Math HPC

Math high performance computing resources

Resources available

Description of Math HPC resources
Hostname (add OS GPU CPU RAM
mathhpavw1202a Red Hat 8 1 X NVIDIA Ampere A100 (80 GB) Intel Xeon 6258R@2.7GHz (112 threads) 754 GB
mathhpavw1202b Red Hat 8 8 X NVIDIA RTX A6000 (48 GB) AMD EPYC 7763@2.45Ghz (256 threads) 4 TB
mathhpavw1202c Red Hat 8 8 X NVIDIA RTX A6000 (48 GB) AMD EPYC 7763@2.45Ghz (256 threads) 2 TB
mathwkmth2207e Ubuntu 20.04 1 X NVIDIA Quadro GV100 (32 GB) Intel Xeon E5-2620@2.4GHz (12 threads) 125 GB
lambdaquad Ubuntu 22.04 4 X NVIDIA GeForce RTX 2080Ti (11 GB) Intel Core i9-9820@3.3Ghz (20 threads) 125 GB

Requesting access

  • Email to request access.
  • If you aren't listed as having a Math affiliation in the campus directory, you need to include who you are working with in Math.

Using the servers

  • Connecting to the servers
  • Storage
    • For mathhpavw1202a, b, and c there is a significant amount of local storage that should be available to each user under /export/<username>.  This storage is local to each server meaning you can't see the storage on one from any of the others.  This local storage is not backed up and should be used as temporary space you need during your computations.  These servers also connect to a file server that provides a small amount of home space that is accessible from any TerpConnect server so it can be seen from all 3 of these servers as well as and
    • If you run out of quota in your home space which is on the file server, leverage the space mentioned above.
      • You can check your quota on the file server by running "fs examine ~<username>"
      • For example, if you are trying to install something like miniconda3, you can do the folllowing to leverage the local space in /export which will create a symbolic link to redirect anything written to your home space in ~<username>/miniconda3 to /export/<username>/miniconda3.  See issues below if you need assistance with this.
      • ln -s /export/<username>/miniconda3 ~<username>/miniconda3
  • Long running jobs
    • For jobs that you think will run over 24 hours, don't use your home space.  Use the temporary /export/<username> directory to store input and output files.  The file server where your home space is located expires access every 24 hours for security reasons.
    • If you end your ssh session, anything you have running will terminate as well.  There are multiple ways to handle this but we recommend using "screen" or "tmux". 
  • Access software
    • If you run a command to execute a piece of software and the command isn't found, you can use the module command to see if the software is available using that method.


Contact, call 301-405-1634, or stop by Martin Hall 0123 for assistance including requests for any software installations.


Zaratan is available, but Math has not made any investments. There are some free allocation options available.

Keywordshigh performance computing   Doc ID131324
OwnerJeff M.GroupUMD Engineering IT
Created2023-09-08 10:24:46Updated2024-06-12 15:03:44
SitesUniversity of Maryland Engineering IT
Feedback  0   0