John Mellor-crummey
Professor at Rice University
Houston, Texas, United States
Summary
👤
Senior🎓
Top SchoolJohn Mellor-Crummey is a professor of computer science at Rice University with over three decades of research and 11 years of documented professional experience focused on high-performance parallel computing. He leads development of the HPCToolkit performance tools and has a long-standing research portfolio spanning compilers, runtime systems, correctness tools, and performance modeling for parallel and scientific applications. His work bridges theory and practice—contributing scalable synchronization algorithms, data-race detection and execution replay techniques, and network performance optimization—while also improving real-world tooling such as Dyninst’s binary instrumentation for better DWARF and symbol handling. Based in Houston, he combines deep academic credentials (Ph.D. Rochester, BSE Princeton) with hands-on open-source impact, including OpenMP parallelization and CUDA binary support in Dyninst, reflecting a pragmatic focus on making parallel software measurable, correct, and performant.
11 years of coding experience
12 years of employment as a software developer
Ph.D., Computer Science, Ph.D., Computer Science at University of RochesterBSE, Electrical Engineering & Computer Science, BSE, Electrical Engineering & Computer Science at Princeton University
Github Skills (8)
c-language10
cprogramming-language10
dwarf10
binarydiff10
openmp9
debug8
debugging8
cuda6
Programming languages (9)
Github contributions (5)
dyninst/dyninst
Sep 2017 - Nov 2022
DyninstAPI: Tools for binary instrumentation, analysis, and modification.
Role in this project:
Back-end Developer
Back-end DeveloperContributions:2 reviews, 62 commits, 17 PRs in 5 years 3 months
Contributions summary:John primarily contributed to the DyninstAPI project by improving the handling of debug information and symbol table operations. They addressed issues related to DWARF line reading, including fixes for global symbol module handling and improved iteration. Additionally, the user added support for new binary types, such as EM_CUDA, and implemented OpenMP parallelization within the parsing functionality for performance improvements.
instrumentationbinary-analysisstackwalkermodificationbinary-instrumentation
jmellorcrummey/spack-configs
May 2019 - Oct 2022
memoized spack configs for some DOE systems
Contributions:39 commits, 6 PRs, 32 pushes in 3 years 5 months
configsdoespack
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.