skip to content
Aaron Jarmusch

About Me

I'm a Computer Science Ph.D. student at the University of Delaware researching High Performance Computing (HPC) and parallel programming models in Computational Research Programming Lab (CRPL). My advisor is Dr. Sunita Chandrasekaran.

My PhD research focusses on developing an analytical model for performance modeling for GPU computing technologies.

My research also develops comprehensive testing frameworks for OpenACC and OpenMP GPU offloading across NVIDIA's latest architectures (H100, GH200). I work on automated CI/CD pipelines for LLVM-based compilers and apply LLMs to compiler validation, making HPC technologies more accessible and robust.

Recent News

[August 2025] Attended ATPESC 2025 (Argonne Training Program on Extreme-Scale Computing) at Argonne National Laboratory!
[July 2025] New preprint "Dissecting the NVIDIA Blackwell Architecture with Microbenchmarks" published on arXiv!
[August 2024] Paper accepted to WACCPD at SC24! - LLM-as-a-Judge for Validation and Verification Testsuites available on arXiv!
[August 2024] Paper accepted at IWOMP! - CI/CD framework for OpenMP Offloading validation and benchmarking!

Research & Impact

GPU Computing & Compiler Validation

Developing comprehensive validation and verification frameworks for OpenACC and OpenMP GPU offloading. My work ensures reliable performance across cutting-edge architectures including NVIDIA H100, GH200, and AMD MI300A systems.

Building CI/CD pipeline for LLVM-based compilers

Advancing HPC software sustainability through automated CI/CD pipelines for LLVM-based compilers. Focus on performance optimization, cross-platform compatibility, and systematic benchmarking across diverse computing architectures.

Recent Publications

Visit Google Scholar for a full list

Aaron Jarmusch , Nathan Graddon , Sunita Chandrasekaran (2025). Dissecting the NVIDIA Blackwell Architecture with Microbenchmarks . arXiv preprint . doi: 10.48550/arXiv.2507.10789

GPU ArchitectureNVIDIA BlackwellMicrobenchmarksPerformance Analysis

Aaron Jarmusch , Felipe Cabarcas , Swaroop Pophale , Andrew Kallai , Johannes Doerfert , Luke Peyralans , Sunita Chandrasekaran (2024). CI/CD Efforts for Validation, Verification and Benchmarking OpenMP Implementations . Advancing OpenMP for Future Accelerators: 20th International Workshop on OpenMP, IWOMP 2024 , pages 111-125 . doi: 10.1007/978-3-031-72567-8_8

OpenMPCI/CDGPU ComputingValidation

Zachariah Sollenberger , Jay Patel , Christian Munley , Aaron Jarmusch , Sunita Chandrasekaran (2024). Exploring LLM-as-a-Judge for Validation and Verification Testsuites . SC24-W: Workshops of the International Conference for High Performance Computing, Networking, Storage and Analysis , pages 1885-1893 . doi: 10.1109/SCW63240.2024.00238

LLMCompiler ValidationAutomated TestingAI

Recent Blog Posts

Insights on HPC research, technical tutorials, and industry perspectives