Ticket 7733

Summary: How to utilize rest over GPU's of the system
Product: Slurm Reporter: Rmula <rahulgoud.mula>
Component: SchedulingAssignee: Jacob Jenson <jacob>
Status: RESOLVED INVALID QA Contact:
Severity: 6 - No support contract    
Priority: --- CC: rahulgoud.mula
Version: 19.05.0   
Hardware: Linux   
OS: Linux   
Site: -Other- Slinky Site: ---
Alineos Sites: --- Atos/Eviden Sites: ---
Confidential Site: --- Coreweave sites: ---
Cray Sites: --- DS9 clusters: ---
Google sites: --- HPCnow Sites: ---
HPE Sites: --- IBM Sites: ---
NOAA SIte: --- NoveTech Sites: ---
Nvidia HWinf-CS Sites: --- OCF Sites: ---
Recursion Pharma Sites: --- SFW Sites: ---
SNIC sites: --- Tzag Elita Sites: ---
Linux Distro: Ubuntu Machine Name:
CLE Version: Version Fixed:
Target Release: --- DevPrio: ---
Emory-Cloud Sites: ---

Description Rmula 2019-09-11 22:58:08 MDT 
Assume we have 4 GPU machine, while trying to allocate/srun 2 GPU of the 4 GPU machine, system is allocated and job is scheduled, I have launched parallel session and trying to allocate/srun restover 2 Gpus of the machine, where job is going to queued state.

Can’t we parallelly utilize the rest over GPU’s of the system ?

Is there any other approach to achieve this ?

Thanks in advance!