Academic Compute Cluster
- Mark Bober
About
The Engineering Compute Cluster (ENGR Cluster) is a medium-scale installation of heterogeneous compute nodes, supporting both common computational tasks, specific Engineering software applications, and specialized nodes for specific research group tasks.
Your class syllabus or assignment document may go into greater detail about how to access these resources for the particular needs of the class.
All faculty, staff, and WUSTL students enrolled in any McKelvey Engineering course are granted access to this resource. Due to how WUSTL processes student data, access may be granted the day after enrollment to a course is processed.
Service Support
Student who are experiencing issues with an assignment should contact thier instructor or teaching assistant. All technical issues can be reported to support@seas.wustl.edu.
Accessing the Academic Compute Cluster
All password prompts for logging into systems in ENGR Cluster are for your WUSTL Key.
Interactive Terminal (SSH) Access
The ENGR Cluster can be accessed via SSH via these hosts:
shell.engr.wustl.edu
shell2.engr.wustl.edu
The hosts above are not meant for academic work. They are access hosts. You may not be able to use some applications on these hosts, or, their performance may be less than expected or ideal.
Interactive Compute Jobs
The script qlogin will start a session on an academic compute host. To use, type qlogin. You may optionally request a number of CPU cores to be allocated, up to 8, as below:
[student@shell.engr ~]$ qlogin -h
Usage: qlogin -c <cores> [-h]
-c, --cpu ? Request a session with ? cores, up to 8.
-h, --help This message.qlogin-gcc will start an interactive session on an academic compute host with a single GPU available. You may optionally request an amount of GPU RAM you wish to use, in specific increments, as below:
Requesting GPU RAM reserves, but does not limit, your GPU RAM usage. It is your responsibility to make sure your code or data stays within your request. Your code may fail if it uses more.
A job requesting 20GB of GPU RAM., the maximum available on the GPUs available, will request exclusive access to the GPU. It may wait longer to start.
[student@shell.engr ~]$ qlogin-gpu -h
Usage: qlogin -g <gpuram> [-h]
-g, --gpuram ? Request a session with a specific amount of GPU RAM available,
from these valid choices:
-g 2 2 GB of RAM
-g 4 4 GB of RAM
-g 10 10 GB of RAM
-g 20 20 GB of RAM
Other values will be rounded to the nearest valid amount.
-h, --help This message.Interactive Application Access
You also may access the Open Ondemand interface at:
https://linuxlab.engr.wustl.edu
There are a number of applications available from this interface under the Interactive Apps or My Interactive Sessions menu choices.. They all run on a compute node, and while they are running you can connect/disconnect from them in your browser at will.
Academic Linux Desktop
This will start a Linux desktop in your web browser. This is a basic Rocky Linux 8 installation, with the majority of development software installed. You may also run several Engineering academic applications from here - you may type the command module avail to see the list, then module add ? to add one of the choices to your environment.
Please see the list of applications below for more about each available.
Other Applications
Other interactive applications, or applications available from within the Academic Linux Desktop, are defined via the links below:
Interactive Applications
Engineering Academic Applications