I don't have specific answers to your questions but one thing you can do is
run the slurmd on one of your "nodes" and see what hardware specs SLURM
auto-detects.
Run "slurmd -C"; from the man page:
-C Print actual hardware configuration and exit. The format of
output is the same as used in slurm.conf to describe a node's
configuration plus it's uptime.
On Sun, Aug 1, 2021 at 10:27 AM Nicholas Yue <yue.nicho...@gmail.com> wrote:
> Hi,
>
> I am hoping to learn more about managing jobs and configuring SLURM
>
> I have a Picocluster 5H which has 5 units of Jetson Nano.
>
> Each Jetson Nano is a Arm A57 (Quad core) + and Nvidia Maxwell GPU
>
> I am using one of the node as head node with the other four as compute
> nodes
>
> The OS is Ubuntu 18.04 LTS and I believe the SLURM version for it is at
> SLURM version 17.11.2
>
> As the cluster nodes have very little RAM and storage, I am hoping to use
> as much software out of the box (that comes with Ubuntu) as possible rather
> than set up a development environment to build the latest of everything.
>
> Has anyone run the stock standard SLURM on Ubuntu 18.04 with a Jetson Nano
> ? I am particularly interested in the setup/configuration so that all the
> resources (cores/RAM/GPU) are fully at the disposal of SLURM. Are there any
> special considerations of GRES for the Jetson Nano's Maxwell GPU ?
>
> Given my hardware configuration, are there any important features I will
> be missing out on if I were to continue using SLURM 17.11.2 ?
>
> Kind regards
>
> --
> Nicholas Yue
> https://www.linkedin.com/in/nicholasyue/
>
