This free workshop series is open to all UT students, faculty, and staff.
| Workshop Title | Date |
|---|---|
| Orientation to ISAAC-NG | Friday, Jan 26 |
| Git | Friday, Feb 2 |
| Project Organization and Data Management | Friday, Mar 1 |
| Slurm Jobs: Scripts, Submissions, and Job Requirements | Friday, Apr 5 |
| Slurm: GPU Computing | Friday, May 3 |
In this workshop, we will cover how to efficiently use the compute resources of ISAAC-NG via the command line and submit, monitor, and manage the computational jobs for production work using the SLURM scheduler.
Git is a distributed version control system that tracks changes in files. It is the industry-standard tool for collaboratively developing software or computational workflows. Its goals include speed, data integrity, and support for distributed, non-linear workflows (thousands of parallel branches running on different computers).
Data organization and documentation are crucial for any research project. A well-organized project ensures efficient analysis and reproducibility, facilitates collaboration, and helps prepare data for publication. This lesson outlines practical strategies for organizing and documenting research projects, including experimental metadata, sequencing information, and bioinformatics workflows.
Estimating compute resources for a Slurm job depends on workload complexity, data type, and desired performance. We will discuss job requirements, including node selection, CPU cores, memory, and wall time. We cover Slurm commands to gather node, job, and historical information.
GPU computing offers significant performance advantages but also introduces additional considerations. Ensure application and library compatibility with specific GPU architectures. Understand parallel programming concepts for effective GPU utilization.
You can find a full list of all workshops offered by OIT HPSC here.
Explore
Write
Chat
Call