Ticket 1951 - fair share consumption at job start
Summary: fair share consumption at job start
Status: OPEN
Alias: None
Product: Slurm
Classification: Unclassified
Component: Accounting (show other tickets)
Version: 16.05.x
Hardware: Linux Linux
: 5 - Enhancement
Assignee: Unassigned Developer
QA Contact:
URL:
Depends on:
Blocks:
 
Reported: 2015-09-16 07:49 MDT by Moe Jette
Modified: 2019-04-09 11:47 MDT (History)
4 users (show)

See Also:
Site: George Washington University, The
Alineos Sites: ---
Atos/Eviden Sites: ---
Confidential Site: ---
Coreweave sites: ---
Cray Sites: ---
DS9 clusters: ---
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: ---
Machine Name:
CLE Version:
Version Fixed:
Target Release: ---
DevPrio: ---
Emory-Cloud Sites: ---

Attachments
Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this ticket.
Description Moe Jette 2015-09-16 07:49:49 MDT 
Account for the full value of active allocations in fairshare immediately, refund unused cycles at job end
Comment 1 Moe Jette 2015-09-16 07:51:25 MDT 
Requested by BYU, GWU and Harvard
Comment 2 Danny Auble 2015-09-18 08:50:45 MDT 
The current idea would be to add the time of the association's grp_used_cpu_run_secs to it's usage_efctv to determine how much time has been used for fairshare.  usage_efctv will grow over time as the grp_used_cpu_run_secs shrinks.  Combining them only when dealing with fairshare should give us the total number of decayed time (usage_efctv) with the number of allocated time (grp_used_cpu_run_secs).

This idea is open for alteration.