糖心Vlog传媒

Compare the performance of using GPUs vs CPUsGPU

HPC applications already rely on significant parallelism running across multiple CPUs and across multiple nodes. As GPUs become increasingly central to modern supercomputing, many researchers view them as requiring entirely new programming models. This talk reframes GPUs as a natural extension of the parallel structures HPC users already employ. By examining parallel computing through a simple motivating example we show how these concepts can make direct use of GPUs. The goal is to give researchers in chemistry, material science, engineering and even finance, an intuitive framework for understanding when and why GPUs accelerate existing workloads and how to approach them effectively.

Compilation and Submission of GPU Jobs on BigBlue GPU

In an effort to expand our collective knowledge on the use of GPU resources in computational research, this seminar will cover compilation and submission of GPU jobs on our HPC cluster. These topics will also transfer to other resources available to researchers on campus.

Basics on Submission Scripts: SLURM Header and Flags SLURM Jobs Beginner

Brinton Eldridge, PhD candidate in Chemistry and GA for the Research Computing group discusses Basics on Submission Scripts: SLURM Header and Flags. On BigBlue, SLURM serves as our job scheduler, and it is in charge of efficiently allocating resources across the cluster. This talk will provide a, hopefully, comprehensive walkthrough of the key headers in SLURM submission scripts. You might have noticed them at the top of your submission files: they start with `--SBATCH`. We will talk about critical resource-requesting flags such as those for CPU cores, memory, time limits, and partition specifications. Additionally, we鈥檒l explain best practices for optimization and common pitfalls to avoid. This talk will be perfect for those of you who have always used the same headers. Understanding what each header is used for is vital if you ever change your scripts or need your job scheduled sooner.

HPC Basics Beginner HPC Linux

Brian will go through some of the fundamentals of concepts needed to use the HPC. He will cover some of the basics of using the Linux command line, submitting several different types of jobs to the cluster, finding metadata on jobs, and using BigBlueWeb.

Version Control for Researchers using GitHub GitHub Reproducibility Beginner

Brian Wentzloff from the Research Computing group will discusses version control for researchers using Github. Much modern research is built on code, data, and collaboration. But too often, our work lives in files named final_final_really_final_v3.sh. In this talk, we鈥檒l explore how Git and GitHub can serve as a lab notebook for your code, capturing every change, every experiment, and every result in a way that鈥檚 searchable, shareable, and reproducible. We鈥檒l look at common problems like lost provenance, overwritten scripts, and vanishing student projects, and show how version control offers a simple, powerful foundation for scientific integrity. Whether you鈥檙e managing pipelines, container definitions, or paper drafts, Git and GitHub can bring order, transparency, and longevity to your research.

An Introduction to Containers: How to set up an environment to run your software - October 31, 2025 Containers Beginner HPC

Eric Spangler demonstrates 鈥淎n Introduction to Containers: How to set up an environment to run your software". This applies to many of the most popular software titles in advanced computing including Conda/MiniConda, AlphaFold, NVIDIA GPU Containers, NAMD, LAMMPS, Git-pulled software, and many more using Singularity, our primary container software (similar to Docker, but for HPCs). It will show the portability of containers and how to move between your computer, BigBlue, iTiger, cloud providers or wherever you might want to run them.

LLMs and Generative AI on iTiger GPU Cluster - September 26, 2025 AI GPU Beginner

Mayira Sharif, a Computer Science student, presents a seminar titled 鈥淐rash Course into LLMs and other Generative AI in the iTiger GPU Cluster鈥