.. _Ant cluster:

Ant cluster
===========

.. _Ant overview:

Overview
--------

Ant is a university cluster available to ICP-affiliated personnel.
Software can be loaded via Spack or EESSI.

All of the storage available on the cluster is handled with a BeeGFS file system.
In total, we provide :qty:`224 <TB>` of storage shared by all users with :qty:`4.7 <Gbps>` write rate.
Note that each user is not given a specified amount of storage on the cluster and
this can cause issues when it is expended.

In total, the cluster consists of 1152 compute cores split across 18 nodes
using a :qty:`25 <Gbps>` interconnect. Each node is equipped with 2 NVIDIA GPUs.

+-----------+--------+-------+-----------------+----------------+--------------------+
| Type      | Nodes  | Cores | RAM             | GPU            | Node list          |
+===========+========+=======+=================+================+====================+
| Standard  | 18     | 2x32  | :qty:`384 <GB>` | 2x NVIDIA L4   | ``compute[01-18]`` |
+-----------+--------+-------+-----------------+----------------+--------------------+
| Login     | 1      | 2x32  | :qty:`384 <GB>` | 1x NVIDIA L4   |                    |
+-----------+--------+-------+-----------------+----------------+--------------------+

This cluster is managed by the ICP. If there are any issues,
please submit an issue `here <https://gitlab.icp.uni-stuttgart.de/icp/hivemind>`_.
The cluster was acquired via a DFG grant in 2022 [#grant_Ant_cluster]_
and became operational in March 2024.

.. _Ant partitions:

Partitions and nodes
--------------------

Contrary to most clusters, compute resources on Ant are not segmented in partitions.
Instead, one has to select resources from a node list.

Standard nodes
^^^^^^^^^^^^^^
There are 18 standard nodes equipped with 2 :cpu-amd-sheet2:`AMD EPYC 9374F <amd-epyc-9374f>`
(32 cores, 64 threads, :qty:`3.85 <GHz>`, :qty:`320 <W>`), :qty:`384 <GB>` of RAM DDR5, `Mellanox ConnectX-4 MCX4121A-ACAT
<https://docs.nvidia.com/nvidia-connectx-4-lx-ethernet-adapter-cards-user-manual.pdf>`__
(:qty:`25 <Gbps>`) interconnect.
Each node has 2 `NVIDIA L4 <https://www.nvidia.com/en-us/data-center/l4/>`__ (:qty:`24 <GB>` GDDR6).
All standard nodes follow the naming convention ``compute[x]`` where they can
take on values ``compute01 ... compute18``.

Login nodes
^^^^^^^^^^^
There is 1 login node with the exact same characteristics as the standard nodes,
hold 1 GPU.

.. _Ant access:

Access
------

Frank Huber is in charge of creating user accounts on Ant.
All ICP students and staff members are entitled to accessing Ant.
For external personnel, Christian Holm must personally approve access.
To log into Ant, refer to :ref:`Logging into clusters`.
See also :doc:`/_pages/_appendix/using_ant` for building software and submitting jobs.

.. _Ant obligations:

Obligations
-----------

DFG funding must be acknowledged as follows: [#DFG_acknowledgement]_

    We acknowledge funding from the Deutsche Forschungsgemeinschaft
    (DFG, German Research Foundation) through the Compute Cluster
    grant no. :dfg-gepris:`492175459`.

____

.. [#grant_Ant_cluster] DFG grant :dfg-gepris:`492175459` for Ant.

.. [#DFG_acknowledgement] See paragraph 15.1 of `DFG-Vordruck 2.18 - 01/20
   <https://www.dfg.de/formulare/2_18/v/dfg_2_18_de_v0120.pdf>`__
   (Besondere Verwendungsrichtlinien für Forschungsgroßgeräte
   nach Art. 91b Abs. 1 S. 1 GG i.V.m. AV-FGH).