I am running Einsten@home gravitational tasks on a Virtual Linux machine with the Tumbleweed development SuSE OS and they run well on the A10-6700 CPU. But the OS does not see the nVidia graphic board and insists that there is a VMWare card, which does not exist. So I could not run GPU tasks on it, while the host Windows 10 makes use of thecard. OK, it is a GTX 750 only, for reason of cable connectors used by the monitor, it was the only one with avGA connector.
Just FYI: Since the current FGRP tasks run on all cards, we have a working solution to prevent the scheduler from sending old, problematic tasks to new cards, and our manpower is pretty limited, I'm postponing further work on that (FGRP) issue and attending to more urgent things. Among that is getting some GPU code for the Gravitational Wave serach to work on E@H.
Back to the original data file topic: after a pause in work issue, the new tasks for the GPU Gamma-Ray Pulsar application use data file 1049M. This for me is a bit of a surprise, and was initially a concern, as the group of data files for which Turing cards have enjoyed success has been of the name form 10nnL.
So I did the obvious thing and promoted two tasks to prompt execution, one each on the GTX 1070 and the RTX 2080 in my primary box. Both ran to completion, with elapsed times the same as the 1049L tasks. So I breathed a sigh of relief, and wondered what the real way to discern Turing-safe tasks may be.
I trust what Bernd posted about not sending offending tasks to Turing and Volta anymore. I just removed my gpu_exclude for the 2080 and let it have at em. Haven't seen any of the 2000 series yet so I think his fix must be working.
I am running Einsten@home
)
I am running Einsten@home gravitational tasks on a Virtual Linux machine with the Tumbleweed development SuSE OS and they run well on the A10-6700 CPU. But the OS does not see the nVidia graphic board and insists that there is a VMWare card, which does not exist. So I could not run GPU tasks on it, while the host Windows 10 makes use of thecard. OK, it is a GTX 750 only, for reason of cable connectors used by the monitor, it was the only one with avGA connector.
Tullio
Bernd Machenschalk wrote:Just
)
YES!!!
Back to the original data
)
Back to the original data file topic: after a pause in work issue, the new tasks for the GPU Gamma-Ray Pulsar application use data file 1049M. This for me is a bit of a surprise, and was initially a concern, as the group of data files for which Turing cards have enjoyed success has been of the name form 10nnL.
So I did the obvious thing and promoted two tasks to prompt execution, one each on the GTX 1070 and the RTX 2080 in my primary box. Both ran to completion, with elapsed times the same as the 1049L tasks. So I breathed a sigh of relief, and wondered what the real way to discern Turing-safe tasks may be.
I trust what Bernd posted
)
I trust what Bernd posted about not sending offending tasks to Turing and Volta anymore. I just removed my gpu_exclude for the 2080 and let it have at em. Haven't seen any of the 2000 series yet so I think his fix must be working.