Summary
To access the Great Lakes cluster, you'll need to request a User Login and those steps are found in Phase 1. To run jobs on the cluster you will need to have access to a Slurm Research Computing Account and those may include: Public, Class, Department, UMRCP or Paid accounts. We are hoping that this Knowledge Base article sheds some light on these different types of Slurm Accounts, their limits, billing, and additional HPC resources outside of the University of Michigan.
Body
Summary
To access the Great Lakes cluster, you'll need to request a User Login and those steps are found in Phase 1. To run jobs on the cluster you will need to have access to a Slurm Research Computing Account and those may include: Public, Class, Department, UMRCP or Paid accounts. We are hoping that this Knowledge Base article sheds some light on these different types of Slurm Accounts, their limits, billing, and additional HPC resources outside of the University of Michigan.
As always, if you have questions or concerns while going through the learning material, please don't hesitate to stop by Office Hours (Monday's 2-3pm and Thursday's 3-4pm), our HPC related workshops, which will be linked below, or submit a ticket to arc-support@umich.edu.
"Go forth and compute!" - Dr. Charles Antonelli
Environment
High Performance Computing (HPC) on the Great Lakes cluster (non-sensitive data)
HPC Accounts
- Understanding Slurm Research Computing Accounts, Limits and Billing
- University of Michigan Research Computing Program (UMRCP)
- UMRCP Accounts usually end with a “0” i.e. <PIuniqname0>
- 80,000 compute hours are granted with this account and are renewed each fiscal year on July 1st through June 30th.
Please Note: We always recommend using and exhausting this account before your own shortcode or department funds, because it will save both you and the college some expense. UMRCP renews July 1st each year.
- Who's eligible for UMRCP?
- Review this link to determine eligibility and follow the “Step-by-Step Guide to Enroll” if eligible.
- If you have any questions about which resources you should be requesting, don’t hesitate to reach out to us at arc-support@umich.edu
- You can check what accounts you have access to with the command line tool: my_accounts or from the Accounts drop down menu within Interactive sessions for Open OnDemand.
- LSA Public Accounts
- LSA1: Users of lsa1 can use up to 24 cores or up to 120 GB of memory across all of their running jobs at any point in time.
- LSA2: Each user of lsa2 can run one job at a time, using a single GPU, up to 2 cores, and up to 10 GB RAM for a maximum of 1 day (24 hours).
- LSA3: The lsa3 account is for running jobs on the largemem partition and there is a limit per user of 36 cores and 180GB of memory.
- FAQ and Overview of computing on high-performance clusters for LSA users
- Information about cost-sharing can be found at the link above.
- Class, Group or Department Accounts
- Accounts created for classes, cooperating research groups or departments are resources shared only between users who are granted access.
- Paid Accounts, also known as Private Accounts
Please note: These accounts are paid for by the researcher and limits for the accounts are optional but can be set upon request. Resources are shared only to whoever is specified as an authorized user by the account owner.
- To request a Paid Account to be created please reach out to us at arc-support@umich.edu with the following information:
- A shortcode for billing (an LSA shortcode is eligible for 80% cost-sharing from the college)
- List of account users (people who can submit jobs to the account, thereby charging the account)
- Account limits (primarily a total spending limit per month but other limits are available too)
- Graduate students in particular may want to use Rackham Graduate Student Research Grants to provision their own, private Great Lakes paid accounts.
- What to do when UMRCP compute resources have been consumed?
- If you’re the PI or Project Admin check the Resource Management Portal (RMP) to see how much of the account has been used.
- Under the header “Great Lakes” select the blue hyper-text “<PIuniqname>_root”, then click the blue hyper-text “<PIuniqname>0 and under header “Subaccount Limits” it will show a percentage of compute hours used with a value “X of 80,000 CPU hours used”
- If you are not the PI or a Project Admin, we will need approval from someone that is before creating a Paid Account.
- HPC Billing Cycles for paid accounts are billed at the start of every month
- HPC alternatives outside of the University and Options when working with collaborators who are not part of the University of Michigan
External resources
Links to external documentation are embedded within this documentation.