GTC ON-DEMAND

 
SEARCH SESSIONS
SEARCH SESSIONS

Search All
 
Refine Results:
 
Year(s)

SOCIAL MEDIA

EMAIL SUBSCRIPTION

 
 

GTC ON-DEMAND

Data Center & Cloud Infrastructure
Presentation
Media
Improving GPU Utilization for Multi-tenant Deep Learning Workloads on DGX-2 and Public GPU Clouds
Abstract:
Learn our solutions for increasing GPU resource utilization on an on-premise DGX-2 node and public clouds. In this talk we present our operational experiences of a set of multi-tenant deep learning workloads selected through an open competition. To host them we use and extend the Backend.AI framework as the resource and computation manager. While tailored for both educational and research-oriented workloads, it offers a topology-aware multi-GPU resource scheduler combined with fractional GPU scaling implemented via API-level CUDA virtualization, achieving higher GPU utilization compared to vanilla setups.
 
Topics:
Data Center & Cloud Infrastructure, GPU Virtualization, Deep Learning & AI Frameworks
Type:
Talk
Event:
GTC Silicon Valley
Year:
2019
Session ID:
S9406
Download:
Share: