# Using all nodes but not all the cpus

**URL:** <https://community.openpbs.org/t/using-all-nodes-but-not-all-the-cpus/3057>\
**Category:** Users/Site Administrators\
**Created:** [March 31, 2022, 7:45am UTC](https://community.openpbs.org/t/using-all-nodes-but-not-all-the-cpus/3057 "2022-03-31T07:45:54Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![watzinki](https://avatars.discourse-cdn.com/v4/letter/w/f04885/32.png) [@watzinki](https://community.openpbs.org/u/watzinki)\
**Post date:** [March 31, 2022, 7:45am UTC](https://community.openpbs.org/t/using-all-nodes-but-not-all-the-cpus/3057/1 "2022-03-31T07:45:54Z")

</div>

Hi All,

Let’s say I have five slave nodes each with 4cores(So total there are 20 cores), I am wondering if it is possible to request all the nodes but 19 cores in total while one core would stay idle?

Thanks

---

<div class="post-metadata">

**Author:** ![adarsh](https://avatars.discourse-cdn.com/v4/letter/a/f07891/32.png) [@adarsh](https://community.openpbs.org/u/adarsh)\
**Post date:** [March 31, 2022, 8:29am UTC](https://community.openpbs.org/t/using-all-nodes-but-not-all-the-cpus/3057/2 "2022-03-31T08:29:23Z")

</div>

Please try this  
`qsub -l select=1:ncpus=4+1:ncpus=4+1:ncpus=4+1:ncpus=4+1:ncpus=3 -- /bin/sleep 1000`