This month
GPUs are in high demand everywhere we turn, and it’s not (just) because folks want to play Half Life 2 at max resolution. It turns out that training ML models for the latest AI applications can be quite expensive, and it’s largely because of this rotten GPU-supply-and-demand predicament. Luckily for us, AWS has been investing in custom silicon and high-perfomance ARM servers to give us cost-effective ways to train our models. In this talk they’ll tell us how to use them to get more bang for our academic bucks.
About this workshop series
UW’s Amazon Web Services (AWS) team cordially invites you to attend a monthly workshop hosted at the eScience Institute. These workshops will consist of a technical component introducing cloud computing concepts, discussing best practices, and an open Q&A session to discuss projects you are working on or issues you are running into.
Participants can attend either online over Zoom, or in person at the WRF Data Science Studio on the 6th floor of the Physics/Astronomy Tower on the UW campus: