GEMS X009.1 - HPC for Ag: Upskilling Agri-Food Researchers to Utilize HPC Resources
Description
Are you an agri-food or natural resources researcher with basic-to-intermediate programming skills in R or Python? Do you find yourself limited by processing time, memory constraints, or dataset size when running analyses, simulations, or models? This self-paced course will teach you how to scale up your computational workflows using readily available High Performance Computing (HPC) and Cloud Computing resources.
Who This Course Is For & What You'll Learn: This course is designed for researchers in computationally intensive agri-food and natural resources domains who want to scale their work beyond laptop limitations. Whether you're processing genetic data in biotechnology, analyzing sensor networks in precision agriculture, running climate models, optimizing supply chains, simulating ecosystem dynamics, or conducting forest inventory analysis—you'll learn to move these computationally intensive workflows from your laptop to professional HPC and cloud infrastructure.
Every exercise, example, and quiz question uses real agricultural research problems rather than generic computer science examples, so you'll directly apply scaling techniques to the types of challenges you actually face.
Course Format:
- Self-paced: Complete at your own schedule
- Time commitment: Approximately 13 hours total (9 hours of video lectures + 4 hours of hands-on quizzes)
- Materials: Video lectures, slide presentations, and interactive Jupyter notebooks
- Assessment: Four mastery-based quizzes (retakeable) covering computer hardware, computer science fundamentals, HPC/cloud concepts, and machine learning best practices
Prerequisites:
- Functional programming knowledge in Python or R at a rudimentary level or higher
- Basic familiarity with GIS concepts recommended but not required
What Tools This Course Uses: Throughout the course you will use three main tools to access and interact with compute-to-scale resources:
- Jupyter notebooks: Interactive coding environment for data analysis
- UNIX command line: Text-based interface for operating system commands
- SLURM: Job scheduling system for HPC clusters
If you do not have a basic functional knowledge of these three tools, GEMS offers non-credit courses on all three topics.
**This course is available to current University of Minnesota faculty, staff and students only**
Outline
- Intro to Cloud Computing
- Expanding HPC Skills
- Basic machine learning concepts
- Classic machine learning algorithms
- Deep learning
- Computer Science Concepts
- Algorithm design
- Data structures
- Computational complexity (Big O)
- Profiling
- HPC and Cloud Concepts
- Divide and conquer
- Parallelization
- Containerization
- Storage types
- Computer Hardware and Resources
- Processor types
- Memory
- Networking speed
- Trends
