Skip to Content

This browser is no longer supported.

For the best WWT.com experience, please use one of our supported browsers.

Safari Logo Safari Chrome Logo Chrome Firefox Logo Firefox Edge Logo Edge
Search wwt.com...
Top Page Results

See all search results

Featured Solutions
What's trending
Help Center
Home
Solutions & Services
Solutions
Automation & Orchestration
Cloud
Data Center
Digital
Digital Workspace
ESG
Mobility
Networking
Security Transformation
See all Solutions
See all Solutions
Services
Application Services
ATC Lab Services
Consulting Services
Customer Success
Infrastructure Services
Mergers & Acquisitions
Strategic Resourcing
Supply Chain & Integration
See all Services
See all Services
Industries
Utilities
Financial Services
Global Service Provider
Healthcare
Life Sciences
Manufacturing
Oil & Gas
Public Sector
Retail
See all Industries
See all Industries
Partners
Cisco
Dell Technologies
HPE
NetApp
VMware
f5
Intel
Microsoft
Palo Alto
See all Partners
See all Partners
Learning & Support
ATC
Communities
Events
Labs
Research
About
Footer Links
Careers
Contact Us
Diversity & Inclusion
Locations
News
Sustainability
ATC WWT Research NVIDIA White Paper Data Analytics & AI Digital
WWT Research • Applied Research Report • September 17, 2019

Deep Learning With NVIDIA DGX-1

We highlight some practical considerations for the Deep Learning practitioner relevant to neural network training on the NVIDIA DGX-1.

The NVIDIA DGX-1 is a state-of-the-art integrated system for deep learning and AI development. In this white paper, you will learn the best practices for dramatic acceleration of deep learning algorithms over CPU-based hardware. This includes how the DGX-1 can bring efficiencies to training on batch size, input image size and model complexity.

Abstract

The NVIDIA DGX-1 is a state of the art integrated system for deep learning and AI development. Making use of 8 interconnected NVIDIA Tesla V100 GPUs, the DGX-1 offers dramatic acceleration of deep learning algorithms over CPU-based hardware. In this paper, we highlight a few best practices that enable the DGX-1 end-user to fully capitalize on its industry-leading performance. Benchmark testing was conducted with a common GPU workload, convolutional neural network (CNN) training, using the Keras deep learning API. We first examined the dependence of training efficiency on 3 factors: batch size, input image size and model complexity. Next, the scalability of training speed was assessed using a multi-tower, data-parallel approach. Finally, we demonstrate the importance of learning rate scaling when employing multiple GPU workers.

"WWT Research reports provide in-depth analysis of the latest technology and industry trends, solution comparisons and expert guidance for maturing your organization's capabilities. By logging in or creating a free account you’ll gain access to other reports as well as labs, events and other valuable content."

Thanks for reading. Want to continue?

Log in or create a free account to continue viewing Deep Learning With NVIDIA DGX-1 and access other valuable content.

© World Wide Technology. All Rights Reserved
  • About
  • Careers
  • Locations
  • News
  • Press Kit
  • Contact Us
  • Privacy Policy
  • Acceptable Use Policy
  • Quality
  • Information Security
  • Supplier Management
  • Cookies