Message boards :
News :
Separation Application Shutting Down on Tuesday, Jun 20th
Message board moderation
Previous · 1 . . . 5 · 6 · 7 · 8 · 9 · 10 · 11 . . . 16 · Next
Author | Message |
---|---|
Send message Joined: 5 Jul 11 Posts: 990 Credit: 376,143,149 RAC: 0 |
At the point I knew there were enough results back for the science, I would have turned the production of new tasks off immediately and without warning, and issued a cancel to all tasks users had in progress. All that would have happened is tasks clients had started running would not get cancelled (stupid Boinc policy) and would have continued to the end without being useful, but that's not many. Everyone's queue would have just been deleted.Remember when Seti did that too and no one knew about for days, they gave credit even though they didn't do anything but trash the results as they were all resends of existing completed tasksWell, at least they have let everyone report the tasks back. If they were smart, they would have significantly reduced the deadlines for those additional resends and used server abort for tasks, that were not needed anymore. That way they could have shut down few weeks earlier. Users are used to and expect things to stop and start without warning, which is why most of us do more than one project. We are basically freelance workers doing work for many universities whenever they have stuff to do. To get work when nothing needs doing is silly. The above was double spaced between sentences, I apologise for the forum software ruining my post. |
Send message Joined: 5 Jul 11 Posts: 990 Credit: 376,143,149 RAC: 0 |
Why did we not just stop releasing tasks first and THEN shut down the project, so a whole bunch of work doesn't get stuck in pending validation?This project is for science, not points. Why should a scientist mess around to please your desires for points? The above was double spaced between sentences, I apologise for the forum software ruining my post. |
Send message Joined: 5 Jul 11 Posts: 990 Credit: 376,143,149 RAC: 0 |
At first i have read buster as booster,I fail to see what Buster has to do with this: The above was double spaced between sentences, I apologise for the forum software ruining my post. |
Send message Joined: 8 May 09 Posts: 3339 Credit: 524,010,781 RAC: 0 |
Remember when Seti did that too and no one knew about for days, they gave credit even though they didn't do anything but trash the results as they were all resends of existing completed tasks I think he was already gone from Seti when that happened but I could be wrong, the Admins said they knew new tasks were on the way and they didn't want to lose crunchers with a pause, they lost more crunchers after people found out about what they did. |
Send message Joined: 8 May 09 Posts: 3339 Credit: 524,010,781 RAC: 0 |
At the point I knew there were enough results back for the science, I would have turned the production of new tasks off immediately and without warning, and issued a cancel to all tasks users had in progress. All that would have happened is tasks clients had started running would not get cancelled (stupid Boinc policy) and would have continued to the end without being useful, but that's not many. Everyone's queue would have just been deleted.Remember when Seti did that too and no one knew about for days, they gave credit even though they didn't do anything but trash the results as they were all resends of existing completed tasksWell, at least they have let everyone report the tasks back. If they were smart, they would have significantly reduced the deadlines for those additional resends and used server abort for tasks, that were not needed anymore. That way they could have shut down few weeks earlier. Like I said before Tom got busy with his dissertation and forgot, I think getting your PHD is more important than a few useless credits and will forgive, but not forget, his transgression. |
Send message Joined: 5 Jul 11 Posts: 990 Credit: 376,143,149 RAC: 0 |
I think he was already gone from Seti when that happened but I could be wrong, the Admins said they knew new tasks were on the way and they didn't want to lose crunchers with a pause, they lost more crunchers after people found out about what they did.They made a stupid decision. Crunchers don't go away with a pause, just look at Climate Prediction (which has just produced a load of Windows tasks by the way, although I think they're all taken now, and they also tend to crash on older machines, only my Ryzen 9 is doing them properly) The above was double spaced between sentences, I apologise for the forum software ruining my post. |
Send message Joined: 5 Jul 11 Posts: 990 Credit: 376,143,149 RAC: 0 |
Like I said before Tom got busy with his dissertation and forgot, I think getting your PHD is more important than a few useless credits and will forgive, but not forget, his transgression.Very few people will get upset about it, there's just a few moaners in here, who cares? I just make my machines do as much science as possible, while trying to short out the National Grid. P.S. apparently Kate was too rude :-) Any ideas for my next one? The above was double spaced between sentences, I apologise for the forum software ruining my post. |
Send message Joined: 8 May 09 Posts: 3339 Credit: 524,010,781 RAC: 0 |
I think he was already gone from Seti when that happened but I could be wrong, the Admins said they knew new tasks were on the way and they didn't want to lose crunchers with a pause, they lost more crunchers after people found out about what they did. They admitted that, in time, but the damage was already done. As for the Climate Prediction tasks my Ryzen 7 is doing them just fine is one of my AMD Threadripper 1920X's, my 5950X but all my other pc's are all having problems with them. |
Send message Joined: 5 Jul 11 Posts: 990 Credit: 376,143,149 RAC: 0 |
