# How to tell the difference between cores used and cores reserved

**URL:** https://forum.kavli.tudelft.nl/t/how-to-tell-the-difference-between-cores-used-and-cores-reserved/163
**Category:** Cluster
**Created:** [28 April 2021 11:53 UTC](https://forum.kavli.tudelft.nl/t/how-to-tell-the-difference-between-cores-used-and-cores-reserved/163 "2021-04-28T11:53:25Z")
**Posts on this page:** 4
**Page:** 1

<div class="post-metadata">

### Author: ![mdepken](https://forum.kavli.tudelft.nl/letter_avatar_proxy/v4/letter/m/848f3c/32.png) [@mdepken](https://forum.kavli.tudelft.nl/u/mdepken)
#### Post date: [28 April 2021 11:53 UTC](https://forum.kavli.tudelft.nl/t/how-to-tell-the-difference-between-cores-used-and-cores-reserved/163/1 "2021-04-28T11:53:25Z")

</div>

Hi All,

We (depken group BN) have had a few instances where it turns out we are using many more cores than what we reserved and thought we were using.

We think we have fixed the problem, but is there an easy way to find out how many cores you actually use, so we can stop ourselves from creating havoc in the future? 🙂

Kind regrads,  
Martin

---

<div class="post-metadata">

### Author: ![pulkin](https://forum.kavli.tudelft.nl/letter_avatar_proxy/v4/letter/p/cdc98d/32.png) [@pulkin](https://forum.kavli.tudelft.nl/u/pulkin)
#### Post date: [28 April 2021 12:43 UTC](https://forum.kavli.tudelft.nl/t/how-to-tell-the-difference-between-cores-used-and-cores-reserved/163/2 "2021-04-28T12:43:45Z")

</div>

What’s the context? hpc05? PBS?

---

<div class="post-metadata">

### Author: ![anton-akhmerov](https://forum.kavli.tudelft.nl/user_avatar/forum.kavli.tudelft.nl/anton-akhmerov/32/4_2.png) [@anton-akhmerov](https://forum.kavli.tudelft.nl/u/anton-akhmerov)
#### Post date: [28 April 2021 12:49 UTC](https://forum.kavli.tudelft.nl/t/how-to-tell-the-difference-between-cores-used-and-cores-reserved/163/3 "2021-04-28T12:49:31Z")

</div>

Yes, this is indeed the question about hpc05 and PBS.

I usually check the monitoring tool [https://hpc05.quantumtinkerer.tudelft.nl/](https://hpc05.quantumtinkerer.tudelft.nl/). This one [uses](https://github.com/basnijholt/cluster-logger/blob/master/logger.py#L21) `qstat -ea`, followed by [simple parsing](https://github.com/basnijholt/cluster-logger/blob/master/logger.py#L99-L105) that collects the statistics in a human-readable form.

However it is usually a better idea to run the task in a similar environment, possibly by getting an interactive PBS job, which will give you a terminal on one of the nodes, and then to check resource usage directly.

---

<div class="post-metadata">

### Author: ![mdepken](https://forum.kavli.tudelft.nl/letter_avatar_proxy/v4/letter/m/848f3c/32.png) [@mdepken](https://forum.kavli.tudelft.nl/u/mdepken)
#### Post date: [29 April 2021 09:40 UTC](https://forum.kavli.tudelft.nl/t/how-to-tell-the-difference-between-cores-used-and-cores-reserved/163/4 "2021-04-29T09:40:22Z")

</div>

Alright. Thanks! I will start with the monitoring tool as it does seem to give the info I was after.
