
AWS Parallel Computing Service now supports node reboots via Slurm commands without triggering instance replacement, enabling more efficient cluster maintenance and troubleshooting.


<div>
<p>
AWS Parallel Computing Service (PCS) has introduced a new feature that allows node reboots via Slurm commands without triggering instance replacement.
</p>
<ul>
<li>Users can now reboot compute nodes for troubleshooting and resource cleanup</li>
<li>Reboots can be scheduled immediately or deferred using 'scontrol reboot' command</li>
<li>Other reboot methods will still trigger instance replacement</li>
<li>Feature is available in all AWS Regions where PCS is supported</li>
<li>Helps maintain cluster health more efficiently and at lower costs</li>
</ul>
<p>
The new capability provides HPC users with more flexible node management options within the AWS Parallel Computing Service.
</p>
</div>


AWS Parallel Computing Service (PCS) now supports node reboot via Slurm

Related articles

Related articles

Oct 17
2025
AWS Parallel Computing Service (PCS) now supports Slurm v25.05

Dec 18
2024
AWS Parallel Computing Service (PCS) now supports Slurm version 24.05

Nov 11
2025
AWS Parallel Computing Service (PCS) now supports Slurm CLI Filter plugins

Nov 20
2025
AWS Parallel Computing Service now supports Slurm REST API