University of Minnesota
Software Engineering Center

You are here

New Trends in Silicon and Hardware - Changes Driven by a New Compute Paradigm

Date of Presentation: 
Wednesday, January 16, 2019
Presented By: 

The huge computing advances made in AI by the Deep Learning revolution of the last 5 years have pushed current hardware to its limits with the venerable CPU, the workhorse of the industry, being forced to run workloads it was not tailored for. This comes at a time when Moore's Law, the driving force of the industry for the past 40 years, has finally started to slow down.

The combination of the above two events has led to a new innovative and creative era in silicon and hardware design, with designs and architectures being considered, researched, and designed all with the singular goal of running AI faster. Parallel Processors, GPUs, reprogrammable FPGA accelerators, custom new devices for both training and inference, liquid cooled systems, and even quantum computers are all being explored to run the AI workloads of tomorrow. This talk will review some of the new ideas and architectures being considered for the future of Cognitive Computing.