Events
Webinar: CPU-based Threading Models Application in Python for HPC Computing
Start
2:00 pm (WET)
2025/11/21
End
3:00 pm (WET)
2025/11/21
Location
Online
About the talk:
The webinar covers three main approaches to overcome Python’s Global Interpreter Lock (GIL) limitation: using compiled C/C++/Fortran code with OpenMP or TBB for maximum performance, implementing PyOMP with Numba for HPC users who prefer Python development, and advanced LLVM integration with GIL-disabled Python for professional developers. Each approach offers different performance gains ranging from 6x to over 10x speedup compared to sequential Python, with varying skill requirements from moderate Python knowledge to expert-level LLVM IR understanding. The webinar is designed for HPC practitioners, researchers, and developers who need to implement parallel computing solutions in Python applications.
About the speaker:
Veselin Kolev (Discoverer Supercomputer)
The lecturer has a research scientist background. He holds a PhD in Chemical Physics. He has an extensive knowledge and practice in designing and applying efficient parallel molecular simulation codes. His most recent scientific position was at the University of Southern California, where he was a member of Professor Ariel Warshel’s group (2013 Nobel Laureate in Chemistry). With more than 20 years of Python programming experience (starting with Python 2.0), he has a strong background in scientific computing, data visualization, data management, and managing high-performance computing (HPC) systems with Python. Currently, he works as an HPC systems engineer and develops both commercial and scientific software. Additionally, he participates in several defense projects for developing intelligent autonomous weapon systems with autonomous decision making abilities.
Moderator:
João Barbosa (IT4I)
João Barbosa joined IT4Innovations in September 2024 to lead the High-Level Support Team, specializing in user support, code optimization, and parallel computing. With expertise in high-performance computing (HPC), in-situ scientific visualization, and modern hardware architectures, he collaborates with developers to enhance scientific applications. Previously, he worked at the Texas Advanced Computing Center and the University of Minho, focusing on scalable computing solutions. João actively contributes to scientific communities, codes, and publications and is passionate about advancing computational.
