Julien Herrmann
Chargé De Recherches at CNRS - Centre national de la recherche scientifique
Greater Lyon Area France
Summary
👤
Senior🎓
Top SchoolJulien Herrmann is a research scientist with 14 years of experience bridging theoretical computer science and practical high-performance computing, currently serving as Chargé de Recherches at CNRS after postdocs at Georgia Tech, Inria and Ohio State. He holds a PhD from ENS Lyon in numerical algebra and scheduling algorithms and specializes in scheduling techniques, numerical algorithms for distributed systems, graph algorithms for HPC and discrete optimization in machine learning. Julien combines deep algorithmic theory with hands-on implementation experience from GPU and PaRSEC projects to applied research, often focusing on scalable linear-algebra solvers and randomized techniques. Based in the Greater Lyon area, he blends academic teaching experience in parallel algorithms with a track record of international collaborations and internships that shaped his practical focus on performance and scalability. An outsider detail: his early work included genetic-algorithm approaches to graph partitioning, hinting at a longstanding interest in hybrid heuristic-plus-theory solutions.
14 years of coding experience
4 years of employment as a software developer
Doctor of Philosophy (PhD), Numerical Algebra, Algorithmic, Doctor of Philosophy (PhD), Numerical Algebra, Algorithmic at Ecole normale supérieure de LyonMathématics, Physics, Mathématics, Physics at Lycée Aux Lazaristes
French, English, German
Github Skills (21)
dataflow-programming10
computation10
numa10
parsec10
heterogeneous10
nodes10
algebra9
architecture9
multicore9
communications9
asic9
backend9
architectures8
gpu-acceleration8
high-performance8
Programming languages (1)
Github contributions (2)
ICLDisco/dplasma
Mar 2012 - Jun 2013
DPLASMA is a highly optimized, accelerator-aware, implementation of a dense linear algebra package for distributed heterogeneous systems. It is designed to deliver sustained performance for distributed systems where each node featuring multiple sockets of multicore processors, and if available, accelerators, using the PaRSEC runtime as a backend.
Contributions:84 commits in 1 year 3 months
heterogeneous-systemslinear-algebra-libraryprocessorsruntimefastest
ICLDisco/parsec
Mar 2012 - Jun 2013
PaRSEC is a generic framework for architecture aware scheduling and management of micro-tasks on distributed, GPU accelerated, many-core heterogeneous architectures. PaRSEC assigns computation threads to the cores, GPU accelerators, overlaps communications and computations and uses a dynamic, fully-distributed scheduler based on architectural features such as NUMA nodes and algorithmic features such as data reuse.
Contributions:89 commits in 1 year 3 months
runtime-systemschedulerparsecreusethreads
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.