Folding Forum

Xeon Gold 6244 is 8C/16T so this must be a 4S server. Either 2x32 or 4x16 should give the best results depending on Windows scheduling and how much time is spent syncing threads on different sockets.

13417 is writing "Global context and integrator variables" 10x less often than 13415. 19:46:25:WU01:FS01:0x22:Project: 13417 (Run 122, Clone 87, Gen 1) 19:46:25:WU01:FS01:0x22:Unit: 0x0000000312bc7d9a5efeb57be7135027 19:46:25:WU01:FS01:0x22:Reading tar file core.xml 19:46:25:WU01:FS01:0x22...

On Windows, you can use TCPView to kill the stalled download. Find FAHClient.exe in the list, right click the connection with vav4.ocis.temple.edu as the Remote Address and then click Close Connection.

The sample I have of p16451 would run on 48 threads. There are rare work units where the atoms have moved enough that the box changes shape or the estimated PME load changes. For 4x4x3 projects, 48 works when PME load is around 0.18. Once it drops towards 0.17, these CPU counts no longer work: 24, 3...

More logs. The first is an excerpt from %APPDATA%, from a stall at an indicated 99.99% again, where it seems like the GPU was paused for my peak utility window, then unpaused, found a checkpoint, but didn't make any progress (04:00:05:WU00:FS01:0x22: Found a checkpoint file). It appears to have act...

Because of the Quick Return Bonus, the more threads on the same work unit the better. I would recommend the default settings since the CPU slot will use 15 threads and that works most of the time. If you ever get a work unit that fails with an error about domain decomposition, temporarily set the sl...

I had the same failure on the same project, but the 2nd line isn't present. Should it be on default log level? Since this project is not supposed to be assigned to 21 threads, the second line being present would mean that the servers realised that and said, "You can have this work unit, but pl...

I have this happen a couple times a month as well on an RX 460 using Windows 8.1. For me the symptoms are 100% load reported in GPU-Z but memory is running at idle clock speed. Attempting to end the FahCore_22.exe in Task Manager results in "Access Denied". I have to reboot, which takes a ...

If you go to Start and type "Data Directory", that should be a shortcut to the FAH folder. GPUs.txt will be in there.

There's less work available for AMD GPUs that are GCN based until the new version of Core 22 is fully tested and generally available.

Another bad WU Can you also include the following lines in your reports? (The first line will always be present but the second one is rare.) 12:57:58:WU03:FS00:Requesting new work unit for slot 00: RUNNING cpu:21 from 128.252.203.4 01:30:04:WARNING:WU02:FS00:AS lowered CPUs from 21 to 18 Then it's ...

It gets tricky when GROMACS decides to break up 21 into 5 pme + 16 pp. A different project might break it up into 6 pme + 15 pp and the results might be different. I thought this was the normal allocation though? This in the first project that didn't want to work with 21 cores. :) Only the smallest...

p14253 uses more memory than most GPU projects. Can you try running MemtestCL and see if it reports any errors?
https://simtk.org/projects/memtest
or
https://www.majorgeeks.com/files/details/memtestcl.html

EDIT: Strangely, none of my other 3 GPUs working in the same time has received any 13409 WUs. Radeon VII is GCN so it's excluded from projects that would throw a shortSortList error. This will be fixed in a new build of core 22 . GCN restrictions for projects will be removed and then a greater vari...

First possible reason might be that the problem does not scale well to highly parallel architectures, parts of code with parallelization are intermixed with and dependent on parts which can not be parallelized, that way limiting the overall performance by those parts which can not be executed in pa...