PowerBI Data Analyst - Create visualizations and dashboards from scratch
AI, Data Science & Cloud Certificates from Google, IBM & Meta
Overview
AI, Data Science & Cloud Certificates from Google, IBM & Meta — 40% Off
One plan covers every Professional Certificate on Coursera. 40% off Coursera Plus Annual.
Unlock All Certificates
Learn how modern C++ standards have evolved to support high-performance computing applications in this seminar presentation by Dr. Tom Deakin from the University of Bristol. Discover the parallel programming capabilities introduced in C++17 and subsequent standards, including built-in parallel patterns for shared-memory and accelerated computing that can serve as alternatives to traditional HPC APIs like MPI and OpenMP. Explore recent C++ features specifically relevant to HPC development, such as standard parallel algorithms, mdspan for multidimensional array access, linear algebra libraries (linalg), and execution policies. Examine real-world applications and research findings from Dr. Deakin's group on implementing these modern C++ features in HPC contexts. Understand the current limitations of C++ for HPC applications and gain insights into the future direction of standards-based parallel programming, moving beyond the traditional approach of adding parallelism through external frameworks to leveraging native language capabilities.
Syllabus
C++ is (nearly) all you need for HPC
Taught by
NHR@FAU