Learner Profiles
Who Should Take This Workshop?
This workshop is designed for researchers new to high-performance computing. Below are three representative learner profiles. Do you see yourself in one of these?
Profile 1: Maya - Graduate Student, New to Computational Research
Background
Maya is a first-year graduate student in molecular biology. Her research involves analyzing genomic sequences and running phylogenetic reconstructions. Until now, she’s used her laptop for data analysis, which has been slow for large datasets.
What She Knows
- Comfortable with basic computer use
- Some experience with Python (learned for a class)
- Has never used a command line extensively
- Not familiar with clusters or supercomputers
Why She Needs HPC
- Her datasets are growing: 50+ GB genomic sequences
- Current analysis takes 2-3 weeks on her laptop
- Needs to run hundreds of sequence comparisons
- Her advisor suggested she learn to use Sagehen HPC
Her Goals
- Get her analysis running 10x faster
- Understand basic job submission
- Learn where to save large datasets
- Eventually run parallel analyses
Profile 2: Dr. James - Faculty Researcher, Transitioning to Computational Methods
Background
Dr. James is an associate professor in chemistry. His lab has traditionally used bench experiments, but he’s starting a new computational chemistry initiative studying molecular reaction mechanisms. He’s hired a postdoc to help develop computational methods.
What He Knows
- Experienced scientist; comfortable learning new tools
- Limited command line experience (occasional Terminal use on Mac)
- Familiar with scientific software (but mostly GUIs)
- Knows research computing exists but never used it
Why He Needs HPC
- Quantum chemistry simulations require massive computation
- Single simulations take weeks on a workstation
- Wants to run parameter sweeps (100+ variations)
- Needs stable, reliable infrastructure for grant-funded work
His Goals
- Set up infrastructure for his lab’s computational work
- Understand what students/postdoc will need
- Know how to request resources and budget time appropriately
- Learn enough to oversee research computing decisions
Profile 3: Alex - Undergraduate Researcher, Data Science Project
Background
Alex is a junior physics/math major who’s enthusiastic about data science. They’re working on an independent study analyzing climate datasets for their senior thesis. They’re taking some computer science courses and are eager to learn computational tools.
What He Knows
- Comfortable with Python (learned in CS courses)
- Some experience with Linux (on their own laptop)
- Knows about cloud computing but hasn’t used HPC
- High learning ability and motivation
Why He Needs HPC
- Climate dataset is 200+ GB (won’t fit on laptop)
- Analysis includes machine learning that’s computationally intensive
- Wants to do more complex analysis than time allows on a laptop
- Plans to feature this work in graduate school applications
His Goals
- Run computationally intensive machine learning
- Learn industry-standard HPC workflows
- Get experience on real computing infrastructure
- Complete thesis analysis in reasonable timeframe
Using These Profiles
During teaching: - “This is like what Maya is doing…” - “Remember, Dr. James might see this from a PI perspective…” - “Alex is excited about optimization…”
When explaining concepts: - Reference their specific challenges - Use relevant examples for each domain
For pacing: - All three profiles represent different pace/depth needs - Provide supplementary exercises for learners like Alex - Offer pre-workshop prep for learners like Maya - Explain “why” for learners like Dr. James
Common Threads
Despite different backgrounds, all three learners share:
- Problem to solve: Something their current computer can’t handle
- Limited HPC experience: First time using a cluster
- Research domain knowledge: They know their field well
- Motivation: Each has a real reason to learn this
- Learning style: Prefer practical, hands-on examples
Teaching to Diverse Learner Types
What Works Well
- Concrete examples: “This genomic analysis took 3 weeks on a laptop, 3 hours on Sagehen”
- Multiple approaches: Show both command line and OnDemand web portal
- Real scenarios: Base challenges on actual research (not toy problems)
- Time to practice: Let them try with their own problems
- Encouragement: “Your first job will work; most issues are simple typos”
Pacing Tips
- For enthusiastic learners like Alex: Have stretch exercises, advanced topics to explore
- For busy learners like Dr. James: Provide executive summary of key concepts
- For anxious learners like Maya: More practice time, explicit confidence-building
- For all: Validate that their anxiety/excitement is normal and healthy
Real vs. Hypothetical
As instructors, whenever possible: - Use real examples from Pomona research - Share actual success stories (with permission) - Show real Sagehen output (not fake) - Discuss real resource constraints and decisions - Celebrate real learner achievements (“Your first job!”)
After the Workshop
These learners won’t become HPC experts after 4 hours. They will:
- Understand what HPC is and why it matters
- Successfully connect and submit a job
- Know where to find help
- Be confident enough to continue learning
Many will return for advanced workshops like: - Workshop 9: SLURM Job Scheduling (more job control) - GPU Computing Workshop (if relevant) - Parallel Programming (if they need it) - Advanced Module Usage (for complex software stacks)
Your role is to get them started confidently, not make them experts.