Name: Scot Swan
Pronouns: He/him/his
Biography:
I am a computational physicist that originally worked on creating material models for simulating high-speed dynamics of solids (things crashing into each other). Now I mostly work with simulating plasmas. I have been at Sandia National Labs for 6 years. I have a bachelors degree in physics and a masters degree in mechanical engineering. I got my PhD in mechanical engineering from the University of Trento, Italy. I speak English (primary language), Italian, and Russian and have lived in both Italy and Russia for 5.5 years total. My hobbies include assembling and painting scale models (like airplanes and space ships), running, and learning how to play the guitar (but I’m not very good).
Institution/Lab: Sandia National Laboratories
Website:
SRP Collaboration Topic/Title: Data Analysis to Speed Up Supercomputers
Field or research area: Data Analysis, Computer Science, High Performance Computing
Please select all the topical areas that apply to your project:
Computational Science Applications (i.e., bioscience, cosmology, chemistry, environmental science, nanotechnology, climate, etc.); Computer Science (i.e., architectures, compilers/languages, networks, workflow/edge, experiment automation, containers, neuromorphic computing, programming models, operating systems, sustainable software); Data Science (i.e., data analytics, data management & storage systems, visualization); High-Performance Computing
Brief Abstract:
Our work would fall under the HPC/Application Metrology umbrella (“”metrology”” is the study of measuring stuff). So we want to measure the performance of the supercomputer, the performance of the applications that run on them, and how the two interact. I am interested in finding someone to help me speed up our High Performance Computing (HPC) supercomputers. We know that there are bottlenecks in our simulations, like memory, CPU, network, or filesystem limitations. Often these bottlenecks are temporary, so they are hard to find or reproduce. We need someone to run applications on the supercomputers and then analyze the performance data to find performance anomalies and then try to find a cause for them. When we have data showing problems, we can pass that to the system administrators for them to address, speeding up the supercomputer so more simulations can be done faster.
Desired relevant skills, background, or interests:
Interest in learning how to use Linux and the Command Line. Interest in learning how to write Python code to analyze data. Interest in running small-to-large simulations on supercomputers. Interest in working with system administrators and computational physicists.
Other comments:
Do any special requirements apply? Minimum GPA (specify what GPA in comments below); In-Person Only; U.S. Citizen Only
Other, specify:
Keywords:
HPC; Supercomputer; Computational Physics; Performance; Applications; System Administration; Data Analysis
Lightning Talk Title: Supercomputer Performance Variability