Questions and Answers :
Windows :
Esimated time increasing......
Message board moderation
Author | Message |
---|---|
Ashish Nain Send message Joined: 7 Feb 13 Posts: 3 Credit: 6,147 RAC: 0 |
my remaining time is going up instead of going time and percentage is also stuck at same point. when i suspend and resume a task it goes down for few seconds then again starts going up. i tried leaving it as it is for some time but it doesn't seem to work... any idea what the problem might be???? |
Zalster Send message Joined: 27 May 99 Posts: 5517 Credit: 528,817,460 RAC: 242 |
Is this a new system you are using? Your remaining time is an estimate of how long it will take for a computer to finish. Since each computer is different (chips, ram, Ghz, graphic cards, etc) there is no way for seti to know exactly how long a work unit will take to complete. So you get an estimate. The times tend to get longer initially. After Seti has run some 12 work units (?) I think, it has a better idea of how long it will take for your computer to finish a work unit. After this, the estimated times should be much closer the actual time. How long are you allowing the computer to run these work units before you abort them? You should let them run for a few days so you can get an idea of if it's truly stuck. I also see you are running 2 project on each of your computers? Are you running them both at the same time or do you run 1 project then switch to the other or are you allowing bonic to make the decision on when to switch between the projects? |
Ashish Nain Send message Joined: 7 Feb 13 Posts: 3 Credit: 6,147 RAC: 0 |
i think i have done more than 12 WU. and it is the same old system on which i have been running boinc since starting. i am letting Boinc decide which task to run but both projects are running at same time. this is happening from past 4 days.. i abort a task after it seems to be stuck 3-4 hours and sometimes even more. |
Zalster Send message Joined: 27 May 99 Posts: 5517 Credit: 528,817,460 RAC: 242 |
I would suspend one project and see what the other does. We know some projects like to "take command" of the GPU and CPUs and will not share with other projects(even thou they are supposed to, this can be overridden by the programers). That might be what is happening, why your times are increasing because they are not getting an equal share time of the resources. I only run 1 project at a time and wait till I finish all work units before I start the other so that I can make sure they get done. If after trying that, it still seems like the times are increasing or stalling let us know. Hopefully another user will have some more ideas after that if it doesn't correct the problem. |
John McLeod VII Send message Joined: 15 Jul 99 Posts: 24806 Credit: 790,712 RAC: 0 |
i think i have done more than 12 WU. and it is the same old system on which i have been running boinc since starting. The problem is that the initial estimate from the projects can be WAY off. The current percent complete can also be way off. Since there are only four things that go into the calculation of the estimate of time left to completion, and 2 of them are subject to errors, the calculation can produce estimates that fluctuate widely. The tasks are probably not actually stuck, and if you keep aborting them, the history of work completion times will never get updated. Mostly just leave things alone. BOINC WIKI |
Ashish Nain Send message Joined: 7 Feb 13 Posts: 3 Credit: 6,147 RAC: 0 |
i allowed boinc to run all day and it completed two tasks that were stuck. the problem still exists when new task is started but i guess i will have to allow the tasks to run for some time for the problem to be solved. |
rob smith Send message Joined: 7 Mar 03 Posts: 22189 Credit: 416,307,556 RAC: 380 |
You need to get about twenty completed and validated tasks before the estimate starts to settle down properly. Bob Smith Member of Seti PIPPS (Pluto is a Planet Protest Society) Somewhere in the (un)known Universe? |
©2024 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.