Discovery

Discovery is the New Mexico State University High Performance Computing Cluster. The cluster is a collection of computers, or nodes, that communicate using InfiniBand, making it an ideal location to scale computational analysis from your personal computer. The high performance computing (HPC) cluster is free to use for current students, faculty, and staff for both research, and teaching. Discovery is equipped with Lenovo nx360, SR650, SR670, SR850, Dell R440, R640, and R740 hardware. It has 54 compute nodes(38 CPU nodes and 16 GPU nodes), with total of 1,536 cores. The operating system is Red Hat Enterprise Linux release 8 (RHEL 8), and it runs Slurm scheduling manager.

NMSU’s Information and Communication Technologies (ICT) provides various ways to assist your research and classroom activities. ICT provides the cyber-infrastructure that’s needed for cluster computing, while the compute resources are expanded and owned by members of the NMSU community. The condo model removes the stress of administering larger computational equipment from researchers and encourages investing in a communal resource.

At the moment, ICT is only looking to expand the current system using similar equipment.

If the noise from your system is distracting you or your students, please be aware that ICT has a user-access room that’s free for use. Your machine will reside in a secure room (you will have key-card access), under stable conditions, with power, and fire-suppression systems. You will be asked to administer to your machine, but this allows for a quieter working environment for all in your space.

If you have machines that you would like to have administered by ICT, you can contact the administrators at hpc-team@nmsu.edu to discuss your needs or fill out the contact form. There will be a yearly cost associated with the administration of a pre-existing machine.

If you would like to donate your machine, please contact the administrators at hpc-team@nmsu.edu or fill out the contact form. It’s likely that the machine can’t be absorbed into the campus cluster, but it can be discussed.

Refer to the above documentation link to learn more.