Ticket 13666

Summary:	Need to set Maxjobperuser and make the additional submitted jobs not to be scheduled to a node until one of the running jobs finishes or timeout
Product:	Slurm	Reporter:	George Nabil <georgeeissa32>
Component:	Scheduling	Assignee:	Jacob Jenson <jacob>
Status:	RESOLVED INVALID	QA Contact:
Severity:	6 - No support contract
Priority:	---
Version:	22.05.x
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:	RHEL	Machine Name:	Imhotep
CLE Version:		Version Fixed:
Target Release:	---	DevPrio:	---
Emory-Cloud Sites:	---

Description George Nabil 2022-03-21 06:12:38 MDT

I have an issue that our system queue is becoming very intense, we were trying to make a maxjobperuser through slurm accounting but that released another issue that if user A will submit a large number of jobs that exceeds his maxjobperuser, his jobs would get scheduled by the backfill scheduler as soon as he submits them which will have more precedence over another user B who will submit his jobs after user A in spite of non-exceeding his maxjobperuser limit.

So I Need to set Maxjobperuser and make the additional submitted jobs submitted by the same user not to be scheduled to a node until one of the running jobs finishes or timeout.