From b5a56c26c794b328d136765dc456144f48cdc415 Mon Sep 17 00:00:00 2001
From: Adam Caprez <acaprez2@unl.edu>
Date: Fri, 31 May 2019 21:18:07 +0000
Subject: [PATCH] Clarify use of gres flag for GPU images.
Resolves #18.
---
content/guides/running_applications/using_singularity.md | 9 ++++++++-
1 file changed, 8 insertions(+), 1 deletion(-)
diff --git a/content/guides/running_applications/using_singularity.md b/content/guides/running_applications/using_singularity.md
index 7ab416ee..b04b7d1f 100644
--- a/content/guides/running_applications/using_singularity.md
+++ b/content/guides/running_applications/using_singularity.md
@@ -142,10 +142,17 @@ the `tensorflow-gpu` image and need the packages `nibabel` and `tables`.
{{% panel theme="info" header="Run an interactive SLURM job" %}}
{{< highlight bash >}}
-srun --pty --mem=4gb --qos=short $SHELL
+srun --pty --mem=4gb --qos=short --gres=gpu --partition=gpu $SHELL
{{< /highlight >}}
{{% /panel %}}
+{{% notice info %}}
+The `--gres=gpu --partition=gpu` options are used here as the `tensorflow-gpu` image is
+GPU enabled. If you are using a non-GPU image, you may omit those options. See the page
+on [submitting GPU jobs]({{< relref "/guides/submitting_jobs/submitting_cuda_or_openacc_jobs/_index.md" >}})
+for more information.
+{{% /notice %}}
+
After the job starts, the prompt will change to indicate you're on a
worker node. Next, start an interactive session in the container.
--
GitLab