Stephanie Brink
Computer Scientist
COMP / CASC DIV / Center for Applied Scientific Computing Division
Biography
Stephanie Brink is a computer scientist in the Center for Applied Scientific Computing at Lawrence Livermore National Laboratory. Her research focuses on power-constrained high performance computing and parallel performance analysis tools.
Stephanie received M.S. and Ph.D. degrees in Computer Science from the University of Oregon in 2016 and 2019, and a B.S. degree in Computer Engineering from the University of the Pacific in 2013. At UO, Stephanie was a member of the CDUX Group led by Hank Childs. Her dissertation work focused on power/performance tradeoffs for scientific visualization workloads on supercomputers.
Software Projects
See Stephanie's GitHub profile for a full list.
- Hatchet: Tool for manipulating call trees in Pandas dataframes.
- Variorum: Vendor-neutral library for exposing low-level control and monitoring of a system’s underlying hardware features, such as power management and performance metrics.
- msr-safe: Kernel module providing whitelisted access to model-specific registers (MSRs).
CV
In progress.
Publications
2020 1 publication
- Stephanie Brink, Ian Lumsden, Connor Scully-Allison, Katy Williams, Olga Pearce, Todd Gamblin, Michela Taufer, Katherine E Isaacs, Abhinav Bhatele. Usability and Performance Improvements in Hatchet. In IEEE/ACM International Workshop on HPC User Support Tools (HUST) and Workshop on Programming and Performance Visualization Tools (ProTools), Virtual Conference, November 2020.
2019 2 publications
- Abhinav Bhatele, Stephanie Brink, and Todd Gamblin. Hatchet: Pruning the Overgrowth in Parallel Profiles. In Proceedings of Supercomputing 2019 (SC19), Denver, Colorado, USA, November 2019.
- Stephanie Labasan, Matthew Larsen, Hank Childs, and Barry Rountree. Power and Performance Tradeoffs for Visualization Algorithms. In Proceedings of IEEE International Parallel and Distributed Processing Symposium (IPDPS), Rio de Janeiro, Brazil, May 2019.
2017 1 publication
- Stephanie Labasan, Matthew Larsen, Hank Childs, and Barry Rountree. PaViz: A Power-Adaptive Framework for Optimizing Visualization Performance. In Proceedings of Eurographics Symposium on Parallel Graphics and Visualization (EGPGV), Barcelona, Spain, June 2017.
2015 2 publications
- Stephanie Labasan, Matthew Larsen, and Hank Childs. Exploring Tradeoffs Between Power and Performance for a Scientific Visualization Algorithm. In Proceedings of IEEE Symposium on Large Data Analysis and Visualization (LDAV), Chicago, IL, October 2015.
- M. Larsen, S. Labasan, P. Navrátil, J.S. Meredith, and H. Childs. Volume Rendering Via Data-Parallel Primitives. In Eurographics Symposium on Parallel Graphics and Visualization (EGPGV), Cagliari, Sardinia, Italy, May 2015.