LC Hotline: 2-4531

From offsite: 925-422-4531

 

Hours

Monday–Friday
8am–12pm, 1–4:45pm
B453 R1103 | Q-clearance area

Nvidia Tools and Unified Memory Workshop

April 11, 2017
 

Date / Time This event has already occurred.  Page is archival.
Description

Max Katz, Steve Rennich and Peng Wang from Nvidia will be presenting a workshop on how to use Nvidia performance tools, how Unified Memory works today on Pascal, and how CUDA and unified memory will be evolving in the Volta time frame.  See agenda below for content.

The content in these presentations will be aimed at people already using GPU platforms, who want to understand more about how to profile their code, how the unified memory system works and the changes coming in Volta.  If you are looking for a more basic getting started workshop please consider our Coral EA Getting Started Workshop.
 
We are able to accommodate remote participants for all talks.  If you are interested please send an e-mail to: karlin1@llnl.gov for webex link.
 
In addition, we have time for developers to bring their code and get hands on help using the Nvidia tools from 2-5 pm on Tuesday.  To book a time please e-mail Ian Karlin (karlin1@llnl.gov).  This time can also be used for more in-depth conversations.

Agenda 9:30-9:35     Welcome and Overview
9:35-10:30   NVIDIA Profilers
10:30-11:15   Unified Memory on P100
11:15-11:30     Break
11:30-12:30  Beyond CUDA 8 & Pascal (Contains NDA Material)
Location Building 453, Armadillo Room
This is a Property Protection Area. Foreign national temporary escorted building access procedures apply.
Audience This is for people already using GPU platforms.  Also, if you are not covered by the CORAL NDA, you will be asked to leave after the second talk.
Fee No cost
Registration No registration is required.
Questions? Please send e-mail to Ian Karlin

 

 

LLNL-WEB-728319