No Travel? No Problem.

Remote Participation
Enabling Cache-Aware Roofline Analysis with Portable Hardware Counter Metrics
Event Type
Online Only
Accelerator-based Architectures
Computational Science
Emerging Technologies
Extreme Scale Comptuing
File Systems and I/O
Heterogeneous Systems
Parallel Programming Languages and Models
Scientific Computing
Software Engineering
Registration Categories
TimeMonday, 15 November 20212:40pm - 3pm CST
DescriptionIn this paper, we seek to provide guidance on how to empirically collect the information required to plot application points on Intel’s Cascade Lake Xeon processor and Fujitsu’s A64FX ARM-based processor. Understanding how to process this information is vital to use the Roofline for empirical performance analysis on these systems. This report is designed to enable users to collect data for these systems and lay the groundwork for us to create a portable method for empirical Roofline-based performance analysis. We do so by presenting our methods and several examples of them in use.
Back To Top Button