| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: The Clean Energy Project - Phase 2 Forum
Thread: All Germanium-containing workunits failing |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 52
|
|
| Author |
|
|
Yarensc
Advanced Cruncher USA Joined: Sep 24, 2011 Post Count: 136 Status: Offline Project Badges: 
|
So basically we're stuck in an endless loop of validating each other until those Ge WU are pulled. It's an endless loop of the host not being marked as "reliable" and not needing a wingman, but its not endless for each workunit, it just means each workunit gets done twice. So yes, it is wasting almost double the time it otherwise would, but we're still going through them and producing valid results (except the Ge and Se ones of course) |
||
|
|
Gatchaman
Cruncher Joined: Feb 29, 2012 Post Count: 49 Status: Offline |
So basically we're stuck in an endless loop of validating each other until those Ge WU are pulled. So yes, it is wasting almost double the time it otherwise would, but we're still going through them and producing valid results (except the Ge and Se ones of course) Imagine you had a tough WU that took 15 hours to complete on a fast-ish machine. It uploads. It's all good. Next work unit is a Ge and it goes and errors out. Now somebody is going to have to re-crunch the first WU for no reason. And if the wingman has a slow cpu then that cruncher is forced to waste time and electricity on a perfectly good WU. Whether it was intentional or not to release these Ge WU into the wild knowing they'd cause this much double, triple+ crunching of the same WU, I just don't know........, but the CEP team should pull the Ge WU's and send them to beta testers only for now.  "Sadly this project is turning into nonscience......" |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
makes you think what kind of CLEAN Energy Project are they running?!  It does say clean energy, nothing about manufacturing, installation, etc. |
||
|
|
Gatchaman
Cruncher Joined: Feb 29, 2012 Post Count: 49 Status: Offline |
I've been thinking of a work around. If the science bods are ignoring this issue and don't want to remove the known error prone Ge WU then how about we just delete them for our buffers.
----------------------------------------I'm using boinctasks to check the state of my WU history and it's not a pretty sight. So why not just download a large batch say 7 days worth of work and just delete the Ge WU from the batch, and while you're there delete every validation work unit. Noobs, you can tell what type of WU it is because the a validation WU ends with any number other than 0, zero. Also set the Boinc to only communicate once a day at a time when you are near your crunching box to see what WU are sent to you. Again delete all Ge and validation jobs. Do this until the science bods at CEP HQ decide we are worthy of a response in relation to this matter. And one last thing imagine all the resource this pointless CEP2 validating is tying up. Resources that could be put towards other projects like the new Help Stop TB project, FightAIDS@Home - Phase 2, Outsmart Ebola Together, Uncovering Genome Mysteries, Mapping Cancer Markers. Whoever gave the go ahead for the Ge WU, in its current state, to be set free should be ashamed of themselves. We, the crunchers, are all here to help. Nothing more, nothing less. We invest our time and our money to help you to help us. At the end of the day when this project comes to an end what are we going to get out of it in the short term anyway. Not a single thing. Not a book token. Not a dime. Nothing. Zero. Maybe a "It was nice knowing you. Now move along, there's nothing to crunch here". Whereas members of the lab will get job opportunities and funding for more research based on the theories and practices proven on our machines. And so they should. Oh I forgot, we'll all get "I completed CEP" badges. Sigh. Just pull the Ge WU and let us get back to real work already! "Sadly this project is turning into nonscience......" |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I haven't downloaded any CEP2 WUs for the past week. There are plenty of other projects to work on.
|
|||
|
||||
|
Gatchaman
Cruncher Joined: Feb 29, 2012 Post Count: 49 Status: Offline |
Turned 3 of my crunchers on and got 8 WU. 7 validations and 1 Ge. Aborted the 7 after an hour. Powered off the crunchers.
----------------------------------------"Sadly this project is turning into nonscience......" |
||
|
|
Yarensc
Advanced Cruncher USA Joined: Sep 24, 2011 Post Count: 136 Status: Offline Project Badges:
|
...[J]ust delete the Ge WU from the batch, and while you're there delete every validation work unit. Since nobody has posted an example of a WU /not/ failing when it has Ge in it, that first part sounds like a good idea. But since so many machines are going to be marked 'unreliable' and put on quorum 1 with the frequent fails, I don't see the point in deleting the other WU's that are validation ones. But I'm starting to agree that it might be better to focus on other sciences until this is fixed so that half the time isn't wasted on validating. Side question: does a user abort status cause a machine to be marked unreliable? |
||
|
|
Gatchaman
Cruncher Joined: Feb 29, 2012 Post Count: 49 Status: Offline |
Side question: does a user abort status cause a machine to be marked unreliable? Was thinking the same thing. Personally I don't really care what my status is. Just as long as one is not wasting time, energy and money on this validating loop rubbish. When CEP decide to pull their finger out I'll very happily go through validating my machines to get back to trusted status. Sadly this project is turning into nonscience...... "Sadly this project is turning into nonscience......" [Edit 1 times, last edit by Gatchaman at Mar 27, 2016 4:38:08 PM] |
||
|
|
svincent
Advanced Cruncher Joined: Jan 3, 2009 Post Count: 53 Status: Offline Project Badges: 
|
While the current situation is irritating, I don't think the extra validation is too much of a waste of time as it helps ensure the integrity of the results. Many projects, such as Einstein@home and apparently the new Help Stop TB project, require two matching results for a valid result.
