GTC ON-DEMAND

 
SEARCH SESSIONS
SEARCH SESSIONS

Search All
 
Refine Results:
 
Year(s)

SOCIAL MEDIA

EMAIL SUBSCRIPTION

 
 

GTC ON-DEMAND

Presentation
Media
Abstract:
We will explore the advantages of combining a model graph that swaps tensors between Volta GPUs with system memory that uses NVLink 2.0 connections between the GPUs and the system cores. GPU memory size can limit model size, image resolution, and batch sizes in neural network training. We'll show how get around those limitations. Our method uses both a graph modification library that adds tensor swap-in/swap-out operations to the graph and NVLink 2.0 connections to system cores and memory to quickly train with models, image resolutions, and batch sizes that were previously impossible. We'll also compare the graph modification module, system architecture, and performance results with standard benchmarks and other models.
We will explore the advantages of combining a model graph that swaps tensors between Volta GPUs with system memory that uses NVLink 2.0 connections between the GPUs and the system cores. GPU memory size can limit model size, image resolution, and batch sizes in neural network training. We'll show how get around those limitations. Our method uses both a graph modification library that adds tensor swap-in/swap-out operations to the graph and NVLink 2.0 connections to system cores and memory to quickly train with models, image resolutions, and batch sizes that were previously impossible. We'll also compare the graph modification module, system architecture, and performance results with standard benchmarks and other models.  Back
 
Topics:
Deep Learning & AI Frameworks, AI & Deep Learning Research, Performance Optimization
Type:
Talk
Event:
GTC Silicon Valley
Year:
2019
Session ID:
S9426
Streaming:
Download:
Share:
 
Abstract:
We will explore what is possible with the unique combination of a model graph that swaps tensors between Volta GPUs and system memory using NVLink 2.0 connections between the GPUs and the system cores. GPU memory size limits the size of models, image resolution, and batch sizes allowed for neural network training. By combining a graph modification library that adds tensor swap-in / swap-out operations to the graph with NVLink 2.0 connections to the system cores and their memory, we can quickly train with models, image resolutions, and batch sizes that were previously impossible. We will review the graph modification module, the system architecture, and the performance results with standard benchmarks and other models.
We will explore what is possible with the unique combination of a model graph that swaps tensors between Volta GPUs and system memory using NVLink 2.0 connections between the GPUs and the system cores. GPU memory size limits the size of models, image resolution, and batch sizes allowed for neural network training. By combining a graph modification library that adds tensor swap-in / swap-out operations to the graph with NVLink 2.0 connections to the system cores and their memory, we can quickly train with models, image resolutions, and batch sizes that were previously impossible. We will review the graph modification module, the system architecture, and the performance results with standard benchmarks and other models.  Back
 
Topics:
Artificial Intelligence and Deep Learning
Type:
Talk
Event:
GTC Europe
Year:
2018
Session ID:
E8336
Streaming:
Download:
Share:
 
 
Previous
  • Amazon Web Services
  • IBM
  • Cisco
  • Dell EMC
  • Hewlett Packard Enterprise
  • Inspur
  • Lenovo
  • SenseTime
  • Supermicro Computers
  • Synnex
  • Autodesk
  • HP
  • Linear Technology
  • MSI Computer Corp.
  • OPTIS
  • PNY
  • SK Hynix
  • vmware
  • Abaco Systems
  • Acceleware Ltd.
  • ASUSTeK COMPUTER INC
  • Cray Inc.
  • Exxact Corporation
  • Flanders - Belgium
  • Google Cloud
  • HTC VIVE
  • Liqid
  • MapD
  • Penguin Computing
  • SAP
  • Sugon
  • Twitter
Next