Name: Very High Performance/Watt Processors: the Road to Exascale
Start: 2012-12-18T19:00:00+01:00
End: 2012-12-18T22:00:00+01:00
Location: Ecole des Mines de Paris - Salle V116

This talk will deal with most efficient (GFLOPS / Watt) processors existing today:

- talk from an FPGA manufacturer: good opportunity to ask questions on HW design choice, techniques on how to best program a parallel chipset, upcoming features, ..

- talk and demo by Kalray of their chipset + MPPA dev. env. under Eclipse.

- demo by Nvidia of the CARMA (ARM + GPU) (http://www.seco.com/en/item/carma-devkit/) board.

- presenting a map of most power-efficient (GFLOPS / Watt) processors

- detailing some hardware architectures

- demo of a power-efficient chipset over OpenCL

We will give away an Nvidia GPU card, some programming books, and other goodies.

Talks in French except one, slides in English, discussion among participants in French or English.

Note: This is a tentative schedule thus subject to change, but you will be notified by email.

Intelligence Artificielle | Machine Learning | Deep Learning

Technology

Artificial Intelligence

Machine Learning

Data Science

Big Data

High Scalability Computing

Cloud Computing

Distributed Systems

Data Mining

Predictive Analytics

Data Analytics

Chatbots

Natural Language Processing

Very High Performance/Watt Processors: the Road to Exascale

Ecole des Mines de Paris - Salle V116

Partager

Intelligence Artificielle | Machine Learning | Deep Learning

Very High Performance/Watt Processors: the Road to Exascale

Intelligence Artificielle | Machine Learning | Deep Learning

Détails

Les membres s'intéressent également à