All's well and good, the problem was caused by the driver version. Went from 531.29 to 537.13 this morning with a clean install as usual, recycled the machine and both GPUs are now working as they should. Many thanks for the suggestions that were given.
FWIW the last remaining result of this run dropped in recently. The run is now finally complete. The current GW run (O3ASHF1) is GPU only, CPUs will not be used for GW searches for a while.
One question here, since there still is quite some GPU downtime with the GW tasks, especially with older CPUs. But also when I run 2 task the same time with offset starting times the GPU drops down to 50% of it's performance while the CPU is calculating one task. A single task is not using all of the GPU on neither of my systems. Wether I run them at 1x or 2x.
Now CPUs have much more cores than a system has GPUs.
Is there a way to load the next task into the GPU already while the current task is finished by the CPU?
And also pre-calculate the next task by the CPU. So with a fast GPU but slow CPU system for example you have 2-3 cores finishing an task, the GPU is calculating with support of two cores and 2-3 cores are preparing a task. That would be full utilisation of a 8 core system for GW tasks. If possible to handle tasks like that within the boinc app wouldn't this be a way to enhance GPU efficency and get a lot more tasks done faster?
One question here, since there still is quite some GPU downtime with the GW tasks, especially with older CPUs. But also when I run 2 task the same time with offset starting times the GPU drops down to 50% of it's performance while the CPU is calculating one task. A single task is not using all of the GPU on neither of my systems. Wether I run them at 1x or 2x.
Now CPUs have much more cores than a system has GPUs.
Is there a way to load the next task into the GPU already while the current task is finished by the CPU?
And also pre-calculate the next task by the CPU. So with a fast GPU but slow CPU system for example you have 2-3 cores finishing an task, the GPU is calculating with support of two cores and 2-3 cores are preparing a task. That would be full utilisation of a 8 core system for GW tasks. If possible to handle tasks like that within the boinc app wouldn't this be a way to enhance GPU efficency and get a lot more tasks done faster?
you're posting in the thread for the Multi-Directional tasks. these tasks are no longer being distributed and that search is over. the new tasks are "All Sky" hence the "AS" in the task type "O3AS"
All's well and good, the
)
All's well and good, the problem was caused by the driver version. Went from 531.29 to 537.13 this morning with a clean install as usual, recycled the machine and both GPUs are now working as they should. Many thanks for the suggestions that were given.
I don't buy computers, I build them!
O3MD1 WUs have stopped being
)
O3MD1 WUs have stopped being distributed. Server Status says Tasks Total are 127,634 and Tasks to Send are none.
I've almost completed the WUs I've received. Will they be restarted or is it time to switch to another project?
Only current tasks for
)
Only current tasks for Einstein are BRP4/BRP7 tasks, FGRP5 cpu tasks and the O3AS Gravity Wave tasks.
Change your project preferences settings.
FWIW the last remaining
)
FWIW the last remaining result of this run dropped in recently. The run is now finally complete. The current GW run (O3ASHF1) is GPU only, CPUs will not be used for GW searches for a while.
BM
One question here, since
)
One question here, since there still is quite some GPU downtime with the GW tasks, especially with older CPUs. But also when I run 2 task the same time with offset starting times the GPU drops down to 50% of it's performance while the CPU is calculating one task. A single task is not using all of the GPU on neither of my systems. Wether I run them at 1x or 2x.
Now CPUs have much more cores than a system has GPUs.
Is there a way to load the next task into the GPU already while the current task is finished by the CPU?
And also pre-calculate the next task by the CPU. So with a fast GPU but slow CPU system for example you have 2-3 cores finishing an task, the GPU is calculating with support of two cores and 2-3 cores are preparing a task. That would be full utilisation of a 8 core system for GW tasks. If possible to handle tasks like that within the boinc app wouldn't this be a way to enhance GPU efficency and get a lot more tasks done faster?
B.I.G wrote: One question
)
you're posting in the thread for the Multi-Directional tasks. these tasks are no longer being distributed and that search is over. the new tasks are "All Sky" hence the "AS" in the task type "O3AS"
post here: https://einsteinathome.org/content/all-sky-gravitational-wave-search-o3-data-o3ashf1
_________________________________________________________________________