Skip to content

CHPC — Research Computing and Data Support for the University Community

Main Navigation

Getting Started in the CHPC General Environment

Welcome to the Center for High Performance Computing (CHPC) at the University of Utah, the university's central resource for research computing and data support. This guide walks new users through the essential steps for accessing the CHPC's General Environment resources and beginning their work.

  The General Environment is suited for data that does not carry any security restrictions. If you have any questions on whether your data is suited for the CHPC General Environment, please contact the CHPC at helpdesk@chpc.utah.edu

Whether you're new to high-performance computing or just new to using CHPC resources, this guide will help you get started.

On this page

The table of contents requires JavaScript to load.

Accounts

To use CHPC systems, you first need a CHPC account. Principal investigators can request an account and create a group here.

Students, researchers, and other non-PIs create accounts by accepting an invitation from their PI to join a CHPC group.

  You must be provisioned a valid, active uNID (University of Utah ID number) at the University of Utah in order to be provisioned an account at the CHPC. Visit our Accounts page for more information on obtaining a uNID and managing your account.

Compute Resources at the CHPC

There are many compute resources offered to researchers within the CHPC General Environment.

These resources include:

 

Resources You Automatically Have Access To

Receiving a CHPC account automatically grants you access to many of the compute resources listed above, enabling you to automatically begin your research endeavors on CHPC systems.

These resources include:

  • 50Gb of home directory space
  • CHPC compute resources, with priority access to the Kingspeak and Lonepeak clusters
  • High-performance scratch spaces - dedicated storage spaces to run computational jobs.
  • Thousands of research software tools
  • Beehive, a dedicated Windows server with access to Windows-only statistics software
  • OnDemand, a web portal for easy access to CHPC systems; this is a great way to access CHPC resources with little to no HPC experience

Logging In

After creating your account, you can connect to CHPC resources in a variety of ways:

  • Open OnDemand
    • Open OnDemand run in your a browser and provides access to many commonly used graphical applications. It is a good starting point if you are not yet comfortable using a terminal.
  • SSH
    • SSH is a common method for logging into a Linux system from a terminal.
  • FastX
    • FastX provides both remote access to Linux systems and to graphical software. FastX is accessible both from a web browser and an application available for installation.
  • Windows Remote Desktop
    • The Windows server provides a remote Windows desktop for researchers who need Windows-specific applications or prefer working in a Windows environment. The server also provides access to home and group storage.

For additional methods of accessing CHPC resources, visit our Accessing CHPC Resources page.

Storing Files

Most researchers work with three main storage options on CHPC systems:

  • Home directories are available by default. Every CHPC user has 50 GB of home directory space. Research groups may purchase additional home directory space, which includes automatic backups.
      Home directories are not backed up by default. Only purchased home directory space includes backups. More information about purchasing additional home directory space is available on our Storage page.
  • Group storage  is available for purchase and is sold by the terabyte. Backups are also available for purchase.
  • Scratch space is available by default to all users and is intended for intermediate files used while jobs are running. To use the scratch space, simply create a directory under your uNID in one of the available scratch spaces.
  Files are removed from the scratch space if they have not been accessed within 60 days. The scratch space is not backed up, so please do not use the scratch space for permanent storage.

Transferring Files in and out of the CHPC

The CHPC has dedicated data transfer nodes, which can transfer data at a high speed. 

Here are examples of different methods you can use to transfer data through the data transfer nodes:

  • Globusprovides tools for efficient and secure data transfers, enabling parallel, load-balanced, and fault-tolerant data movement.
  • Rclone is a command-line tool that lets you easily transfer and sync files between your computer and various cloud storage services, such as Google Drive, Box, Dropbox, and S3-compatible providers.
  • SCPis a Linux command-line tool used to securely transfer files on CHPC's interactive nodes or on a Data Transfer Node (DTN).
  • OnDemand has built-in file transfer tools, as well as terminal access to the data transfer nodes.

Accessing Software

The CHPC manages thousands of software packages for researchers. Most applications are packaged as modules, allowing you to load and unload software without managing installations and dependencies yourself.

The CHPC also enables researchers to install their own software. Environment management systems such as Miniforge (a replacement for conda), Python Virtual Environments, Apptainer/Singularity, and Spack are all popular tools that can be used to install and manage software.

The CHPC also provides access to a variety of AI tools and AI agents. Claude, Codex, and OpenCode are installed as modules. 

Running your First Job

Slurm is the CHPC's job-scheduling system. All computational workloads are submitted to and managed by Slurm. After logging in, run the mychpc batch command to see which compute resources you can access through Slurm.

Access our Slurm documentation to learn how to request computational resources from our clusters.

  Try using our tool that helps users find which accounts, partitions and qualities of service you can use when submitting jobs on Center for High Performance Computing systems.

Slurm Allocations & Reserved Computing Service

A compute allocation is required to conduct research on CHPC-owned resources on the Granite and Notchpeak clusters. Allocations are free to obtain and only need to be applied for. The older General Environment clusters, Kingspeak and Lonepeak, do not require an allocation.

CHPC offers two types of allocations: Small Allocations and Regular Allocations. Both are requested using the same form. If you do not already have an allocation, we recommend requesting a Small Allocation, which provides 30,000 CPU hours and 300 GPU hours and is granted automatically. Regular Allocations provide up to 300,000 CPU hours and 3,000 GPU hours and may be requested for any quarter within the next year. If you request a Regular Allocation without already having an allocation, you will automatically receive a Small Allocation while your request is under review.

Research groups can also purchase a reserved computing service on CHPC clusters. The reserved computing service would provide the purchasing private access to dedicated compute resources for a period of time that they choose. To inquire about this, please contact the CHPC at helpdesk@chpc.utah.edu.

Training and Support

The CHPC supports researchers with a wide range of academic backgrounds and computing experience. Whether you are an undergraduate beginning your first research project, an experienced faculty researcher, or somewhere in between, our team is here to help.

  The Center for High Performance Computing hosts virtual office hours with its research computing facilitators at 1:00 p.m. MT most Fridays. Please see the events page for details. You are also welcome to reach out to helpdesk@chpc.utah.edu for assistance.

For more information, visit our FAQ page.

Last Updated: 10/9/26