YoVDO

Sustainable Development of Stencil-based HPC Applications - JuliaCon 2024

Offered By: The Julia Programming Language via YouTube

Tags

Julia Courses Parallel Computing Courses High Performance Computing Courses GPU Computing Courses

Course Description

Overview

Save Big on Coursera Plus. 7,000+ courses at $160 off. Limited Time Only!
Explore a successful approach for sustainable development of stencil-based High Performance Computing (HPC) applications in this JuliaCon 2024 conference talk. Learn how the three "P"s - Performance, Portability, and Productivity - are addressed through the implementation of HPC building blocks using Julia packages ParallelStencil.jl, ImplicitGlobalGrid.jl, and CellArrays.jl. Discover how numerical algorithms can be formulated with architecture-agnostic, math-close code, leading to productivity similar to classical prototyping environments. Understand how these HPC building blocks transform codes into massively scalable high-performance multi-GPU/CPU applications or frameworks. Gain insights into automatic performance optimization, parallelization, optimized memory layouts, and GPU-aware distributed parallelization. Examine the extensibility of math-close notation, integration with the Julia ecosystem, and compatibility with Julia's extension feature. Review performance and scaling benchmarks from real-world applications on leading supercomputers, demonstrating near-optimal performance and scaling on thousands of GPUs.

Syllabus

Sustainable Development of Stencil-based HPC Applications | Räss, Omlin, Utkin | JuliaCon 2024


Taught by

The Julia Programming Language

Related Courses

Моделирование биологических молекул на GPU (Biomolecular modeling on GPU)
Moscow Institute of Physics and Technology via Coursera
Practical Deep Learning For Coders
fast.ai via Independent
GPU Architectures And Programming
Indian Institute of Technology, Kharagpur via Swayam
Perform Real-Time Object Detection with YOLOv3
Coursera Project Network via Coursera
Getting Started with PyTorch
Coursera Project Network via Coursera