1 of 33

Introduction to the Minnesota Supercomputing Institute

Benjamin Lynch, Ph.D.

Director, MSI

Minnesota Supercomputing Institute

© 2025 Regents of the University of Minnesota. All rights reserved.

2 of 33

Overview

  • Mission
  • Organization
  • Resources and Services
  • Eligibility, Access, and Accounts

Minnesota Supercomputing Institute

© 2025 Regents of the University of Minnesota. All rights reserved.

3 of 33

Minnesota Supercomputing Institute

  • Founded in 1984 to provide advanced research computing infrastructure and expertise to the State of Minnesota in order to advance and accelerate research and foster innovation and discoveries through advanced computing technologies, scientific computing, and informatics.

Minnesota Supercomputing Institute

© 2025 Regents of the University of Minnesota. All rights reserved.

4 of 33

Are you a new faculty member?

z.umn.edu/msiaccess

Minnesota Supercomputing Institute

© 2025 Regents of the University of Minnesota. All rights reserved.

5 of 33

Research Groups

Social Science

Physics

Other

Mathematical Sciences

Informatics

Health Sciences

Veterinary Medicine

Agriculture

Astronomy

Biochemistry

Computer

Science

Earth Sciences

Economics

Chemistry

Biology

Engineering

Genetics

1041 Groups

122 Departments

5409 Users

Minnesota Supercomputing Institute

© 2025 Regents of the University of Minnesota. All rights reserved.

6 of 33

Minnesota Supercomputing Institute

(Ben Lynch)

U-Spatial

(Len Kne)

Research Computing�Administration

(Brian Carlton)

Research Computing

(Jim Wilgenbusch)

Research Informatics

Outreach and Training

(Laura Fingerson)

Data Science Initiative

(Haley Borck)

International Institute for Biosensing

(Patricia Khashayar)

Operations and Development

Research Computing

Minnesota Supercomputing Institute

© 2025 Regents of the University of Minnesota. All rights reserved.

7 of 33

Minnesota Supercomputing Institute

  • 58 full time employees and 4 students
  • > 1000 active Faculty groups using MSI resources
  • ~5000 active users
  • Most resources and expertise are provided as a common good to University of Minnesota faculty and their groups.

Minnesota Supercomputing Institute

© 2025 Regents of the University of Minnesota. All rights reserved.

8 of 33

High-performance Hardware and Software

  • >80,000 processor cores
  • 448 GPUs
  • 30 Petabytes of high-performance storage
  • 523 Software Packages
  • Automated testing

Minnesota Supercomputing Institute

© 2025 Regents of the University of Minnesota. All rights reserved.

9 of 33

Remote Interfaces

Minnesota Supercomputing Institute

© 2025 Regents of the University of Minnesota. All rights reserved.

10 of 33

Research Informatics

Advanced Support

Collaborations

Partnerships

Services

Minnesota Supercomputing Institute

© 2025 Regents of the University of Minnesota. All rights reserved.

11 of 33

High Performance Computing (HPC)

  • Highly parallel computing environment with high-speed network connections, graphics processing units, high memory, and high-performance storage
  • batch HPC and interactive HPC
  • HPC access controlled by the SLURM Fairshare scheduler

Minnesota Supercomputing Institute

© 2025 Regents of the University of Minnesota. All rights reserved.

12 of 33

Linux

  • Unix was developed at AT&T Bell Labs in the 1970’s.

  • Linux was developed by Linus Torvalds in the 1990’s, and is now developed by a community as a free operating system.

  • Linux is: free, stable, extensible, and fast.

  • Unix/Linux are the dominant OSes for large HPC systems.

Minnesota Supercomputing Institute

© 2025 Regents of the University of Minnesota. All rights reserved.

13 of 33

Batch HPC

  • Using batch scheduling is the primary way to make use of MSI’s shared supercomputing resources.
  • You will submit jobs to the cluster scheduler, and your analysis eventually runs on the computer without any further intervention.
    • Requires jobs to be written as shell scripts and submitted via the command line.
  • Jobs are queued and priority is based on a combination of the size of the request and how many jobs your group has run recently.
  • Jobs will run when resources are available.

