GTC ON-DEMAND

 
SEARCH SESSIONS
SEARCH SESSIONS

Search All
 
Refine Results:
 
Year(s)

SOCIAL MEDIA

EMAIL SUBSCRIPTION

 
 

GTC ON-DEMAND

Presentation
Media
Abstract:

Learn how a successful implementation of a low memory footprint, multi-GPU iterative method makes it possible to efficiently resolve localization of spontaneous nonlinear flow in deforming porous media. Grasping this physical process is essential to ensure safe underground waste storage and understand natural fluid migration in reservoirs. We'll describe our parallel, matrix-free solver design, which provides a short time to solution and can solve a variety of coupled and nonlinear systems of partial differential equations in 3D. We will unveil the key algorithmic and optimization concepts that enable our stencil-based solvers to converge in few iterations, while tackling the hardware limit on the most recent NVIDIA high-bandwidth GPU accelerators. Also, we will explain how we achieved 98 percent parallel efficiency on 5000 NVIDIA Tesla P100 GPUs on the hybrid Cray XC-50 Piz Daint supercomputer at the Swiss National Supercomputing Centre, CSCS.

Learn how a successful implementation of a low memory footprint, multi-GPU iterative method makes it possible to efficiently resolve localization of spontaneous nonlinear flow in deforming porous media. Grasping this physical process is essential to ensure safe underground waste storage and understand natural fluid migration in reservoirs. We'll describe our parallel, matrix-free solver design, which provides a short time to solution and can solve a variety of coupled and nonlinear systems of partial differential equations in 3D. We will unveil the key algorithmic and optimization concepts that enable our stencil-based solvers to converge in few iterations, while tackling the hardware limit on the most recent NVIDIA high-bandwidth GPU accelerators. Also, we will explain how we achieved 98 percent parallel efficiency on 5000 NVIDIA Tesla P100 GPUs on the hybrid Cray XC-50 Piz Daint supercomputer at the Swiss National Supercomputing Centre, CSCS.

  Back
 
Topics:
Computational Fluid Dynamics, HPC and Supercomputing
Type:
Talk
Event:
GTC Silicon Valley
Year:
2019
Session ID:
S9368
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