Dr Johan Barthelemy1, Dr Wei Fang1, Dr Gabriel Noaje1
1Nvidia, Australia
Biography:
After his PhD in Applied Mathematics at the University of Namur (Belgium), Dr. Johan Barthélemy joined the SMART Infrastructure Facility of the University of Wollongong (Australia) where he was a Senior Lecturer and the head of the Digital Living Lab researching AIoT solutions for smart cities and environmental monitoring in extreme conditions such as Antarctica. Passionate about applied AI and how to accelerate it, he is now part of the Strategic Researchers Engagement team at NVIDIA, helping scientists to build the next generation of AIs.
Dr. Wei Fang joined NVIDIA in 2021 as a Solutions Architect for High-Performance Computing (HPC) and Artificial Intelligence (AI). In this role, he actively engages with NVIDIA customers across the Asia Pacific South region, focusing on the Higher Education & Research and Supercomputing sectors. Dr. Fang is dedicated to promoting the adoption of NVIDIA's latest technologies to deliver large-scale HPC and AI solutions, driving innovation, and enhancing research productivity. Before his tenure at NVIDIA, Dr. Fang served as an HPC Specialist at Intersect Australia from 2013 to 2021, where he managed Intersect’s HPC services and delivered cutting-edge HPC and cloud solutions to Australian universities. He also worked as a Cloud Specialist at the National Computational Infrastructure (NCI) in 2013 and as a Satellite Ground-station Software Engineer at Geoscience Australia in 2012, where he developed scientific platforms and software solutions. Prior to that, he contributed to the private sector by developing software and algorithms for telecom network optimization. Dr. Fang earned his Ph.D. in Electrical Engineering from Nanyang Technological University, Singapore, in 2004.
Gabriel Noaje has more than 15 years of experience in accelerated computing solutions for High Performance Computing and Artificial Intelligence. He has performed a variety of roles both in the enterprise and public sector that allowed him to manage, design and work on state of the art HPC solutions across a broad range of industries and domains. In his current role at NVIDIA, Gabriel is spearheading the development of HPC business in Asia Pacific.
Prior to joining NVIDIA, he was a Senior Solutions Architect with SGI and HPE where he has worked closely with major supercomputing centers in APAC. Previously, he was a Senior Computational Scientist at A*STAR Computational Resource Centre in Singapore (A*CRC) supporting users with deploying their applications on GPUs and large HPC systems. Gabriel holds a PhD in Computer Sciences from the University of Reims Champagne-Ardenne, France and a BSc and MSc in Computer Sciences from the Polytechnic University of Bucharest, Romania.
Abstract:
This workshop explores how to use Numba—the just-in-time, type-specializing Python function compiler—to accelerate Python programs to run on massively parallel NVIDIA GPUs. You will learn how to: Use Numba to compile CUDA kernels from NumPy universal functions (ufuncs); Use Numba to create and launch custom CUDA kernels; Apply key GPU memory management techniques.
At the conclusion of the workshop, you will be able to use Numba to compile and launch CUDA kernels to accelerate your Python applications on NVIDIA GPUs. You will have an understanding of the fundamental tools and techniques for GPU-accelerated Python applications with CUDA and Numba:
• GPU-accelerate NumPy ufuncs with a few lines of code.
• Configure code parallelization using the CUDA thread hierarchy.
• Write custom CUDA device kernels for maximum performance and flexibility.
• Use memory coalescing and on-device shared memory to increase CUDA kernel bandwidth.
The following topics and technologies will be covered:
• CUDA Python with Numba;
• CUDA programming general practices.