Getting Started on our HPC Systems

To obtain an account (or multiple accounts), the sponsoring Stanford faculty member must send an email request to srcc-support@stanford.edu that includes the name and SUNet ID of each team member who will be using Sherlock.
 
For additional basic information about Sherlock, see www.sherlock.stanford.edu/docs/getting-started/#how-to-request-an-account.

All users must have a Stanford SUNet ID and a Sherlock account to log in to Sherlock. Your Sherlock account uses the same username/password as your SUnet ID. 

Connections to Sherlock are provided via Secure Shell (SSH) login. For step-by-step instructions, see Connecting to Sherlock.

The Research Computing team gives a monthly onboarding presentation that covers Sherlock's layout, scheduler limits, file systems, and job submission/software installation best practices for using our shared High-Performance Computing environment. 

You can find links to attend via Zoom (first Wednesday of the month, 1:00-2:00 pm Pacific Time), access a video of the session on-demand, download the presentation slide deck on the Sherlock Onboarding Session page.

See also: Sherlock Documentation.

See Storage.

A number of methods allow transferring data in/out of Sherlock and other clusters, and managing data within your project’s directory space. You can use SSH-based file transfer commands, such as scp, sftp, or rsync, or the more full-featured Globus tool. For more information and links, see Sherlock Data Transfer.

To learn about running compute jobs on Sherlock, see Submitting Jobs.

See Available Software on the Sherlock website.

Carina is available to any Stanford researcher who has Principal Investigator (PI) privileges or who is part of a team led by a Stanford PI. Carina is exclusively available for research using high-risk data, and each new data set brought into the Carina environment will undergo a Data Risk Assessment.

See Request a Carina Account on the Carina website.

For instructions and links, see Connect to Carina on the Carina website.

For introductory instructions, see Connect to Carina.

In addition, the Research Computing team offers classes on topics that can help you build skills on Carina. Find out what opportunities are coming up on the Classes & Workshops page.

The standard Carina setup includes 1TB of data storage for each PI (Principal Investigator). Additional storage may be purchased in 10TB units for a monthly charge.

A number of methods allow transferring data in/out of Carina and other clusters, and managing data within your project’s directory space. You can use SSH-based file transfer commands, such as scp, sftp, or rsync, or the more full-featured Globus tool. For more information and links, see Tools for Moving and Managing Data.