A Runtime Framework for Regular and Irregular Message-Driven Parallel Applications on GPU Systems

A Runtime Framework for Regular and Irregular Message-Driven Parallel Applications on GPU Systems

The effective use of GPUs for accelerating applications depends on a number of factors including effective asynchronous use of heterogeneous resources, reducing data transfer between CPU and GPU, increasing occupancy of GPU kernels, overlapping data transfers with computations, reducing GPU idling a...

Full description

Bibliographic Details
Main Author: Rengasamy, Vasudevan
Other Authors: Vadhiyar, Sathish
Language:en_US
Published: 2018
Subjects:
Graphics Processing Unit (GPU)
Parallel Programming (Computer Science)
Parallel Programming Models
Parallel Programming Frameworks
Charm++ (Computer Program Language)
HybridAPI-GPU Management Framework
G-Charm Framework
Accelerator Based Computing
Cholesky Factorization
Computer Science
Online Access:http://hdl.handle.net/2005/3193
http://etd.ncsi.iisc.ernet.in/abstracts/4055/G26576-Abs.pdf

Internet

http://hdl.handle.net/2005/3193
http://etd.ncsi.iisc.ernet.in/abstracts/4055/G26576-Abs.pdf

Similar Items