Merge pull request #50 from flindersuni/feature/dash-and-cst-studio

CST Studio & HPC Dashboard

docs/source/index.rst

@@ -3,11 +3,10 @@ Welcome to the DeepThought HPC
 
 The new Flinders University HPC is called DeepThought. This new HPC comprises of AMD EPYC based hardware and next-generation management software, allowing for a dynamic and agile HPC service. 
 
-.. _/cluster: storage/storageusage.html
-
 .. attention:: 
-    The new BeeGFS Parallel Filesystem mounted at /cluster has just been deployed, and is **now ready for usage**. It will appear in any disk 
-    usage listings on the HPC. For further information please read the `/cluster`_ section of the Storage Usage & Guidelines.
+    The new HPC Dashboard has been released. The default, 2-Week Window 'Quick View' can be accessed at https://deepweb.flinders.edu.au/,
+    https://deepweb.flinders.edu.au/public-dashboards/9f967160b3d0441c8fe9c7eee9f231c5, or https://deepweb.flinders.edu.au/dtdash/. 
+    Be advised the the 'full' version ('/dtdash') will be removed in the future. 
 
 .. attention::
     This documentation is under active development, meaning that it can
@@ -47,6 +46,37 @@ The following files are provided for integration into your reference manager of
 - RIS_
 
 
+
+.. _Default URL: https://deepweb.flinders.edu.au/
+.. _Public Dashboard: https://deepweb.flinders.edu.au/public-dashboards/9f967160b3d0441c8fe9c7eee9f231c5
+.. _Full Version: https://deepweb.flinders.edu.au/dtdash
+
+
+HPC Visual Dashboard
++++++++++++++++++++++++++++
+DeepThought has a time-series based visual statistics dashboard that be viewed via a web-browser while on campus.  
+
+THe 'Default' and 'Public' URL's are using an *Alpha Release* feature set, and may not display correctly. In that case, please use the 
+'Full Version' link, and any display strangeness will be resolved. 
+
+The following URLS all link to the dashboard. 
+
+1. `Default URL`_ 
+2. `Public Dashboard`_ 
+3. `Full Version`_  
+
+
+The 'Full Version' has the capability to change the time-period viewed, and will be removed when the 'Public' and 'Default' feature set reach 
+maturity. 
+
+A screen shot of a small portion of the Dashboard is below.
+
+.. figure:: _static/GrafanaDashSnapshot.png
+    :align: center
+    :alt: HPC Visual Statistics Dashboard
+
+
+
 Table of Contents
 ====================
 
@@ -71,6 +101,7 @@ Table of Contents
 
     software/softwaresuitesoverview.rst
     software/ansys.rst
+    software/cststudio.rst
     software/delft3d.rst 
     software/gromacs.rst
     software/jupyter.rst

docs/source/software/ansys.rst

@@ -14,7 +14,7 @@ ANSYS 2021R2 is the current version of the ANSYS Suite installed on the HPC. Bot
 Before You Start
 ==============================
 
-APDL, Fluent and the EM Suite / HFSS all have a common requirement. All script, journal or run files **must** have been generated in *TUI* Mode. This means that all command must been created, and saved via the 
+APDL, Fluent and the EM Suite / HFSS all have a common requirement. All script, journal or run files **must** have been generated in *TUI* Mode. This means that all commands must been created, and saved via the 
 programs' Command Line Interface. An example of this is - in Fluent, you **cannot** use the "Journal Recording" function via the GUI. When running the job, the GUI will not be available and Fluent **will fail**. 
 
 

docs/source/software/cststudio.rst

@@ -0,0 +1,82 @@
+-----------------------------------------
+CST Studio Suite 
+-----------------------------------------
+
+.. _CST Studio 2022: https://www.ansys.com/
+
+========================================
+CST Studio Suite Status
+========================================
+
+`CST Studio 2022`_ is the current version of the CST Studio Suite installed on the HPC. Currently, Single-Node Module and GPU acceleration are tested and confirmed to be working via a manual CLI call. 
+
+The ``cst_`` job-submission scripts will work, however they *do not* allocate resources in an effective manner for DeepThought. Use at your own risk, as a manually created SLURM job will 
+give you much better results. 
+
+Below are the modes of operation and their status.
+
++----------------------+--------------------+
+| Run Mode             | Status             |
++======================+====================+
+| GPU Accelerated      | Working/Released   |
++----------------------+--------------------+
+| Single-Node Parallel | Working/Released   |
++----------------------+--------------------+
+| Multi-Node Parallel  | Testing/Unreleased |
++----------------------+--------------------+
+
+
+========================================================================
+CST Studio Suite Quickstart Command Line Guide
+========================================================================
+
+To run a CST Studio job on the HPC you will need your ``.cst`` file, and the associated SLURM script. Modify any <OPTIONS> to your requirements.  
+
+1.  Single-Node, GPU Accelerated
+
+    ``cst_design_environment -defaultacc -with-gpu=<NUMBER_OF_GPUS_REQUESTED> -m -r -num-threads=$SLURM_CPUS_PER_TASK -project-file /path/to/projet.cst``
+
+2. Single-Node, CPU Only 
+
+   ``cst_design_environment -m -r -num-threads=$SLURM_CPUS_PER_TASK -project-file /path/to/project.cst``
+
+
+3. Multi-Node GPU Accelerated 
+
+   ``Coming Soon, when MPI is released``
+
+
+4. Multi-Node, CPU Only 
+
+   ``Coming Soon, when MPI is released``
+
+
+
+
+++++++++++++++++++++++++++++++++++++++++++++++
+CST Studio Suite CLI Quick List
+++++++++++++++++++++++++++++++++++++++++++++++
+
++---------------------------------+--------------------------------------------------------------+
+| CLI Option                      | Description                                                  |
++=================================+==============================================================+
+| -defaultacc                     | Enable Compute Acceleration Detection.                       |
++---------------------------------+--------------------------------------------------------------+
+| -with-gpu=<NUM_GPU>             | Per Node, the number of GPU's (Max 2). Requires -defaultacc. |
++---------------------------------+--------------------------------------------------------------+
+| -m -r                           | Enable 'Batch' Mode                                          |
++---------------------------------+--------------------------------------------------------------+
+| -num-threads=<COUNT>            | Number of threads (CPU's) to use                             |
++---------------------------------+--------------------------------------------------------------+
+| -num-cpudevices=<COUNT>         | MPI Only. Use only if you know what NUMA is                  |
++---------------------------------+--------------------------------------------------------------+
+| -withmpi                        | Enable MPI                                                   |
++---------------------------------+--------------------------------------------------------------+
+| -machinefile=/path/to/file      | Path to the CST Machine File. Required for MPI.              |
++---------------------------------+--------------------------------------------------------------+
+| -autompi                        | Automatically detect the MPI Type. Required for MPI.         |
++---------------------------------+--------------------------------------------------------------+
+| -shared-dir                     | A shared space for CST. Defaults to $TMPDIR                  |
++---------------------------------+--------------------------------------------------------------+
+| -project-file /path/to/file.cst | Path to .CST Project FIle                                    |
++---------------------------------+--------------------------------------------------------------+