Folding Forum

Hi all,

I am joining Folding@Home after 10 years of hiatus, especially for helping to develop the cure for the COVID-19.
Currently, my server has 4xGTX1080Tis and 2x24 Cores Xeon.
I divide the CPU Cores into 3 slots: 16 cores, 16 cores, and 12 cores, leaving 4 cores for GPU-enabled work.
Then, another 4 slots go into GPUs.

I have been running the Folding@Home client for about 12 hours++.
I notice that only four Fahcore_xx are running in a given time (i.e., only 4 slots are active despite the fact that I have 7 slots).
The four Fahcore_xx that are running can be in the combination of single Fahcore_a7 and three Fahcore_22, two Fahcore_a7 and two Fahcore_22, or four Fahcore_22 without any Fahcore_a7.

Is there a way to increase the number of Fahcore_xx that can be run at the same time?

Thank you.

Subject: Temporary server outages
is probably related to you only having some of your slots active at any one time.

Hi Joe,

Thanks for the reply.
I have been running Folding@Home more than 2 days now and currently, I have two servers where each of them has 4 x NVIDIA GTX1080Ti and 2x24 Cores Xeon (for a total of 7 slots: 3 slots go to CPU (16+16+12) and 4 slots go to GPU)
According to my observation, only four slots are utilized at the same time on each server, whether it is running single Fahcore_a7 and three Fahcore_22, two Fahcore_a7 and two Fahcore_22, or four Fahcore_22 without any Fahcore_a7.
This makes me thinking whether there is a limit on how many slots can be run at the same time. If so, where can I change the limit so that all of the 7 slots on my two servers can be utilized?

Thanks,

bagus wrote:Hi Joe,

Thanks for the reply.
I have been running Folding@Home more than 2 days now and currently, I have two servers where each of them has 4 x NVIDIA GTX1080Ti and 2x24 Cores Xeon (for a total of 7 slots: 3 slots go to CPU (16+16+12) and 4 slots go to GPU)
According to my observation, only four slots are utilized at the same time on each server, whether it is running single Fahcore_a7 and three Fahcore_22, two Fahcore_a7 and two Fahcore_22, or four Fahcore_22 without any Fahcore_a7.
This makes me thinking whether there is a limit on how many slots can be run at the same time. If so, where can I change the limit so that all of the 7 slots on my two servers can be utilized?

Thanks,

For Nvidia GPUs, its necessary to reserve one CPU core as well because the CUDA/OpenCL calculations require some CPU. That might be why you aren't seeing full utilization, if that makes sense.

Hi all, thanks for the answer.

Joe was right. It depends on the availability of the WUs. Now, all my CPU cores (16+16+12) and my GPU (4 GPUs) can be utilized simultaneously.

Thanks and happy folding