Getting Efficient DNN Inference Performance: Is It Really About the TOPS?


Date:  Tuesday, September 22, 2020

Start Time: 11:30 am

End Time: 12:00 pm

In this presentation, we will look at how performance is measured among deep learning inference platforms, starting with the simple peak TOPS metric, why it’s used and why it might be misleading. We will look at compute efficiency as measured by real benchmark workload performance and how it relates to peak TOPS, comparing performance across Intel’s inference platforms. We’ll also discuss how developers can use Intel’s DevCloud for the Edge to quickly access Intel’s inference platforms.

There will be a live Q&A session with the presenter immediately following the presentation.

Track

Session Speakers

  • Gary Brown is Director of AI marketing at Intel’s IoT Group, and has most recently overseen the product launches of edge AI products such as Intel Movidius VPUs (Vision Processor Units) and other IoT Vision products that include AI technologies. Gary has over 25 years of experience leading applications engineering and technical marketing teams in vision, imaging and audio. Prior to Intel, he was VP of Marketing at Movidius, and he led product marketing for imaging/video products at Tensilica and Cadence. Early in his career, Gary enjoyed leading functional verification for an HD Video processor at Panasonic Semiconductor in Osaka, Japan. Gary holds a Masters in EE from Stanford and a Bachelors in Engineering from UC Berkeley.


Share

Getting Efficient DNN Inference Performance: Is It Really About the TOPS?