The Roofline performance model offers an insightful and intuitive method for extracting the key execution characteristics of HPC applications and comparing them against the performance bounds of modern CPUs and GPUs. Its ability to abstract the complexity of memory hierarchies and identify the most profitable optimization techniques have made Roofline-based analysis increasingly popular in the HPC community. The tutorial will introduce the fundamental aspects behind different Roofline modeling principles as well as providing several practical use case scenarios that highlight their efficacy for application optimization on CPUs and GPUs. This tutorial presents a unique combination of instruction to Roofline, hands-on instruction in using Roofline within Intel, NVIDIA, and AMD production performance tools and discussions of real-world Roofline use cases at ALCF, NERSC and OLCF computing centers. The tutorial presenters have a long history of collaborating on the Roofline model and have presented several Roofline-based tutorials.
Sunday, November 12th, 8:30am-5pm CST
|Welcome and Administration
|Introduction to the Roofline Model
|Software Setup for Hands-on Components
|Session I Q&A and Break
|NVIDIA NSight Compute Roofline Hands-On
|NERSC Application Use Cases
|Session II Q&A and Lunch
|Intel Advisor Roofline Hands-On
|ALCF Application Use Cases
|Session III Q&A and Break
|AMD Omniperf Roofline Tool Hands-on
|OLCF Application Use Cases
|Session IV Q&A
We have created a slack space for tutorial discussions. Join here.
Slides and Materials
The slides for this tutorial are available on our google drive.
Intel Advisor Setup
Please see instructions here
NVIDIA Nsight Setup
Please register or log in to https://courses.nvidia.com/dli-event/ (code provided live in tutorial session)
AMD Roofline Setup
Please see documentation here