I agree though that the lack of communication from the devs is pretty remiss: not even a one sentence post to say they're aware of it and are looking into it. It seems that communication for many of these distributed computer projects is great at the beginning but becomes ever more infrequent as the project progresses. Sigh. |
||
|
|
Gatchaman
Cruncher Joined: Feb 29, 2012 Post Count: 49 Status: Offline |
These were the last batch of WU I got from CEP. A couple slipped by to completion(error) but most were aborted after a few mins staring at the last digit of the WU.
----------------------------------------E236402_ 871_ S.324.C35H27N1S2Se1Si1.HBJHHMVDOHJLMD-UHFFFAOYSA-N.5_ s1_ 14_ 4-- E236429_ 647_ S.316.C36H23N3O5S1.DHNHBIAASJDTRF-UHFFFAOYSA-N.16_ s1_ 14_ 2-- E236503_ 865_ S.342.C39H29N1O1S1Se1Si1.JZMVZQCYDOWTBV-UHFFFAOYSA-N.3_ s1_ 14_ 0-- E236502_ 277_ S.340.C36H29N1O3S1Se1Si1.KSMWBVBVDBCMPT-UHFFFAOYSA-N.4_ s1_ 14_ 3-- E236503_ 748_ S.324.C26H12N6O2S4Se1.YAGGIVCXALCMHJ-UHFFFAOYSA-N.16_ s1_ 14_ 2-- E236503_ 754_ S.324.C26H12N6O2S4Se1.YAGGIVCXALCMHJ-UHFFFAOYSA-N.22_ s1_ 14_ 2-- E236503_ 621_ S.290.C21F4H7N7O1S4.XDYIUVKIMJBQEP-UHFFFAOYSA-N.14_ s1_ 14_ 2-- E236503_ 644_ S.310.C29H13N7O1S2Se1.HMGFWKVRWNHPOX-UHFFFAOYSA-N.12_ s1_ 14_ 2-- E236411_ 845_ S.330.C32H21N3S3Se1Si1.HEDMHURIILNQET-UHFFFAOYSA-N.14_ s1_ 14_ 3-- E236411_ 452_ S.292.C27H12N4O3S2Se1.NVMFYVWLJLYFTN-UHFFFAOYSA-N.7_ s1_ 14_ 1-- E236501_ 875_ S.330.C34H20N6S4.LNQZNROEAKKBIR-UHFFFAOYSA-N.18_ s1_ 14_ 1-- E236501_ 292_ S.328.C31H19N1S3Se2.UYGKUAVCVMZNTB-UHFFFAOYSA-N.19_ s1_ 14_ 3-- E236495_ 788_ S.328.C34H28N2S3Se1.HMSWSXUXUMVGCV-UHFFFAOYSA-N.5_ s1_ 14_ 3 E236403_ 950_ S.316.C34H17N1O1S5.ZKUSUWHHQZTNHJ-UHFFFAOYSA-N.10_ s1_ 14_ 2 E236494_ 357_ S.324.C35H27N1S2Se1Si1.NQGYZGMMWHICPI-UHFFFAOYSA-N.16_ s1_ 14_ 3-- Even if they are legit, something's not right with CEP these days. And I've just bought my first ever rack mount servers, which I was going to slip under my desk at work ;-) , to replace my box shape crunching boxes as I'm losing my hack space this summer. As for communication with the volunteers who are doing the vast amount of donkey work crunching for CEP well it doesn't take 30 seconds to type something that could change the current perception of the CEP. Nice thing about the internet is that one can be virtually anywhere on the planet and one could still send a message, however short, to millions of people. If I was running WCG I'd be having words with the CEP team right about now....but I'm not, so I can't. CEP should be setting an example to other labs on how to run a project, and not, how not to run a project. WCG is always trying to get people to join up, but whats the point when this is how projects treat volunteers who spend their time, energy and money to crunch for them. For me it was one of the hardest things to do but I had to un-tick CEP from my list of things to crunch  Maybe if enough of us un-tick CEP they'll finally talk to us. STRIKE!  If CEP can't be bothered then neither can I. "Sadly this project is turning into nonscience......" [Edit 1 times, last edit by Gatchaman at Mar 28, 2016 11:29:44 AM] |
|||
|
||||
|
|