| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Active Research
Forum: OpenPandemics - COVID-19 Project
Thread: "September 2020 Update: Fight AIDS@Home" article is relevant to OpenPandemics |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 15
|
|
| Author |
|
|
Aurum
Master Cruncher The Great Basin Joined: Dec 24, 2017 Post Count: 2391 Status: Offline Project Badges: 
|
Isn't that call made by the research team? Keith Uplinger is the only one I recall ever saying anything about it. I don't recall any reference to Forli Lab or any comments from Forli. Do you? Since they've gone into radio silence no one knows what their current thinking is.  ...KRI please cancel all shadow-banning |
||
|
|
Aurum
Master Cruncher The Great Basin Joined: Dec 24, 2017 Post Count: 2391 Status: Offline Project Badges:
|
I know of several projects that use both CPUs and GPUs on the same OpenCl work units. At the moment, Einstein does on the gravity wave work units, and MilkyWay is planning to do so shortly on the N-body work. In the past, there was also POEM. GPUGrid did it and it was a disaster....KRI please cancel all shadow-banning |
||
|
|
Jim1348
Veteran Cruncher USA Joined: Jul 13, 2009 Post Count: 1066 Status: Offline Project Badges: 
|
GPUGrid did it and it was a disaster. No, GPUGrid is CUDA only, and has never used OpenCl on the GPUs. They had a Quantum Chemistry project for the CPUs, but that was entirely separate from the GPU work (and not OpenCl anyway). [Edit 1 times, last edit by Jim1348 at Sep 16, 2020 1:34:05 AM] |
||
|
|
Rickjb
Veteran Cruncher Australia Joined: Sep 17, 2006 Post Count: 666 Status: Offline Project Badges: 
|
Aurun420 said: "And yet IBM wants to shove all OPN WUs down the same pipe so how will that be possible???"
----------------------------------------When HCC ran on GPUs, it was run on CPUs at the same time, drawing WUs from the same pool. The techs had to get the software to run so that the results from a WU would be the same, independent of whether it had run on a GPU or CPU. The definition of "the same" may have needed to be relaxed just a little because some GPUs may handle roundoff of the least significant bit of calculated values differently to the method used by the FPUs in standard CPUs, and the GPUs may have been using "single"-precision FP (32-bit) vs "double"-precision (64-bit) on the CPUs. It's not just the false positives from inaccurate simulations that are a problem. The positives can be further filtered to weed out the false ones, but the false negatives may be missed completely. [Edit]: Forgot to remind people that VINA and BEDAM, as deployed so far, do not have the covalent bonding protocol that has been added to AutoDock for OPN, so unless they are updated they will not be useful to filter any of the AutoDock positives that might bond covalently. [/Edit] [Edit 2 times, last edit by Rickjb at Sep 17, 2020 4:20:42 PM] |
||
|
|
Aurum
Master Cruncher The Great Basin Joined: Dec 24, 2017 Post Count: 2391 Status: Offline Project Badges:
|
GPUGrid did it and it was a disaster. No, GPUGrid is CUDA only, and has never used OpenCl on the GPUs. They had a Quantum Chemistry project for the CPUs, but that was entirely separate from the GPU work (and not OpenCl anyway). ...KRI please cancel all shadow-banning |
||
|
|
|