| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Support
Forum: Website Support
Thread: Bandwidth |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 36
|
|
| Author |
|
|
Sgt.Joe
Ace Cruncher USA Joined: Jul 4, 2006 Post Count: 7545 Status: Offline Project Badges: 
|
Btw, I do have a question for you (and possible other long time, multi-host crunchers): Have you ever had the case where one of your hosts turned out to be the wing man for another one of your's? Yes, I have that happen. It was some uncounted number of years ago and it has only happened once of which I am aware. I kept track of the work unit ID's in a spread sheet and noticed the duplicate. Cheers
Sgt. Joe
*Minnesota Crunchers* |
||
|
|
TPCBF
Master Cruncher USA Joined: Jan 2, 2011 Post Count: 1928 Status: Offline Project Badges:
|
Btw, I do have a question for you (and possible other long time, multi-host crunchers): Have you ever had the case where one of your hosts turned out to be the wing man for another one of your's? Yes, I have that happen. It was some uncounted number of years ago and it has only happened once of which I am aware. I kept track of the work unit ID's in a spread sheet and noticed the duplicate. Cheers I will take a look tonight if this is a bigger issue to fix (I don't think so, I just should need to make one change in my add/modify routine to fix this. Don't think that this will happen a lot, possibly uses like Marist College or similar with a really large number of hosts mght otherwise run into it. But as I don't know for how long this program might be used for, I rather make it right...  Ralf  |
||
|
|
adriverhoef
Master Cruncher The Netherlands Joined: Apr 3, 2009 Post Count: 2069 Status: Offline Project Badges: 
|
Btw, I do have a question for you (and possible other long time, multi-host crunchers): Have you ever had the case where one of your hosts turned out to be the wing man for another one of your's? That would mean that one member (i.e. same account) gets to crunch two tasks from the same workunit on two (assumed different) devices. I've dived into my logfiles that span over 2 million results and couldn't find any proof of that happening. Despite this outcome, I'm not saying that it is impossible.So far, I process WUs by processing the "workunitId", only to just realize that there could be the case where a WU shows up twice (or more), with different status, because two different devices could get the same WU, with the same "workunitId" but with a different "resultId." It is possible that the workunitId will be reused after resetting the counter before growing too large. For a fact, the resultId has been reset on 2018-03-03 before it could reach 2^31 (2147483648) to a number just below 2^27, as far as I can tell. The workunitId has also been reset, one year earlier: on 2017-05-19, before it could reach 2^31 it was reset to a number in the vicinity of 40,000,000 (the two lowest numbers that I found were 37600344 and 41688790). They must both be 32-bit numbers; signed, I'm afraid.EDIT: Strange as it may seem, the workunitId was again reset on 2021-11-01 from a number well below 900 million to a number in the vicinity of 2^24 (16,777,216); the lowest I saw was 18188151. Do I hear anybody saying 'Right before WCG moved to Krembil'?  [Edit 1 times, last edit by adriverhoef at Sep 20, 2022 11:41:42 PM] |
||
|
|
TPCBF
Master Cruncher USA Joined: Jan 2, 2011 Post Count: 1928 Status: Offline Project Badges:
|
I remember that this did happen a few years ago that one of my hosts was the wing man of another one of mine. Though I think the chances are relatively low. Could be that this "resetting of the workunitID" was what I remembered uplinger/knreed to mention back then.
----------------------------------------Doesn't look to me as if the numbers are in danger of overflowing even a signed 32bit integer, right now they are about a factor of 10 smaller than that. And I seriously hope that by the time the overflow could indeed happen, Krembil got their shit together and the stats are working on the web site again... Anyway, I had time to take a quick look inbetween my regular work today, and it was indeed only one line in the code that needed to be changed to switch from the WorkUnitID to ResultID. Ok, plus two spots where the WU iD was shown on the screen... Ralf  |
||
|
|
Sandman192
Cruncher United States Joined: Apr 26, 2007 Post Count: 20 Status: Offline Project Badges: 
|
Failing on downloading WUs. Seem uploading WU still working.
Went to main page of WCG site and the 3 pictures that come up for updates are loading really slow. Diel up all over again. |
||
|
|
cjslman
Master Cruncher Mexico Joined: Nov 23, 2004 Post Count: 2082 Status: Offline Project Badges: 
|
The download retry/pending issue seems to be a recurring problem. The first time I noticed it (back in August) I sorta panicked thinking it was something to do with my machine, but now I just think "meh" and hope that it will be gone by morning. On a side note, it is annoying.
----------------------------------------CJSL |
||
|
|
cjslman
Master Cruncher Mexico Joined: Nov 23, 2004 Post Count: 2082 Status: Offline Project Badges:
|
The download retry/pending issue seems to be a recurring problem. The first time I noticed it (back in August) I sorta panicked thinking it was something to do with my machine, but now I just think "meh" and hope that it will be gone by morning. On a side note, it is annoying.
----------------------------------------CJSL |
||
|
|
duanebong
Advanced Cruncher Singapore Joined: Apr 25, 2009 Post Count: 134 Status: Offline Project Badges:
|
I thought the download retry / pending issue was getting better with time. But in the last week it seems to be happening more often again. I often find my machines idling with no tasks as they failed to download WUs. And the more powerful the machine the more the idle time, as it crunches work too fast - which is a pity.
----------------------------------------When WCG first resumed I would make an effort to hit the retry. But it's not possible to keep babysitting all the machines several months down the road... especially when there are several spread over different physical locations. It is what it is..  |
||
|
|
BobbyB
Veteran Cruncher Canada Joined: Apr 25, 2020 Post Count: 598 Status: Offline Project Badges:
|
I solved the idle problem by adding TN-Grid to the mix. Yes that means less processing power for WCG or does it.
----------------------------------------I turned off TN-Grid once or twice (Ryzen 9 and 5) and eventually ended up with idle time. So as we can all see WCG still can't fill the queues. I'm actually at position112 worldwide at TN-Grid  All credits other than WCG are since Feb 2022. [Edit 2 times, last edit by BobbyB at Oct 4, 2022 8:43:55 PM] |
||
|
|
Sgt.Joe
Ace Cruncher USA Joined: Jul 4, 2006 Post Count: 7545 Status: Offline Project Badges:
|
I solved the idle problem by adding TN-Grid to the mix. Yes that means less processing power for WCG or does it. I turned off TN-Grid once or twice (Ryzen 9 and 5) and eventually ended up with idle time. So as we can all see WCG still can't fill the queues. I'm actually at position112 worldwide at TN-Grid All credits other than WCG are since Feb 2022. If it doesn't improve pretty soon, I will fire up TN-Grid and Denis again.They both appear to be worthwhile projects that deserve some time. I don't remember how big the TN-Grid work units were (a couple of hours I think) and the Denis workunits were a bit bigger than hour. Pretty reasonable I think. They both kept the queues filled so they appear to have plenty of work and the ability to get it out to the crunchers. Cheers
Sgt. Joe
*Minnesota Crunchers* |
||
|
|
|