I'm a Ph.D. student in Computer Science (HPC concentration) at the University of Tennessee, Knoxville, where I'm a graduate research assistant in the Global Computing Lab with Dr. Michela Taufer. I build performance analysis tools for high-performance computing. Most of my work asks one question: why did this code run the way it did?
- 🔬 Now: building an LLVM pass plugin that exposes compiler optimization remarks to annotation-based profilers such as Caliper, so runtime performance can be traced back to compiler decisions.
- 🏛️ Labs: two summers at Lawrence Livermore National Laboratory (2024, 2025) and one at Los Alamos National Laboratory (2023).
- 🎓 Recognition: UTK Graduate Fellowship · ACM Student Research Competition at SC24 and SC25 · Lead Student Volunteer at SC25.
- 🐧 Off the clock: tinkering with NixOS and Hyprland, writing toy compilers, and slowly learning Rust.
A not-very-scientific flame graph of my time. Wider bars mean more time.
| project | what it does | my role |
|---|---|---|
| Thicket | Python toolkit for exploratory analysis of ensembles of HPC performance profiles | developer · researcher |
| Hatchet | Analyzes calling-context trees from many HPC profilers through one common data model | developer · researcher |
| RAJAPerf | RAJA Performance Suite. I use it to study performance portability across CPUs and GPUs | researcher |
| Compiler provenance | Records which compiler optimizations fired and links them to runtime data in Caliper and Thicket | developer · researcher |
| ANACIN-X | Trace-based analysis of non-deterministic behavior in MPI applications | developer · researcher |
RAJA Performance Suite: Performance Portability Analysis with Caliper and Thicket
O. Pearce, J. Burmark, R. Hornung, B. Bogale, I. Lumsden, M. McKinsey, D. Yokelson, D. Boehme, S. Brink, M. Taufer, T. Scogland
P3HPC @ SC'24· Workshop on Performance, Portability & Productivity in HPC
Towards Affordable Reproducibility Using Scalable Capture and Comparison of Intermediate Multi-Run Results
N. Tan, K. Assogba, W. J. Ashworth, B. T. Bogale, F. Cappello, M. M. Rafique, M. Taufer, B. Nicolae
Middleware '24· 25th ACM/IFIP International Middleware Conference
posters: ACM Student Research Competition
- An Approach for Correlating Compiler Optimizations with Runtime Performance ·
ACM SRC @ SC'25 - Cluster-Based Methodology for Characterizing the Performance of Portable Applications ·
ACM SRC @ SC'24
| when | where | what |
|---|---|---|
| 2024 → now | UTK · Graduate Research Assistant | LLVM pass plugin for compiler remarks; how compilers and -O levels shape application performance |
| summer 2025 | LLNL · Defense Science & Technology Intern | Lightweight runtime provenance for compiler optimizations, integrated with Caliper and Thicket and validated on RAJAPerf |
| summer 2024 | LLNL · Graduate Computing Scholar | Cluster-based method for characterizing portable application performance across CPU and GPU architectures |
| summer 2023 | LANL · Parallel Computing Intern | Parallelized X-ray transport simulations with Kokkos, reaching a >13× speedup over serial |
| 2022 → 2024 | UTK · Undergraduate Research Assistant | Apptainer/Singularity containers for reproducibility, non-determinism in HPC, and dedup-based NN checkpointing |
exit · thanks for stopping by