Mikhail Kurnosov
Professor at Rzhanov Institute of Semiconductor Physics Siberian Branch of Russian Academy of Sciences
Novosibirsk, Novosibirsk Oblast, Russia
Summary
🤩
Rockstar🎓
Top SchoolMikhail Kurnosov is a professor and senior researcher with 14+ years of experience specializing in high-performance communication algorithms, parallel programming (MPI, OpenMP, CUDA) and software vectorization. He combines academic leadership—advising bachelors, masters and PhD students—with hands-on systems engineering, designing and administering HPC clusters (InfiniBand, SLURM/TORQUE, NFS) and benchmarking MPI collectives. As a frequent contributor to the flagship Open MPI project, he implements and optimizes collective operations (Reduce, Allreduce, Scan) and hardens algorithmic code paths for production use. His background spans research institutes, university teaching, industry lecturing (Yandex) and a practical early career in educational language and tooling development, giving him a rare blend of pedagogy and production-grade optimization experience. Notably, his work on adding and tuning collective algorithms demonstrates both deep theoretical understanding and measurable performance impact on widely used open-source middleware.
14 years of coding experience
7 years of employment as a software developer
Doctor of Science, Computer Technology/Computer Systems Technology, Doctor of Science, Computer Technology/Computer Systems Technology at Siberian State University of Telecommunications and Information SciencesM.Sc. (Specialist), Mathematics and Computer Science, M.Sc. (Specialist), Mathematics and Computer Science at Gorno-Altaysk State University
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (10)
algorithm10
c1710
data-structures10
algorithms10
mpi10
hpc10
openmpi10
c1110
data-structure10
fortran9
Programming languages (1)
Github contributions (5)
open-mpi/ompi
May 2017 - Nov 2021
Open MPI main development repository
Role in this project:
Back-end Developer & Algorithm Implementer
Back-end Developer & Algorithm ImplementerContributions:1 review, 56 commits, 34 PRs in 4 years 7 months
Contributions summary:Mikhail primarily contributes to the Open MPI project by implementing and optimizing collective communication algorithms, particularly for MPI operations like Reduce and Allreduce. Their work involves adding new algorithms, such as Rabenseifner's and recursive doubling for Scan and Exscan, and improving existing ones like Reduce_scatter_block, emphasizing performance and efficiency within intra-communicator contexts. They also fix potential bugs by adding checks for bounds of sindex[] and rindex[], which can lead to improved code stability. The user's contributions include both algorithm implementation and debugging with the core focus on optimizations.
mpicluster-computingfortranopenmpipetsc
mkurnosov/ompi
May 2017 - Dec 2023
Open MPI main development repository
Contributions:109 pushes, 62 branches in 6 years 8 months
halojavampiswig
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.