As for the Climate Prediction tasks my Ryzen 7 is doing them just fine is one of my AMD Threadripper 1920X's, my 5950X but all my other pc's are all having problems with them.Looks like we're both crashing them on older machines, I've let them know here: https://www.cpdn.org/forum_thread.php?id=9149&postid=68939#68939 The above was double spaced between sentences, I apologise for the forum software ruining my post. |
Send message Joined: 13 Oct 16 Posts: 112 Credit: 1,174,293,644 RAC: 0 |
Will all submitted tasks be validated, and points given for work completed, or will any unvalidtaed tasks just be left in limbo? ^ This? I have 3003 tasks in "Validation pending" and 1189 tasks in "Validation inconclusive". WTF! We processed them...we should be given credit for them. Absolute horrible way to shut down an app! |
Send message Joined: 5 Jul 11 Posts: 990 Credit: 376,143,149 RAC: 0 |
WTF! We processed them...we should be given credit for them. Absolute horrible way to shut down an app!It's just credit, it's meaningless, it's not like they stole money from you for goodness sake. The above was double spaced between sentences, I apologise for the forum software ruining my post. |
Send message Joined: 12 Jun 10 Posts: 57 Credit: 6,171,817 RAC: 56 |
Mr Hucker that's a shame you find it pointless. I find tasks to run much faster on my GPU compared to my CPUHow much faster? I can't see any CPU tasks on your history to make a comparison. And are you accounting for the CPU being able to do 24 at a time? My 3080 GPU is roughly 27 minutes faster. The CPU 3900X (using at most 50% of my CPU so effectively no hyperthreading) runtime 31 minutes CPU GPU run time 4 minutes |
Send message Joined: 5 Jul 11 Posts: 990 Credit: 376,143,149 RAC: 0 |
I get 50 minutes with HT on with the same CPU, looks like you can get more out of it with HT.Mr Hucker that's a shame you find it pointless. I find tasks to run much faster on my GPU compared to my CPUHow much faster? I can't see any CPU tasks on your history to make a comparison. And are you accounting for the CPU being able to do 24 at a time? Since you get 12 done every 31 minutes, that's a task every 2.58 minutes from your CPU. Faster than your GPU, or is the GPU doing more than one at a time? The above was double spaced between sentences, I apologise for the forum software ruining my post. |
Send message Joined: 8 May 09 Posts: 3339 Credit: 524,010,781 RAC: 0 |
As for the Climate Prediction tasks my Ryzen 7 is doing them just fine is one of my AMD Threadripper 1920X's, my 5950X but all my other pc's are all having problems with them. It looks like they have an idea of where to look for the problem, thanks for posting all that |
Send message Joined: 19 Jul 10 Posts: 622 Credit: 19,254,980 RAC: 1 |
He forgot it on Tuesday, but the creation of new tasks should have been stopped the moment they had enough for science and not a minute later.Why did we not just stop releasing tasks first and THEN shut down the project, so a whole bunch of work doesn't get stuck in pending validation?Tom answered that when he said he got busy with his dissertation and forgot Also that's not an excuse for HOW they stopped Sepatation, look at the server status page, "workunits waiting for validation" growing as "tasks in progress" falls, people have thousands of tasks waiting for validation, apparently the server creates also new replacement tasks for those that have missed the deadline, as ready to send buffer is now over 62k and only 1k of that is nBody. I guess most of those 595 workunits waiting for assimilation are also Separation and they got stuck at that position when they ripped it out from the server. All this mess (and wasting computing ressources) could have been avoided if they stopped Separation in the right time and the right way. |
Send message Joined: 8 May 09 Posts: 3339 Credit: 524,010,781 RAC: 0 |
Cavalary wrote: ODLK, Moo Wrapper, LHC and LHC-dev all have 32bit tasks, don't know the ram requirements though, there are probably several more as well. https://stats6.free-dc.org/apps that's a list of every app at every project, even some closed down ones, click on the column that says Short and scroll down a few pages and you will see the projects with Windows X86 apps. You can then click on the Project name and then click the Project name at the top of the page for Projects stats and it will take you to that individual project so you either signup or see if the apps you want to run actually has any tasks available. |
Send message Joined: 19 Jul 10 Posts: 622 Credit: 19,254,980 RAC: 1 |
World Community Grid has 32-bit apps and in particular MCM is using very little RAM (about 80MB per task for the 64-bit app running here, 32-bit should be even lower). Einstein's Gamma-ray pulsar search #5 is 32-bit only for Windows and works even on XP, but requires more RAM, IIRC about 700MB per task, Binary Radio Pulsar Search has also 32-bit application and needs about 250MB per task. So yes, there are actually many 32-bit apps available.Only doing CPU work, and was looking for something to switch to that still worked on 32-bit Win 7 and used as little RAM as possible.ODLK, Moo Wrapper, LHC and LHC-dev all have 32bit tasks, don't know the ram requirements though, there are probably several more as well. |
Send message Joined: 30 Dec 09 Posts: 21 Credit: 75,543,393 RAC: 0 |
WTF! We processed them...we should be given credit for them. Absolute horrible way to shut down an app!It's just credit, it's meaningless, it's not like they stole money from you for goodness sake. Of course they did. In form of electricity. On a lot of machines. |
Send message Joined: 18 Jun 09 Posts: 35 Credit: 11,811,888 RAC: 0 |
did anyone FORCE you to run through lot's of electricity on lots of machines? no. so why the drama i do not understand |
Send message Joined: 5 Jul 11 Posts: 990 Credit: 376,143,149 RAC: 0 |
ODLK, Moo Wrapper, LHC and LHC-dev all have 32bit tasks, don't know the ram requirements though, there are probably several more as well.LHC almost always need virtualbox, nevermind 64 bit. Only the sixtrack might work on 32bit, and they hardly ever have those. The above was double spaced between sentences, I apologise for the forum software ruining my post. |
©2024 Astroinformatics Group