# CPU bursting in a job

**URL:** <https://community.openpbs.org/t/cpu-bursting-in-a-job/866>\
**Category:** Users/Site Administrators\
**Created:** [February 16, 2018, 7:44pm UTC](https://community.openpbs.org/t/cpu-bursting-in-a-job/866 "2018-02-16T19:44:34Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![sscluster](https://avatars.discourse-cdn.com/v4/letter/s/c37758/32.png) [@sscluster](https://community.openpbs.org/u/sscluster)\
**Post date:** [February 16, 2018, 7:44pm UTC](https://community.openpbs.org/t/cpu-bursting-in-a-job/866/1 "2018-02-16T19:44:34Z")

</div>

CentOS 7  
PBSPro 14.1.0

I am trying to configure the PBS MOM daemon to allow for CPU bursting. Following the Admin guide I put the following in /var/spool/pbs/mom\_priv/config and restarted MOM on the compute nodes.

$enforce mem  
$enforce cpuaverage  
$enforce cpuburst  
$enforce delta\_percent\_over 50  
$enforce delta\_cpufactor 1.05  
$enforce delta\_weightup 0.4  
$enforce delta\_weightdown 0.1  
$enforce average\_percent\_over 50  
$enforce average\_cpufactor 1.025  
$enforce average\_trialperiod 120

After restart jobs are still killed with the following errors.

cormhap\_slancy\_defaults.o437013.31:=\>\> PBS: job killed: ncpus 17.4 exceeded limit 16 (burst)  
cormhap\_slancy\_defaults.o437013.32:=\>\> PBS: job killed: ncpus 17.7 exceeded limit 16 (burst)  
cormhap\_slancy\_defaults.o437013.35:=\>\> PBS: job killed: ncpus 17.8 exceeded limit 16 (burst)  
cormhap\_slancy\_defaults.o437013.36:=\>\> PBS: job killed: ncpus 17.8 exceeded limit 16 (burst)  
cormhap\_slancy\_defaults.o437013.40:=\>\> PBS: job killed: ncpus 17.6 exceeded limit 16 (burst)  
cormhap\_slancy\_defaults.o437016.36:=\>\> PBS: job killed: ncpus 17.8 exceeded limit 16 (burst)

Have I missed a step in getting this configured correctly?

-SS-

---

<div class="post-metadata">

**Author:** ![scc](https://yyz2.discourse-cdn.com/flex030/user_avatar/community.openpbs.org/scc/32/234_2.png) [@scc](https://community.openpbs.org/u/scc)\
**Post date:** [February 19, 2018, 9:24pm UTC](https://community.openpbs.org/t/cpu-bursting-in-a-job/866/2 "2018-02-19T21:24:44Z")

</div>

I believe, but am not 100% sure, that with these parameters set (which all appear to be the default aside from turning actual enforcement on) a 16 cpu job will be killed if it uses more than 17.3, which these all are. From the manual:

The job is killed if the following is true:  
new\_cpupercent \> ((ncpus \* 100 \* delta\_cpufactor) + delta\_percent\_over)

((16 \* 100 \* 1.05) + 50 ) = 1730

---

<div class="post-metadata">

**Author:** ![sscluster](https://avatars.discourse-cdn.com/v4/letter/s/c37758/32.png) [@sscluster](https://community.openpbs.org/u/sscluster)\
**Post date:** [February 19, 2018, 10:31pm UTC](https://community.openpbs.org/t/cpu-bursting-in-a-job/866/3 "2018-02-19T22:31:28Z")

</div>

Scott,  
So the delta\_percent\_over is not a percentage of the number of cpus where in this instance it would be 800? Is the percentage just a mislableling?

I would assume that the delta\_percentage\_over would be used in the following method to follow the formula above if it were to make sense in a percentage model.:

((16 \* 100 \* 1.05) + 800))

-SS-

---

<div class="post-metadata">

**Author:** ![scc](https://yyz2.discourse-cdn.com/flex030/user_avatar/community.openpbs.org/scc/32/234_2.png) [@scc](https://community.openpbs.org/u/scc)\
**Post date:** [February 20, 2018, 3:26pm UTC](https://community.openpbs.org/t/cpu-bursting-in-a-job/866/4 "2018-02-20T15:26:44Z")

</div>

I am not sure it is actually mislabeled, but there are multiple ways in which it COULD make sense. The implementation is expressed in absolute percentage of cpus, that is, for a value of 50 you are saying: “allow usage of an extra half of a cpu” where the amount of extra cpu that can be used is expressed in terms of percentage of a cpu. It does not allow for one to say"allow usage of an extra 50% over whatever the job’s (ncpus \* 100 \* delta\_cpufactor) happens to be".

---

<div class="post-metadata">

**Author:** ![sscluster](https://avatars.discourse-cdn.com/v4/letter/s/c37758/32.png) [@sscluster](https://community.openpbs.org/u/sscluster)\
**Post date:** [February 20, 2018, 8:35pm UTC](https://community.openpbs.org/t/cpu-bursting-in-a-job/866/5 "2018-02-20T20:35:20Z")

</div>

Scott,  
Think I have figured out how to get the 50% bursting overhead on jobs. I just needed to change $enforce delta\_cpufactor 1.05 to $enforce delta\_cpufactor 1.50 and it seems to have done trick.

As a side note there does seem to be an error in the docs as it states in the table labelled 5-15 that the default is 1.50, but it is actually 1.05.

-SS-