Recommended prerequisites:
None
Overview:
High-performance computing (HPC) is here to help you get your work done faster! Whether you're trying to run a parameter sweep across thousands of model runs, processing a dataset that is hundreds of GB in size, fine-tuning an AI/ML model, or running a simulation that distributes its work across tens or hundreds of compute cores, understanding how to leverage HPC will help you get those jobs done.
This workshop will cover the high-level concepts you need to understand to effectively run resource intensive software and code on Rice's high-performance computing (HPC) clusters, including:
- What is HPC? What is a cluster? – Computation at scale
- How do I run my code? – Software environments and making software available on a cluster.
- Where do my files go? – Types of storage and what they are for.
- Why don't my computations run immediately? – Job scheduling on a shared computing system:
- The SLURM scheduler
- Partitions and queues
- Requesting resources (CPU, memory, time, etc.)
- Submitting, managing and monitoring jobs
Contact information:
Please contact researchdata@rice.edu if you have questions about the Data@Rice workshop series.
Date/Time
-
Location
Fondren Library Basement B43A (Collaboration Space)
Registration Form