| 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: 7
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hi!
I'm new to this, but it seems really cool! One thing I've noticed is that my elapsed time will frequently stay at a certain time for longer than it should. This leads to the time elapsed showing, for example, 6 seconds instead of 10. Same goes for the estimated time left. I was wondering why this is, and what I could to to fix this issue. Thanks in advance! |
||
|
|
Sgt.Joe
Ace Cruncher USA Joined: Jul 4, 2006 Post Count: 7847 Status: Offline Project Badges:
|
Hi! I'm new to this, but it seems really cool! One thing I've noticed is that my elapsed time will frequently stay at a certain time for longer than it should. This leads to the time elapsed showing, for example, 6 seconds instead of 10. Same goes for the estimated time left. I was wondering why this is, and what I could to to fix this issue. Thanks in advance! Don't worry about it. If your computer is doing something else, it doesn't necessarily have time to update that field every second. Pretty much, if the field is updating, your computer is working on the work unit. Since BOINC runs at the lowest priority, everyother task your computer does takes precedence. Cheers
Sgt. Joe
*Minnesota Crunchers* |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Thanks! This helps a lot.
----------------------------------------Oh yeah, one clarifying question, it'll still do this while it's running, right? Because it does pause when the CPU is busy, but it will also do this (stopping for one extra second) if it's running the project, leading to me needing to wait much longer than the estimated time. [Edit 1 times, last edit by Former Member at Jan 16, 2021 3:00:17 AM] |
||
|
|
Bryn Mawr
Senior Cruncher Joined: Dec 26, 2018 Post Count: 385 Status: Offline Project Badges:
|
Thanks! This helps a lot. Oh yeah, one clarifying question, it'll still do this while it's running, right? Because it does pause when the CPU is busy, but it will also do this (stopping for one extra second) if it's running the project, leading to me needing to wait much longer than the estimated time. It is quite normal on my systems for the times to only update every 15 seconds or so but that’s ok because it updates by 15 seconds. Select a work unit and click on properties (or look at the WUs in the results status tab) and compare the CPU time with the elapsed time, this will give you a measure of how efficiently your machine is processing - they should be within one or two percent unless you are using the pc for other work. |
||
|
|
floyd
Cruncher Joined: May 28, 2016 Post Count: 47 Status: Offline Project Badges:
|
I'm not sure which of two effects you describe:
A) The elapsed time does not update every second but increases at normal speed overall. That can be caused by very high CPU load, permanent disc activity or network congestion when monitoring a remote client. Anything that can cause a delay effectively. The progress bar and remaining time updating in larger steps can be normal, that's application specific. Also see the post by Sgt.Joe. B) The elapsed time increases slower than real "wall clock" time. B1) Other CPU load does not allow the applications to run full speed. That's what Bryn Mawr describes. B2) It's a client setting. First check where your settings come from. The Manager's computing preferences will tell you at the top if you're using local preferences or web preferences. For local preferences look for "Usage limits" below. It's common practice to use 100% of CPU time and control the computing power by setting the percentage of CPUs to use. Else the applications would switch between on and off in quick succession, not running full time. For web preferences check your device profile at WCG for equivalent settings. There's pre-defined sets of preferences, the one you chose could limit the CPU time. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hi everyone! Thanks for helping once again.
It is quite normal on my systems for the times to only update every 15 seconds or so but that’s ok because it updates by 15 seconds. Select a work unit and click on properties (or look at the WUs in the results status tab) and compare the CPU time with the elapsed time, this will give you a measure of how efficiently your machine is processing - they should be within one or two percent unless you are using the pc for other work. It doesn't update every 15 seconds, it just runs a little slowly. For example, it'll say: Elapsed: 3:52:40 and linger on that for two seconds instead of one before saying Elapsed: 3:52:41 which results in the elapsed time being shorter than it actually is. Anyway, here is the CPU time with the Elapsed time in Properties: ![]() I'm not sure what to make of this, although there's a clear gap between the two times. ---------- I'm not sure which of two effects you describe: A) The elapsed time does not update every second but increases at normal speed overall. That can be caused by very high CPU load, permanent disc activity or network congestion when monitoring a remote client. Anything that can cause a delay effectively. The progress bar and remaining time updating in larger steps can be normal, that's application specific. Also see the post by Sgt.Joe. B) The elapsed time increases slower than real "wall clock" time. B1) Other CPU load does not allow the applications to run full speed. That's what Bryn Mawr describes. B2) It's a client setting. First check where your settings come from. The Manager's computing preferences will tell you at the top if you're using local preferences or web preferences. For local preferences look for "Usage limits" below. It's common practice to use 100% of CPU time and control the computing power by setting the percentage of CPUs to use. Else the applications would switch between on and off in quick succession, not running full time. For web preferences check your device profile at WCG for equivalent settings. There's pre-defined sets of preferences, the one you chose could limit the CPU time. I think it may be something like B2, although I'm not sure. Here's the data on my preferences: I've figured out that the pattern is something like this by the second: 00:00:01 00:00:02 00:00:02 00:00:03 00:00:03 [repeat] It'll first update by one second, then another time, then it'll freeze, then update, then freeze again, repeating. Since my percentage is at 60%, it seems like it's the source. Setting my percentage at 100% solves it. Thank you so much! This is solved now Best, |
||
|
|
floyd
Cruncher Joined: May 28, 2016 Post Count: 47 Status: Offline Project Badges:
|
![]() I'm not sure what to make of this, although there's a clear gap between the two times. By making a change here you will have switched from web-based preferences to local preferences. That doesn't matter in your case but you should keep it in mind. |
||
|
|
|