From 7a21732a9540fa6d756466cfc02c209b8d5f96c1 Mon Sep 17 00:00:00 2001
From: Taras Lazariv <taras.lazariv@tu-dresden.de>
Date: Fri, 11 Feb 2022 13:28:21 +0100
Subject: [PATCH] Explained multithreading memory
---
.../docs/jobs_and_resources/partitions_and_limits.md | 7 ++++---
1 file changed, 4 insertions(+), 3 deletions(-)
diff --git a/doc.zih.tu-dresden.de/docs/jobs_and_resources/partitions_and_limits.md b/doc.zih.tu-dresden.de/docs/jobs_and_resources/partitions_and_limits.md
index 973378cbc..cec8c5b2e 100644
--- a/doc.zih.tu-dresden.de/docs/jobs_and_resources/partitions_and_limits.md
+++ b/doc.zih.tu-dresden.de/docs/jobs_and_resources/partitions_and_limits.md
@@ -75,9 +75,10 @@ be aware of the limits shown in the following table.
!!! note
- The ML nodes have 4way-SMT, so for every physical core allocated (e.g., with
- `SLURM_HINT=nomultithread`), you will always get 4*1443 MB, because the memory of the other
- threads is allocated implicitly, too.
+ Some nodes have multithreading (SMT) enabled, so for every physical core allocated
+ (e.g., with `SLURM_HINT=nomultithread`), you will always get `MB per Core`*`number of threads`,
+ because the memory of the other threads is allocated implicitly, too.
+ Those nodes are marked with an asterisk.
Some of the partitions, denoted with a double asterisk, have a counterpart for interactive
jobs. These partitions have a `-interactive` suffix (e.g. `ml-interactive`) and have the same
configuration.
--
GitLab