Minnesota Supercomputing Institute

© 2025 Regents of the University of Minnesota. All rights reserved.

14 of 33

Example batch SLURM script

#!/bin/bash -l

#SBATCH --time=8:00:00

#SBATCH --ntasks=4

#SBATCH --mem=20g

#SBATCH --gres=gpu:a100:2

#SBATCH --partition=msigpu,some-other-partition

#SBATCH --mail-type=ALL

#SBATCH --mail-user=sample_email@umn.edu

#

cd ~/program_directory

module load (...software modules you need….)

Command_to_run_your_application (cmd line options)

Example for GPU usage

Minnesota Supercomputing Institute

© 2025 Regents of the University of Minnesota. All rights reserved.

15 of 33

Rightsize Your Resource Request

Time until execution

Minnesota Supercomputing Institute

© 2025 Regents of the University of Minnesota. All rights reserved.

16 of 33

Rightsize Your Resource Request

Does it going to use these resources?

Time until execution

Minnesota Supercomputing Institute

© 2025 Regents of the University of Minnesota. All rights reserved.

17 of 33

Rightsize Your Resource Request

If so, it’s going to have to wait.

Time until execution

Minnesota Supercomputing Institute

© 2025 Regents of the University of Minnesota. All rights reserved.

18 of 33

Rightsize Your Resource Request

If it’s needs fewer resources, it will start faster!

Time until execution

Minnesota Supercomputing Institute

© 2025 Regents of the University of Minnesota. All rights reserved.

19 of 33

Interactive HPC

  • Types of interactive HPC offered:
    • srun: SLURM batch method to get an interactive shell on HPC. Partition to use is called “interactive”
    • Open OnDemand: Web browser interface with access to local and cluster files, custom applications, Jupyter, or a terminal window. ondemand.msi.umn.edu

Minnesota Supercomputing Institute

© 2025 Regents of the University of Minnesota. All rights reserved.

20 of 33

Open OnDemand

Minnesota Supercomputing Institute

© 2025 Regents of the University of Minnesota. All rights reserved.

21 of 33

Software Resources

  • Scientific and numerical software packages are built into “modules” and available to be loaded as you need them.
  • Common software tools used in many fields are available on our systems
    • Genetics
    • Proteomics
    • Image processing
    • Computational chemistry
    • … and many more
  • 500+ software applications msi.umn.edu/software
  • We can work with your research groups to install software that you need
    • Contact the help desk!

Minnesota Supercomputing Institute

© 2025 Regents of the University of Minnesota. All rights reserved.

22 of 33

Data Storage

Primary Storage Tier-2 Tape

Minnesota Supercomputing Institute

© 2025 Regents of the University of Minnesota. All rights reserved.

23 of 33

Primary Storage

  • Primary storage is where you will do most of your work
  • Watch your disk usage:
    • If you are up against file number or disk usage quota, your entire group cannot make new files
  • Important directories in your primary storage space:��/users/<some number>/Your_userID�Your home directory. Store scripts and small files here��/home/groupname/shared�PI group shared directory. Store shared software, common datasets, reference databases, etc, here.

Minnesota Supercomputing Institute

© 2025 Regents of the University of Minnesota. All rights reserved.

24 of 33

Scratch Storage

  • Use it for intermediate/temporary files in analytical workflows
  • Files are kept for 30 days, so be sure to copy important files to your group directory
  • Scratch space is not backed up

  • Location:��/scratch.global/

Minnesota Supercomputing Institute

© 2025 Regents of the University of Minnesota. All rights reserved.

