GTC ON-DEMAND

 
SEARCH SESSIONS
SEARCH SESSIONS

Search All
 
Refine Results:
 
Year(s)

SOCIAL MEDIA

EMAIL SUBSCRIPTION

 
 

GTC ON-DEMAND

GPU Virtualization
Presentation
Media
Ask the Expert: Accelerated DataScience workloads on vGPU
Abstract:

Today, containers are an easy way to deploy GPU accelerated libraries on development environment. RAPIDS is a set of DataScience libraries accelerated on GPU available today to a data scientist within an organization. Deploying RAPIDS on cloud in containers hepls democratizing access to accelerated computing. Deploying these containers in data centers on vGPUs help maximize hardware utilization and optimize budget-driven capital expenditures by sharing hardware resources among a team of users. With GPU-accelerated CUDA-X AI, data scientists can realize value from insights faster than with CPUs systems.

 
Topics:
GPU Virtualization
Type:
Talk
Event:
VMWorld
Year:
2019
Session ID:
VM9050
Share: