Maximum PC

It has never taken this long for a WU anyone think there is something wrong. All cores at 100% Normaly it is a little over a day, not sure if I am running a new project but my PPD drop to 200 from around 8000.

I have one that is looking to take a long time too. 230 PPD (7 days from now, 14% done) on a P7611 (R4, C57, G154). Last I checked, my CPU was pegged w/ 8 threads @ 100%. My GPUs don't seem to be negatively affected though.

http://gero1369.com/folding/

Gero1369 and EasyE

There is a known issue with some P7611 work units. There is a report about it here on the Stanford forum.

My take on the forum comments is that if you're sure that these very long frame segments are way out of line with other work you fold, and that no other issues exist on your PC, that you should add your PRGC (P7611 R0 G23 C288 in EasyE's case) numbers to the Stanford forum so they have a report of the bad work unit.... then Dump the bad work unit.

One comment from Stanford leader tjlane said,

"I am sorry for the re-occurrence of this issue. Unfortunately I can't always guarantee this issue won't crop up, but I've done my best to mitigate it. The root cause is a bug in the A4 core, and I've reported it to the correct people. I think that because this issue occurs in >.1% of WUs, it hasn't been a priority for our dev team, which is usually swamped.

This issue will be resolved eventually, but in the mean time please feel free to dump these WUs. Not only are the points bad, but when this issue occurs the returned WU is meaningless. Therefore it's beneficial for the science (and the donor) to dump the WU.

I do apologize and will bug the core dev team again. Please let me know if there are any further questions or concerns."

Thanks Gary after looking at the description of p7611 looks like it should have never been sent out to low end systems, like most of us folders have.




Folding@home project descriptions

Project 7611

This project involves additional sampling of the FiP35 WW domain shot from the ultra-long trajectories run by DE Shaw on their new supercomputer ANTON. We are testing the differences between these new ultra-long trajectories and shorter ones from FAH, to test how simulations run on FAH stack up to more traditional methods.

Yep. i posted on the Folding Forum after I posted here. I have yet to dump the WU because I'm at work, but when I get home and have a few spare minutes, I'll get on it. I should be vigilant for the next several days to ensure that there are no more super-long WU times.

At least I'm not on vacation or something

70xx & 76xx WU's seem to be especailly problematic with this known bug. Still you do need to verify that it's a bad work unit, & not something else before dumping the work unit.

Checklist before dumping work unit:

1> verify that nothing else is hogging significant CPU cycles. Best way to do this (Windows 7 or older) is to open task manager, go to processes tab, select the "Processes from all users" option, then select the view menu & "Select Columns" check the "CPU Time" box then click on the new CPU Time headder to sort processes by most CPU Time used. After a week of "up time", I only have 11 processes with more than 1 minute CPU time & they're all FAH or Windows system related. Basically, you're just looking to verify that no process(es) are stealing significant CPU Time from FAH. This is one my biggest single gripes with Windows 8, they've removed this option... It is/was truly an invaluable tool for troubleshooting slow systems.

If you can verify that nothing else is hogging your CPU resources then, then continue to step 2, otherwise stop & fix your computer

2> report work unit to Stanford Forums, including Run, Clone, Generation data, and the first 5 - 10% of the FAH log for that work unit. Here's a bad work unit post I made a whie back that even shows what I'm talking about in step 1. http://foldingforum.org/viewtopic.php?f=19&t=22821

3> Save the log files, in case Stanford wants you to supply additional information later. I save mine to an old 256MB thumb drive that I don't otherwise use anymore.

4> dump the work unit & enjoy the return to normality. According to Stanford, this bug causes the finished work unit data to be unusable to the project. So even though you can fold it through to completion, & receive points for it, the effort is 100% wasted time & work.