25 of 33

Second Tier Storage

  • Second tier (Tier-2) storage is separate from primary storage - you must use special tools to access it.
  • From the command line, you can use the Amazon “s3cmd” software package to interact with your storage:
  • For a graphical interface, you can use Globus
  • Not backed up, so if you delete a file, it will be gone for good
    • Quotas are 120 TB per research group, typically under the PI’s userID

Minnesota Supercomputing Institute

© 2025 Regents of the University of Minnesota. All rights reserved.

26 of 33

Data Storage Allocation

  • Default storage allocation: 150GB/group primary storage
  • Requests of up to 5TB:
    • No justification or review required
  • For requests of 5 TB–20 TB:
    • Justification
    • Review by MSI staff
  • ANNUAL renewals for all non-default storage allocations
    • If you haven’t made active use of the storage you’ve been allocated, you’ll be asked to cut down on your primary storage allocation.

Minnesota Supercomputing Institute

© 2025 Regents of the University of Minnesota. All rights reserved.

27 of 33

Service Costs

  • Free for UMN faculty and their associates
  • Free for other MN academic institutions
  • Free for expert consultation
  • Fee for data storage over allocations limits
  • Fee for extended expert collaboration
  • Fee for service for external organizations (any organization that is not MN higher-ed)

Minnesota Supercomputing Institute

© 2025 Regents of the University of Minnesota. All rights reserved.

28 of 33

PI Group Structure

  • PI: Controls everything

  • Group Administrators: Users within the group designated to do everything a PI can do except add other Group Administrators�
  • Users: Added by PI, usually lab members (postdocs, techs, undergrads, grad students)
    • Users may be part of multiple accounts if they perform analysis for multiple PIs
    • The PI is a user, too, and their SU and disk usage counts the same as their users’

Minnesota Supercomputing Institute

© 2025 Regents of the University of Minnesota. All rights reserved.

29 of 33

Accounts and Renewals

  • UMN faculty members, department staff, and researchers at other MN institutions for higher education are eligible for MSI accounts

  • PI accounts are renewed annually
    • Renewal period starts in late October
    • Extra storage allocations all must be checked and justified

Minnesota Supercomputing Institute

© 2025 Regents of the University of Minnesota. All rights reserved.

30 of 33

Research Support�Services

  • MSI has ~33 consulting staff
  • The majority of consultants have an advanced degree
  • We provide >20 workshops and tutorials annually
  • Informational sessions for proposal development
  • Expertise including:
    • Research Computing; Informatics
    • Big Data Analysis
    • Parallel Algorithm Optimization and Development
    • Custom Analysis Pipeline Development
    • Genomics, Chemistry
    • Application Development

Minnesota Supercomputing Institute

© 2025 Regents of the University of Minnesota. All rights reserved.

31 of 33

Training and Tutorials

  • We host tutorials on a range of topics�https://msi.umn.edu/tutorials
  • Introductory tutorials
    • Intro to MSI
    • Intro to Linux
    • Intro to Job Submission
  • Bioinformatics tutorials
    • Single Cell Genomics
    • RNAseq data analysis
  • Advanced tutorials
    • Parallelization
    • Python for scientific computing
  • https://www.youtube.com/@UofMMSI

Minnesota Supercomputing Institute

© 2025 Regents of the University of Minnesota. All rights reserved.

32 of 33

Summary

  • High Performance Computing and Consulting
  • Free access for researchers
  • Online access through ood.msi.umn.edu
  • Additional dedicated resources if you have a budget
  • help@msi.umn.edu if you have any questions

Minnesota Supercomputing Institute

© 2025 Regents of the University of Minnesota. All rights reserved.

33 of 33

Questions?

  • MSI Help Desk
    • Monday-Tuesday-Thursday in person Walk-in help is available in Walter Library 587. 9:00 AM – 4:00 PM
    • Wednesday & Friday, schedule a virtual appointment to meet with helpdesk staff by visiting the page below and selecting a time block
    • https://www.msi.umn.edu/content/helpdesk-video-consultations

Minnesota Supercomputing Institute

© 2025 Regents of the University of Minnesota. All rights reserved.