...
Node Type | CPU | Memory | Accelerator |
|---|---|---|---|
Standard Compute | 128 Cores (2x AMD EPYC 7713 Zen3) | 512 GiB | N/A |
High Memory | 128 Cores (2x AMD EPYC 7713 Zen3) | 2048 GiB | N/A |
GPU A100 | 48 Cores (2x AMD EPYC 7413 Zen3) | 512 GiB | 4x NVIDIA A100 80GiB |
GPU A30 | 48 Cores (2x AMD EPYC 7413 Zen3) | 512 GiB | 3x NVIDIA A30 24GiB |
GPU MIG | 48 Cores (2x AMD EPYC 7413 Zen3) | 512 GiB | 28x 16x NVIDIA A100 sliced into 20GiB and 10GiB segments“slices” |
Xlienx FPGA | 48 Cores (2x AMD EPYC 7443 Zen3) | 256 GiB | 1x Xilinx U280 |
Bitaware FPGA | 52 Cores (Intel Xeon Gold 6230R) | 376 GiB | 1x BittWare 520N-MX |
| Info |
|---|
There is privately owned hardware that may have slightly different specs. See the Sol Status Page for the full features of every node |
Requesting Resources
| Excerpt |
|---|
Requesting CPUs |
By default, Slurm will attempt to schedule jobs as locally proximate as possible, so requesting -c 5 cores will try to get 5 cores on one node -N 1, unless otherwise specified.
To request a given number of CPUs sharing the same node, you can use the following in your
This will create a job with 5 CPU cores on one node. To request a given number of CPUs spread across multiple nodes, you can use the following:
The above example will allocate |
...
a total of 50 cores spread across as few as 2 nodes or as many as 4 nodes. Take note of the inclusion or omission of
This reduced example will still allocate 50 cores, 5 cores per task on any number of available nodes. Note, that unless you are using MPI-aware software, you will likely prefer to always add |
...
, to ensure that each job worker has sufficient connectivity |
...
. -c and -n have similar effects in Slurm in allocating cores, but -n is the number of tasks, and -c is the number of cores per task. MPI processes bind to a task, so the general rule of thumb is for MPI jobs to allocate tasks, while serial jobs allocate cores, and hybrid jobs allocate both. See the official Slurm documentation for more information: https://slurm.schedmd.com/sbatch.html Requesting Memory |
...
Cores and memory are de-coupled: if you need only a single CPU core but ample memory, you can do so like this:
If you do not specify --mem, you will be allocated 2GiB per CPU core OR 24GiB per GPU To request more than 512GiB of memory, you will need to use the highmem partition
To request all available memory on a node |
...
:
Requesting GPUsTo request a GPU, you can specify the -G option within your job request: This will allocate the first available GPU that fits your job request
To request multiple GPU’s specify a number greater than 1:
To request a specific number of GPU’s per node when running multi node:
To request a specific type of GPU (a100 for example):
|
Requesting FPGAs
Sol has two nodes with a Field Programmable Gate Array (FPGA) accelerator. One is an Intel-based node with a Bitaware 520N-MX FPGA, the other is an AMD-based node with a Xilinx U280. Because there is only FPGA per node, it is recommended to allocate the entire node.
Bitaware:
| Code Block |
|---|
#SBATCH --exclusive
#SBATCH -L bittware
or
interactive --exclusive -L bittware |
Xilinx:
| Code Block |
|---|
#SBATCH --exclusive
#SBATCH -L xilinx
or
interactive --exclusive -L xilinx |
Or via the web portal using the “Additional Sbatch options” section:
Note there should not be a space between “-L” and the FPGA name on the web portal
 |  |
Additional Help
If you require further assistance on this topic, please don't hesitate to contact the Research Computing Team. To create a support ticket, kindly send an email to rtshelp@asu.edu. For quick inquiries, you're welcome to reach out via our #rc-support Slack Channel or attend our office hours for live assistance
...