| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: Help Conquer Cancer
Thread: Something is nuts with To Completion on my tasks |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 7
|
|
| Author |
|
|
Richard Mitnick
Veteran Cruncher USA Joined: Feb 28, 2007 Post Count: 583 Status: Offline Project Badges: 
|
On all of my projects except HCC from WCG, as the CPU time runs up, the To completion time runs down, which is what I would expect.
----------------------------------------But on HCC,currently running tasks on three computers, on devices named Laptop and Desktop, the To completion time is actually growing, on LaptopII it is erratically going up and then down: 1:01:01 to 1:01:02 to 1:00:47. What's up with this? My main reason for looking at this is that several tasks on Laptop have failed and quit, just wasting everybody's time. On Laptop and Desktop, I have temporarily detached from all but WCG projects to try and find out just what the heck is going on. >>RSM |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
It just means the original time estimate was wrong.
Work unit sizes vary, as you know. Sometimes BOINC simply doesn't have enough information for a good time estimate. However, as the work unit progresses, BOINC recalculates based on the time it took to get to the point you're at. Work units that failed can also give BOINC bad data to base its time estimates on. After a few work unit failures, it may take a while for BOINC to readjust. |
||
|
|
Richard Mitnick
Veteran Cruncher USA Joined: Feb 28, 2007 Post Count: 583 Status: Offline Project Badges:
|
Thanks, that is reassuring news.
----------------------------------------I need to watch it because one machine has errored several times, wasting time for me and for the project. >>RSM |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hello, did you check http://www.worldcommunitygrid.org/forums/wcg/viewthread?thread=17407
Is yours a separate issue from this? Can't say why your tasks keep on failing, though. |
||
|
|
mattyfd99
Cruncher Joined: Nov 23, 2005 Post Count: 7 Status: Offline Project Badges: 
|
I have this problem too and I brought this up before. If you let the problems run without viewing them on the desktop it will run fine. If you decide to watch them with graphics or the task manager you will encounter this problem. Try just checking the progress periodically and you will see that it is probably working OK.
|
||
|
|
Richard Mitnick
Veteran Cruncher USA Joined: Feb 28, 2007 Post Count: 583 Status: Offline Project Badges:
|
I looked at that thread. My problem seems to be worse.
----------------------------------------On 2 PIII machines, at about the same time, tasks for HCC quit for "computation errors." This is just wasting time for the project and for me. There does not seem to be a problem with the deadline dates, and I detached from the two BOINC projects on these machines to give HCC a good test, and the machines failed. I am running HCC on two Core 2 Duo Vista machines with 2 gig and 2.5 gig CPU's and 2 gigs of DDR2 DRAM, all with no problems. So, I think something about the PIII's, with CPU's around 1 gig and 512 megs of DRAM is just not enough power and capacity for HCC. I will be happy to run other WCG projects on these machines if it makes any sense, so that they lift some burden off others who can run HCC. I will entertain any ideas here. >>RSM |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hello mitrichr,
512 MB ought to be enough as long as you allow 100% of everything, without any limits imposed on BOINC. But theory does not always mirror reality, so you had better switch to FAAH, DDDT and/or HPF2 on those Pentium III machines. Sorry for causing you trouble at the start of this new year. Lawrence |
||
|
|
|