Hardware

Hazel Linux cluster

The Hazel cluster is a heterogeneous cluster that includes state-of-the-art CPUs, GPUs, and networking architecture while maintaining older resources as long as feasible. The cluster consists of Intel Xeon and AMD EPYC based compute nodes in dual-socket servers, some with attached GPUs.

  • On the order of 400 compute nodes with over 14,000 cores
  • Majority of nodes connected with InfiniBand
  • Several nodes with attached GPUs including A100, H100, H200, L40S, and other models
  • Most nodes have more than 128 GB of memory; standard configuration is now 512 GB with some 1024 GB nodes

See the Compute Resources page for full details on available hardware, queues, and scheduling.

Cluster status monitoring

Real-time availability of resources can be monitored on the cluster status pages:

Storage

Storage is provided at no cost to all HPC accounts:

  • Home directory (/home) — 1 GB per account for source code, scripts, and small executables
  • Scratch space (/share) — 20 TB per project for running applications and working with large data
  • Application storage (/usr/local/usrapps) — 100 GB per project by request, backed up space for installing larger applications and conda environments

See the Storage page for details on directory locations and size limits.

Partner Program

If existing compute resources are inadequate for the needs of a project, there is an opportunity to purchase additional compute nodes or storage under the HPC Partner Program.

Software

Officially supported applications

HPC staff maintain a core set of commonly used applications including commercial, open-source, and community-supported packages. These are available to all users via the module system.

User-installed software

Users can install their own software in their project directories under /usr/local/usrapps. HPC staff will assist with installation questions, but users are asked to first consult the documentation and make efforts to do installs themselves when feasible.

Conda and containers

Conda is available for managing Python environments and installing packages. Apptainer (formerly Singularity) is provided for running containerized applications, which helps with reproducibility and ease of deployment.

Staff

HPC has a team of computational scientists responsible for maintaining the cluster, supporting users, and helping researchers run their applications effectively. Staff services include:

Consulting

HPC staff are available for consultations to help with:

  • Getting started on the cluster
  • Optimizing job scripts and resource requests
  • Porting and scaling applications
  • Software installation and environment setup
  • Troubleshooting job failures and performance issues

Contact HPC to schedule a consultation.

Training

HPC staff provide training workshops and resources for users at all levels:

Documentation

Extensive documentation is available covering all aspects of using the HPC cluster, from logging in to running parallel jobs. See the documentation overview for a full index.

Faculty Oversight

The HPC Faculty Oversight Committee advises on matters related to the HPC service, including acceptable use policy, resource allocation, and strategic planning.