| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
| World Community Grid Forums
|
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 45
|
|
| Author |
|
|
phillipspencer
Advanced Cruncher France Joined: Apr 9, 2015 Post Count: 71 Status: Offline Project Badges:
|
i'm ready to start crunching for WCG again, but one of two things need to happen first: 1) transient upload/download errors resolved, or 2) a significant increase in task deadlines. Unlike some lucky people, i don't have the time to babysit network issues, so if it doesn't work and BOINC keeps backing off for 1-2 days both up & down (which happened to me with an ARP tasks few months back), then i expect a longer deadline to counteract it. If longer deadlines are not possible, then apologies but i'll continue on other projects, and occasionally check back on WCG... You make some good points. In particular, I see no reason why deadlines should be restricted to six days for normal and three days for resend work units, especially given some of us only run WCG in the background, when our PCs are needed to be on (not 24 hours a day!) and we are not "nursemaiding" it through all the transient download errors. Cheers Phillip |
||
|
|
TPCBF
Master Cruncher USA Joined: Jan 2, 2011 Post Count: 2173 Status: Offline Project Badges:
|
In particular, I see no reason why deadlines should be restricted to six days for normal and three days for resend work units Well, I see a very good reason for not doing that.Rather than spend time to implement a temporary workaround, the time and effort should much rather be used to actually fix the download problems instead. That would be much more beneficial in the long run... Ralf PS: Despite the declaration in the email referred in this thread, WCG as a whole should not be considered "operational" yet... |
||
|
|
phillipspencer
Advanced Cruncher France Joined: Apr 9, 2015 Post Count: 71 Status: Offline Project Badges:
|
As someone else who never received the email (and it wasn't in my spam folder either!) I have to wonder about the parameter selection for sending it and / or data collection to verify since there have been multiple reports of non-receipt by crunchers like me who apparently meet the criteria of currently active user who has been returning test WU results over the past few months.
(I suppose, looking on the bright side, not receiving it meant I wasn't confused by it! ![]() |
||
|
|
TPCBF
Master Cruncher USA Joined: Jan 2, 2011 Post Count: 2173 Status: Offline Project Badges:
|
As someone else who never received the email (and it wasn't in my spam folder either!) I have to wonder about the parameter selection for sending it and / or data collection to verify since there have been multiple reports of non-receipt by crunchers like me who apparently meet the criteria of currently active user who has been returning test WU results over the past few months. I can't tell why you (and as others claim) haven't gotten that email (I think it has been posted in its entirety here in the forum by one of the recipients), my best guess is that you might have missed it as it could have ended up in your spam/junk folder.(I suppose, looking on the bright side, not receiving it meant I wasn't confused by it! ![]() For me, receiving this email felt like a slap in the face without the ability to return the favor, as it showed up after the download problems had already shown up again, then followed a couple of hours later by a nose dive of web site and forum... Ralf |
||
|
|
phillipspencer
Advanced Cruncher France Joined: Apr 9, 2015 Post Count: 71 Status: Offline Project Badges:
|
In particular, I see no reason why deadlines should be restricted to six days for normal and three days for resend work units Well, I see a very good reason for not doing that.Rather than spend time to implement a temporary workaround, the time and effort should much rather be used to actually fix the download problems instead. That would be much more beneficial in the long run... Ralf PS: Despite the declaration in the email referred in this thread, WCG as a whole should not be considered "operational" yet... Hi Ralf, I realise I was not clear. I was not thinking of it as a temporary fix. As you rightly say would be a distraction from fixing the underlying problems. I have been seeing the download problems exacerbate a "time-out" missed deadline for those of us who do not run our PCs every day, let alone 24x7. While SETI@home's deadlines were exceptionally long, I personally found Einstein@home's fortnight deadlines worked better than under a week for WCG. I suppose a counter-argument is that the current deadlines work and given recent experience of WCG anything that works should be left completely alone! Phillip [Edit 1 times, last edit by phillipspencer at Oct 7, 2022 7:30:07 PM] |
||
|
|
TPCBF
Master Cruncher USA Joined: Jan 2, 2011 Post Count: 2173 Status: Offline Project Badges:
|
