| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Active Research
Forum: Africa Rainfall Project
Thread: Work Available |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 3596
|
|
| Author |
|
|
Macroman
Advanced Cruncher Joined: Jun 4, 2005 Post Count: 112 Status: Offline Project Badges: 
|
Since 2020-08-03, I notice a project slow down. It is becoming more and more difficult for my machines to get ARP1 WUs. Cheers, Yves Not sure if there is some kind of reason, a cyclicalness to these slowdowns, but received a rush job 48 hours ago which finished but no further backfill to my standard profile requesting to have 2. No rush i suppose. From what I've seen I think that there is usually a pause when we're running out of an iteration (001, 002 etc) before starting the next. I'd surmise that there is only room for so many work units to be simultaneously loaded so we may need to wait for stragglers from the previous iteration to complete before the new iteration can be loaded. I have seen almost no overlap between the iterations, (ie now that 019 is out there you don't see any 018's any more). |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Not sure if there is some kind of reason, a cyclicalness to these slowdowns, but received a rush job 48 hours ago which finished but no further backfill to my standard profile requesting to have 2. No rush i suppose. I am interested to know by "rush job" do you mean a job that had not been returned before it's deadline? The job for whatever reason processed in high priority mode although my queue is only 1 day deep. BOINC gets nervous when a task is having a remaining deadline that is less than double of the estimated runtime or less than 24 hours (whichever is greater), so when the previous ARP1 job finished it had already eaten 24 hours of that deadline, leaving 2.5 days. That triggered the rush processing aka HP from the start. (app_config only allows execution of 1). [Edit 1 times, last edit by Former Member at Aug 7, 2020 5:57:29 AM] |
||
|
|
Mike.Gibson
Ace Cruncher England Joined: Aug 23, 2007 Post Count: 12594 Status: Offline Project Badges: 
|
My very short queue of arp has dried up but we only started on 019 a couple of days ago, so that shouldn't be it. More likely the server needs another kick where it has worked before, as has happened several times on a Thursday or Friday.
Let's hope that kick will be forthcoming soon. |
||
|
|
sam6861
Advanced Cruncher Joined: Mar 31, 2020 Post Count: 107 Status: Offline Project Badges:
|
The work is there, one of my ARP work is pending validation. But, there is "waiting to be sent" for more then 12 hours.
https://www.worldcommunitygrid.org/ms/device/...s.do?workunitId=209495086 |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
That's a repair job, not new work. That hung in limbo on Friday afternoon does not bode well for outlook towards a resume this weekend.
|
||
|
|
Mike.Gibson
Ace Cruncher England Joined: Aug 23, 2007 Post Count: 12594 Status: Offline Project Badges:
|
That does indicate that the queue has stuck.
Mike |
||
|
|
nei1
Cruncher United States Joined: Jan 30, 2014 Post Count: 38 Status: Offline Project Badges: 
|
> "waiting to be sent"
----------------------------------------Stuck! I never saw that before. There should be a distributed computing project to help predict when Africa Rainfall Project work units will become available. Trying all my superstitious tricks to get an ARP wu, but they won't give. [Edit 1 times, last edit by nei1 at Aug 7, 2020 7:03:41 PM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
> "waiting to be sent" Stuck! I never saw that before. There should be a distributed computing project to help predict when Africa Rainfall Project work units will become available. Trying all my superstitious tricks to get an ARP wu, but they won't give. For all we know the 'waiting to be sent' is in holding pattern just in case the No Reply still shows up. We've had cases too where the WTBS were withdrawn because it crossed paths with a late return, so I'm in the 'jury is out' camp ATM. |
||
|
|
Macroman
Advanced Cruncher Joined: Jun 4, 2005 Post Count: 112 Status: Offline Project Badges:
|
OPN1 has really picked up with ARP stopped. Last 5 days in a row all over a million WU.
----------------------------------------[Edit 1 times, last edit by Macroman at Aug 8, 2020 4:25:20 AM] |
||
|
|
Mike.Gibson
Ace Cruncher England Joined: Aug 23, 2007 Post Count: 12594 Status: Offline Project Badges:
|
Returns are down from 8000 per day to under 700 this morning.
Anyone still holding arp units should reduce their cache for the future because their machines will not be classified as 'reliable' if they do not regularly return units within 3.5 days. Being classified as 'reliable' means that those machines will get short-dated re-sends. Mike |
||
|
|
|