Stanford Seminar - NVIDIA GPU Computing: A Journey from PC Gaming to Deep Learning

EE380: Computer Systems Colloquium Seminar
NVIDIA GPU Computing: A Journey from PC Gaming to Deep Learning
Speaker: Stuart Oberman, NVIDIA
Deep Learning and GPU Computing are now being deployed across many industries, helping to solve big data problems ranging from computer vision and natural language-processing to self-driving cars. At the heart of these solutions is the NVIDIA GPU, providing the computing power to both train these massive deep neural networks as well as efficiently provide inference and implementation of those networks. But how did the GPU get to this point?
In this talk I will present a personal perspective and some lessons learned during the GPU's journey and evolution from being the heart of the PC gaming platform, to today also powering the world's largest datacenters and supercomputers.
About the Speaker:
Stuart Oberman is Vice President of GPU ASIC Engineering at NVIDIA. Since 2002, he has contributed to the design and verification of seven GPU architectures.
He currently directs multiple GPU design and verification teams. He previously worked at AMD, where he was an architect of the 3DNow! multimedia instruction set and the Athlon floating-point unit.
Stuart earned the BS degree in electrical engineering from the University of Iowa, and the MS and PhD degrees in electrical engineering from Stanford University, where he performed research in the Stanford Architecture and Arithmetic Group. He has coauthored one book and more than 20 technical papers. He holds more than 55 granted US patents.
For more information about this seminar and its speaker, you can visit ee380.stanford.edu/Abstracts/1...
Support for the Stanford Colloquium on Computer Systems Seminar Series provided by the Stanford Computer Forum.
Colloquium on Computer Systems Seminar Series (EE380) presents the current research in design, implementation, analysis, and use of computer systems. Topics range from integrated circuits to operating systems and programming languages. It is free and open to the public, with new lectures each week.
Learn more: bit.ly/WinYX5

[@bennettwinters7278]

It's incredible that this stuff is actually happening.

[@Bestietvcute]

nice summary presentation, thanks for uploading!

[@Interestingworld4567]

AWESOME SPEECH 🤗🤗🤗🤗

[@PamirTea]

Great talk, thank you for uploading.

[@rahaf.azabeen1676]

ك

[@ben6]

57:25 Comparing the E5-2690 (from 2012) to a SMX (released on Mar 27th, 2018). Classic NVIDIA.

[@ky8920]

v4 is released in 2016

[@Sun4Niebieskieoczy]

As if Heaven knew how much humanity needs help to finally liberate itself from oppression, and empowers the amazing Nvidia team with their visionary thinking and products - and they ride in on a White Horse - delivering the tools which open the entrance into... and will continue to assist in that most virtuous of endeavors...

[@sczoot6285]

HUH?

[@TheLeontheking]

6:50

[@user-eb6xb7ol5t]

請各國前往台灣會議

[@nooheesa5376]

Active Guy, 1.5 Mhz not 1150 Ghz

[@user-eb6xb7ol5t]

和我說話

[@TheLeontheking]

such a dumb question at around 50 minutes (how far are we from being able to implement deep-learning using a single gpu). Simple Answer is we are more than able to do so, the question is just how much time you want to have the system running, and the time needed obviously decreases when you use multiple gpus.