600AF 6C/12T 3.2 GHz - 85W
AMD Ryzen 5 2500U 4C/8T 2.0 GHz - 28W
AMD Ryzen 7 7730U 8C/16T 3.0 GHz
[Jun 9, 2020 2:51:07 PM]   Link   Report threatening or abusive post: please login first  Go to top 
TheBizii
Advanced Cruncher
Slovenia
Joined: Jan 4, 2016
Post Count: 51
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: Weird WU

I've seen several WUs that run to 50% in 10 minutes and then take a long time till 100%, ending up running somewhere between 3 and 3 and half hours, which is the normal runtime on my computer.

Seems like normal behaviour, work units are validated correctly.


Yeah. that's exactly what's happening with that WU right now.
[Jun 9, 2020 3:07:03 PM]   Link   Report threatening or abusive post: please login first  Go to top 
uplinger
Former World Community Grid Tech
Joined: May 23, 2005
Post Count: 3952
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: Weird WU

Generally what you're seeing in this case is that the work unit you were given was a combination of multiple ligands (probably 2 since it got to 50% quickly). We estimate the runtime needed for each ligand. For this work unit, you may have been given a really small ligand which is really quick. Then got paired with a more difficult ligand that could take upwards of 12 hours. We split the second ligand so that it would finish the current work unit around 3 hours. The way the percentages work on the progress bar is based on ligands and number of runs in a given ligand. Which is why you saw it at 50% quickly.

Thanks,
-Uplinger
[Jun 9, 2020 3:29:43 PM]   Link   Report threatening or abusive post: please login first  Go to top 
sterl
Cruncher
Joined: Jul 23, 2013
Post Count: 5
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: Weird WU

I have noticed some of the really quick work units,

Its the ones that last 2 or 3 days that worry me, I dont want to spend 3 days on a unit and then it arrives late.
[Jun 11, 2020 7:12:10 AM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Weird WU

Why worry if the standard deadline is 7 days. Those that run devices with days of buffered work and know their's tasks can take 3 days (Android probably), might consider to go into the device profile > profile> custom profile radio button and limit OPN1 near the bottom to equal the number of active BOINC threads. My phone computes 1 at the time, so I've set the profile to 1. When the result is returned a fresh one is fetched with again full 7 day deadline. Even when it's a repair job it gets 3.5 days to work it out. More than enough.
[Jun 11, 2020 8:04:36 AM]   Link   Report threatening or abusive post: please login first  Go to top 
[ Jump to Last Post ]
Post new Thread