Hello Support and Developers,
One of customer is experiencing issues running Ubuntu 12.04 LTS using kernel 3.13.0-52 on the following system:
Brief system specs:
2x Intel Xeon E5-2680v3 12 core CPU’s
HyperThreading Enabled for a total of 48 cores per system
8X Nvidia Tesla K80 GPU Card
256 GB of RAM
The issue they are experiencing is very sporadic and I haven’t been able to reproduce the issue reliably yet but it does happen often. During system install or at random times the system load will be as high as 6000+ with the top processes being acpi_pad, I realize that I can simply blacklist this module but I would like to understand exactly what is happening to generate this amount of load on the system?
Also, your Nvidia 629 log collecting script doesn’t support K80’s. Do you have newer version of the script so we can collect the logs?
Penguin Tech Support