DevNation 2016: William Cohen on “Optimizing code for modern processors”
DevNation sneak peek is a behind-the-scenes preview of sessions and information that will take place at DevNation 2016. Sign up for DevNation at www.devnation.org. Learn more. Code more. Share more. Join the Nation.
The difference between processor peak performance and actual delivered performance can be very large. As such, the goal of my DevNation 2016 presentation is to talk about the hardware mechanisms that modern processors use to provide that incredible peak performance such as caches, branch prediction, and superscalar execution. I’ll show show how developers might write code to better exploit those mechanisms, and how to identify problem areas in code.
The presentation isn’t going to instantly make you an expert in optimizing code. However, it provides a good starting point for developers that want their code to run more quickly and efficiently on modern processors.
Performance is a topic that many people care about. I am happy that I am able to help people better understand performance issues by giving presentation and talking with people at DevNation and Red Hat Summit.
Additionally, last year Joe Mario organized a performance expert Bird-of-a-Feather session that allowed people to talk to many people with expertise in a number of different performance areas. This year there is a similar session scheduled for Thursday evening that I am really looking forward to. So many knowledgeable people in one place!
William Cohen has been a developer of performance tools at Red Hat for the past decade (and beyond) and has worked on a number of the performance tools in Red Hat Enterprise Linux and Fedora — tools such as OProfile, PAPI, SystemTap, and Dyninst.