| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Support
Forum: Suggestions / Feedback
Thread: Work Units returned after the Deadline |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 9
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hi,
I have 3 computers processing work units which has normally been ok but with being sick over the last week some work units have got behind and been returned after the deadline. They have not been days late but more like in terms of hours. In most cases they were still accepted and marked valid but for the Human Proteome Folding - Phase 2 project I now have 2 work units marked as too late and I know there were only a few hours late. While it is fine for each project to define rules independently I think these 13 hours cpu time are now effectively wasted and could have been spent doing something useful. So if I now see a Human Proteome work unit is not going to be completed before the deadline then I think it better to kick it out and use processing time for something else. The work units are named, qh678_00046, qh711_00170 Regards, |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hope you're well again.
That's perfectly fine... to abort tasks you think will not make deadline. This is because immediately upon the deadline is passed, a new copy is send out as your copy would then be marked "No Reply". Normally if a task has *not* been started before the deadline and another computer has done it quickly, the server sends a message to your computer to automatically abort the redundant copy. If your computer has though started the task, the task will not be aborted. That's up to the user to decide as then a message appears in the client log to say it could be too late and time is wasted [no credit]. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hi, ....not quite sure why you being sick causes them to be returned late.......are you working them by hand? I have 3 computers processing work units which has normally been ok but with being sick over the last week some work units have got behind and been returned after the deadline.  |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hi Scribe,
----------------------------------------Well one of the PC's is my work PC, You are expected to turn off before you leave for the day. Hi, Rob, Yes you are correct but in the case of this Human Protoeme project 19 copies are sent out and 15 are expected to be returned. But it is not so tragic, I am still learning here. :) [Edit 1 times, last edit by Former Member at Oct 10, 2012 6:37:15 PM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
....aaaah, now I understand
 |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hello,
I admit, that I have not read this thread in complete, so my post might sound a bit "out of place" but it maches with the topic "after deadline". I was crunching for a project, which has too high system requirements compared with my PC. If I had gone on, I would have ended sending only units "too late". Ok, this is my problem, but what about people following the grids "install and forget" strategy and having too weak PCs? On their PC WCG will run on and on sending only worthless results. We can discuss this problem, but in the end it is WCG and only WCG who can analyse, whether a certain machine sends back masses of outdated results. As a solution to the problem of too weak machines, I can propose, that in the preferences there could be a button"Inform me if my machine returns more then x outdated results in a sequence" Mabe in a next version of BOINC? Thank you. Bye Kafejka |
||
|
|
Falconet
Master Cruncher Portugal Joined: Mar 9, 2009 Post Count: 3315 Status: Offline Project Badges: 
|
Hello, I admit, that I have not read this thread in complete, so my post might sound a bit "out of place" but it maches with the topic "after deadline". I was crunching for a project, which has too high system requirements compared with my PC. If I had gone on, I would have ended sending only units "too late". Ok, this is my problem, but what about people following the grids "install and forget" strategy and having too weak PCs? On their PC WCG will run on and on sending only worthless results. We can discuss this problem, but in the end it is WCG and only WCG who can analyse, whether a certain machine sends back masses of outdated results. As a solution to the problem of too weak machines, I can propose, that in the preferences there could be a button"Inform me if my machine returns more then x outdated results in a sequence" Mabe in a next version of BOINC? Thank you. Bye Kafejka Was the project CEP2?  - 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 |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hello Falconet,
it was Say No to Schistosoma Greetings Kafejka |
||
|
|
Falconet
Master Cruncher Portugal Joined: Mar 9, 2009 Post Count: 3315 Status: Offline Project Badges:
|
They usually aren't very long.
----------------------------------------Try doing HCC CPU. They finish fairly fast. What's you CPU by the way? - 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 [Edit 1 times, last edit by Falconet at Dec 29, 2012 4:12:19 PM] |
||
|
|
|