Overview:

• Practical introduction to high-performance computing for life-science and biotechnology research
• Getting on and getting around a cluster: accounts, login nodes, file systems, and data transfer
• Submitting, monitoring, and scaling analyses with a job scheduler
• Running research workflows that do not fit on a laptop — genomics, imaging, simulation, and machine learning

Lecture Topics:

• What HPC is, and when a problem actually needs it
• Cluster architecture: nodes, cores, memory, storage tiers, and queues
• The command line and remote access essentials
• Job scripts and schedulers: requesting the right resources
• Software environments: modules, containers, and reproducibility
• Parallelism in practice: array jobs, multithreading, and GPUs
• Moving and managing research-scale data
• Benchmarking, troubleshooting, and good citizenship on shared systems

Hands-On Topics:

• Connecting to the cluster and navigating the file system
• Writing and submitting a first batch job
• Running a bioinformatics pipeline at scale as an array job
• Building and running a containerized analysis environment
• GPU-accelerated analysis of a research dataset
• Capstone: port a workflow of your own onto the cluster