IHi Ralf, I don't think that you can easily compare for example Einstein@Home to most WCG projects. I recall that in the description of some of those projects, when they first came out, it was mentioned that subsequent WUs might depend on previously returned work. Now if you extend that deadline, considering that there ALWAYS is some kind of chance that a host misses a deadline on long running projects like ARP1, and missed deadline would result in a resend (_2, _3, etc), which also would have to have an extended deadline per your argument. And that in turn could lead to a situation where no new WU batches can be send out because to create them, they would still be waiting for those delayed returns. I realise I was not clear. I was not thinking of it as a temporary fix. As you rightly say would be a distraction from fixing the underlying problems. I have been seeing the download problems exacerbate a "time-out" missed deadline for those of us who do not run our PCs every day, let alone 24x7. While SETI@home's deadlines were exceptionally long, I personally found Einstein@home's fortnight deadlines worked better than under a week for WCG. I suppose a counter-argument is that the current deadlines work and given recent experience of WCG anything that works should be left completely alone! Phillip Yes, it's frustrating, I had early on a whole set of ARP1 WUs that ended up on slower hosts, being delayed by the download issues and when returned where considered "too late" and thus ended up being discarded. There is also a case from Adrian (?)), where an original WU missed the deadline, someone got a resend, and got that discarded upon return because the original (delayed) WU had come in 2h earlier and validated. Tough nuts... Ralf |
||
|
|
Link64
Senior Cruncher Joined: Feb 19, 2021 Post Count: 206 Status: Offline Project Badges:
|
It's hilarious that some of y'all think posting "I HAVE MANY COMPUTERS SO YOU BETTER STRAIGHTEN UP OR I WILL TAKE MY CYCLES ELSEWHERE GOOD SIR" is making anything better in any way. It's just you, centering yourself, so you can feel righteously indignant. We're not all gonna stand up and clap for your brave defense of your hardware. Yeah, this project has problems. And when you post like this you become one of them, so congrats. Just go. Or don't. But if you're gonna stick around, stop with this kind of noise. I think you are misreading the intentions of these parties. Some of us wish to help with medical research by supplying our cycles to projects which we deem worthwhile. It is a bit frustrating when we want to help, have the ability to do so, and yet can't. It is the scientists and researchers who are being shortchanged. Perhaps some here need to find different way to express their "wish to help with medical research" in that case, than no one will "misread" it. I've seen that many years on SETI with all the server issues over there and never had the impression, that the main reason for all those rants was to help with science. Even when the project should shut down, instead of helping with fast and clean shutdown, those that used to complain most about their computers being idle, used modified clients to bypass server limits and bunker two months worth of work and crunched all of it just for credits, even though the WUs were already validated by additional resends. 0 value for science. Yes, the situation sucks, but like back than on SETI, it does not help to hammer on the servers which can't handle the load even without it, post here what hardware someone has bought or threaten to leave. Report issues and crunch other projects in the meantime, there are like two dozens other scientifically valuable BOINC projects out there (+ non-BOINC projects), also some with medical research. That member with servers with hundreds GB of RAM could for example run those new rosetta phyton tasks, which most standard PCs can't run. That research isn't far from medical. There were other projects suggested in several threads before the move. No CPU core needs to idle even for one second if someone wants to help the science. That was the whole idea behind BOINC, one platform for many projects, if one has issues or simply no work, crunch for an other one. You can even configure backup projects, so you don't need to check yourself too often if your devices have enough work. I got the email too, gave it a try, seen stuck downloads, back to Einstein. I will keep checking the forums for news and will come back, when I can actually help instead of making it even worse by creating additional load on the already overloaded servers. ![]() [Edit 2 times, last edit by Link64 at Oct 8, 2022 6:32:10 PM] |
||
|
|
Sgt.Joe
Ace Cruncher USA Joined: Jul 4, 2006 Post Count: 7844 Status: Recently Active Project Badges:
|
those that used to complain most about their computers being idle, used modified clients to bypass server limits and bunker two months worth of work and crunched all of it just for credits, even though the WUs were already validated by additional resends. 0 value for science. I agree with you. All the credits/badges/points etc. are all fun window dressing (which I enjoy), but they are not the point of the crunching. The point is to provide a resource to the researchers/scientists which they would have difficulty getting another way. If I have this resource, I will put it to use either here or elsewhere if here is not available and if I see merit in the project. When you want to help, but can't, it is frustrating. Cheers
Sgt. Joe
*Minnesota Crunchers* |
||
|
|
Speedy51
Veteran Cruncher New Zealand Joined: Nov 4, 2005 Post Count: 1326 Status: Offline Project Badges:
|
Could you provide a link to the latest ? Thank you Cheers https://www.worldcommunitygrid.org/forums/wcg/viewthread_thread,44362 ![]() [Edit 1 times, last edit by Speedy51 at Oct 9, 2022 2:21:27 AM] |
||
|
|
nyanthiss
Cruncher Joined: Nov 23, 2012 Post Count: 15 Status: Offline Project Badges:
|
those that used to complain most about their computers being idle, used modified clients to bypass server limits and bunker two months worth of work and crunched all of it just for credits, That's a nice theory, and may even be true in case of SETI, but it's completely wrong for WCG. Nobody crunches biology projects on BOINC solely for credits, because pretty much all of them have an abysmal credits/work ratio. You can get 10x more credits for the same CPU time by crunching astronomy or math projects.
Intel Xeon E3-1231 v3
AMD A10 7800 AMD Ryzen 5 3500U AMD Ryzen 1700X AMD Ryzen 5900X 2x RaspberryPi, 1x Odroid |
||
|
|
|