# How to define GPU load

**URL:** <https://community.openpbs.org/t/how-to-define-gpu-load/2922>\
**Category:** Users/Site Administrators\
**Created:** [January 4, 2022, 11:51am UTC](https://community.openpbs.org/t/how-to-define-gpu-load/2922 "2022-01-04T11:51:46Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![ddivad](https://avatars.discourse-cdn.com/v4/letter/d/f05b48/32.png) [@ddivad](https://community.openpbs.org/u/ddivad)\
**Post date:** [January 4, 2022, 11:51am UTC](https://community.openpbs.org/t/how-to-define-gpu-load/2922/1 "2022-01-04T11:51:46Z")

</div>

Hello SMEs,

I configured OpenPBS 20.0 for job allocations to GPU servers based on server load  
Configured /var/spool/pbs/mom\_priv/config like this-\>  
$max\_load 10.0  
$ideal\_load 5.0

This is working fine and allocation is happening based on CPU load of the GPu servers.  
Can you please suggest how I can configure job allocations based on GPU usage rather than CPU load ?

---

<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:** [January 4, 2022, 6:28pm UTC](https://community.openpbs.org/t/how-to-define-gpu-load/2922/2 "2022-01-04T18:28:58Z")

</div>

> [@ddivad](#):
>
> Can you please suggest how I can configure job allocations based on GPU usage rather than CPU load ?

For example: if there are 3 nodes , each with 2 GPU cards, and you would like to schedule a job on the node which has more unused GPU cards, then you can use node\_sort\_key against the ngpus (host consumable resource type long - resources\_available.ngpus=XY)

For example: if you have an applicaiton that would report total load on the gpu load on the GPU-compute node and this value is updated to a custome resource value of compute node ( resources\_available.gpuload=XYZ), then this value can be used in the node sort key and sorted based on the gpu load of the compute node.

Hope this helps.