| 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: 3203
|
|
| Author |
|
|
TPCBF
Master Cruncher USA Joined: Jan 2, 2011 Post Count: 1950 Status: Offline Project Badges: 
|
It should get better as queues are filled. I find it better to make a shorter queue now and lengthen it slowly. much better to be downloading pieces of one WU at a time, then 5 at a time. That is the problem! By default, it is one (1) WU for one host. And that was chosen way back when for a reason. But too many people get greedy and thus creating more problem for everyone...It will be interesting to see if they keep the 7 day deadline. I don't see a real reason why not. Just the first WU that I got, on an older, slower i5 but that doesn't do much else but running WCG these days, had a 2 day dead line and that could get tight...Ralf  |
||
|
|
Mike.Gibson
Ace Cruncher England Joined: Aug 23, 2007 Post Count: 12359 Status: Offline Project Badges: 
|
My one came through but MCMs were stuck.
|
||
|
|
imakuni
Advanced Cruncher Joined: Jun 11, 2009 Post Count: 103 Status: Offline Project Badges: 
|
It should get better as queues are filled. I find it better to make a shorter queue now and lengthen it slowly. much better to be downloading pieces of one WU at a time, then 5 at a time. That is the problem! By default, it is one (1) WU for one host. And that was chosen way back when for a reason. But too many people get greedy and thus creating more problem for everyone...Ralf Quite the opposite: "greedy" users are actually HELPING the situation, not making it worse. A single person downloading 16 tasks is actually only pulling 2 download requests at a time, assuming they didn't mess with max_file_xfers in cc_config. Now yeah, sure, that one person is taking a bunch of tasks, but in terms of download requests, they're making 2 connections at a time. Compare those same 16 tasks being sent to one host each. That means 32 instances of boinc pinging the server to download stuff. The strain is much, much greater. On a side note, if a single user getting multiple tasks is evil, then I'm sorry, it's the server's job to say "no" and only give them one single task a time. It's not the user's fault that the project's scheduler is assigning them work: if you can complete a WU in the deadline given, no one can complain.  Want to have an image of yourself like this on? Check this thread: https://secure.worldcommunitygrid.org/forums/wcg/viewthread_thread,29840 |
||
|
|
TPCBF
Master Cruncher USA Joined: Jan 2, 2011 Post Count: 1950 Status: Offline Project Badges:
|
It should get better as queues are filled. I find it better to make a shorter queue now and lengthen it slowly. much better to be downloading pieces of one WU at a time, then 5 at a time. That is the problem! By default, it is one (1) WU for one host. And that was chosen way back when for a reason. But too many people get greedy and thus creating more problem for everyone...Ralf Quite the opposite: "greedy" users are actually HELPING the situation, not making it worse. A single person downloading 16 tasks is actually only pulling 2 download requests at a time, assuming they didn't mess with max_file_xfers in cc_config. Now yeah, sure, that one person is taking a bunch of tasks, but in terms of download requests, they're making 2 connections at a time. Compare those same 16 tasks being sent to one host each. That means 32 instances of boinc pinging the server to download stuff. The strain is much, much greater. On a side note, if a single user getting multiple tasks is evil, then I'm sorry, it's the server's job to say "no" and only give them one single task a time. It's not the user's fault that the project's scheduler is assigning them work: if you can complete a WU in the deadline given, no one can complain. This all has been proven time and time again, and was even worse when at the same time OPNG was available. But "some people" are just not understanding that they contribute to the overall system... Ralf |
||
|
|
Iceberg007
Cruncher Joined: Nov 12, 2017 Post Count: 27 Status: Offline |
Horray!!!
---------------------------------------- |
||
|
|
spRocket
Senior Cruncher Joined: Mar 25, 2020 Post Count: 274 Status: Offline Project Badges: 
|
Great to see something other than MCM in my queue for a change! Looks like the logjam has finally cleared as well. Hopefully it can stay cleared, but we'll see.
 Checking my results, there's even a validated unit among the few that I've managed to return. |
||
|
|
Unixchick
Veteran Cruncher Joined: Apr 16, 2020 Post Count: 948 Status: Recently Active Project Badges: 
|
I've managed to return 2 completed ARP WUs !! The upload went smoothly. I'm getting the dreaded no tasks are available message now. UG!
I predict a flood of resends on Nov 6 as people may have made their queues too long. |
||
|
|
Mike.Gibson
Ace Cruncher England Joined: Aug 23, 2007 Post Count: 12359 Status: Offline Project Badges:
|
If they can't increase their download capacity, why don't they release ARP units more slowly?
|
||
|
|
Mike.Gibson
Ace Cruncher England Joined: Aug 23, 2007 Post Count: 12359 Status: Offline Project Badges:
|
My Sunday reports will be back on 10 November.
|
||
|
|
bfmorse
Senior Cruncher US Joined: Jul 26, 2009 Post Count: 296 Status: Offline Project Badges: 
|
Unixchick
The flood of resends has already happened: MCM -2's and MCM -3's and the ARP WU's were still being sent. Seems to me, excessively long queues only CONTRIBUTE to the overall problem, not make it hidden (other than for the volunteer that HAS the excessively loooooong Queue and never really sees a loss of WU's when there is a technical problem at WCG with new WU's). I have my own ideas as to how to handle them, but my ideas carry no weight nor to they get responded to. Happy Crunching!  |
||
|
|
|