Free Lunch in In Situ Visualization: Leveraging Idle CPU Resources to Mitigate GPU Contention

Download PDF
Open Webpage

Victor A. Mateevitsi, Andres Sewell, Jens Henrik Göbbert, Mathis Bode, Paul F. Fischer, Joseph A. Insley, Ioannis Kavroulakis, Damaskinos Konioris, Yu-Hsiang Lan, Misun Min, Dimitrios Papageorgiou, Steve Petruzza, Silvio Rizzi, Ananias Tomboulides, Michael E. Papka

Published: 2025, Last Modified: 01 Mar 2026LDAV 2025EveryoneRevisionsBibTeXCC BY-SA 4.0
Abstract: In high-performance computing (HPC), in situ analysis and visualization avoid costly I/O by extracting insight during the simulation run. When these tasks execute on the same GPUs that drive the simulation, resource contention can degrade performance. We propose an asynchronous framework that offloads visualization to idle CPU cores and overlaps in situ work with simulation execution. The framework uses ASCENT for efficient data movement and rendering and applies core pinning. Evaluated with NekRS on Polaris and JUWELS Booster, it reduced end-to-end runtime by 21–40% for slice-based visualizations compared to inline GPU instrumentation. Slice outputs incurred little overhead and overlapped cleanly with the simulation. Heavier filters like isovolumes, multilevel contours, and volume rendering were "free" (i.e., visualization time was encapsulated by simulation time) only at lower node counts, consistent with a CPU-budget model: filter cost and output cadence must fit the cores available per node.