Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
This technical presentation explores the critical role of thread affinity in optimizing MPI/OpenMP job performance, explaining how improper configuration can silently degrade performance, especially when porting applications to new architectures like aarch64 from x86_64. Learn how bad thread affinity impacts job performance through a practical demonstration of Linaro Forge's Thread Affinity Advisor, which helps identify and resolve affinity issues to maximize performance on HPC systems. The presentation includes a worked example showing how to effectively use this tool to optimize thread distribution across cores and achieve peak performance on new computing environments.