Paper Info
Title
DART-CUDA: A PGAS Runtime System for Multi-GPU Systems
Abstract
The Partitioned Global Address Space (PGAS) approach is a promising programming model in high performance parallel computing that combines the advantages of distributed memory systems and shared memory systems. The PGAS model has been used on a variety of hardware platforms in the form of PGAS programming languages like Unified Parallel C (UPC), Chapel and Fortress. However, in spite of the increasing adoption in distributed and shared memory systems, the extension of the PGAS model to accelerator platforms is still not well supported. To exploit the immense computational power of multi-GPU systems, this work is concerned with the design and implementation of a Partitioned Global Address Space model for multi-GPU systems. Several issues related to the combination of logically separate GPU memories on multiple graphic cards are addressed. Furthermore, the execution model of modern GPU architectures is studied and a task creation mechanism with load balancing is proposed. Our work is implemented in the context of the DASH project, a C++ template library that realizes PGAS semantics through operator overloading. Experimental results suggest promising performance of the design and its implementation.
Year
DOI
Venue
2015
10.1109/ISPDC.2015.20
2015 14th International Symposium on Parallel and Distributed Computing
Keywords
Field
DocType
PGAS,Partitioned Global Address Space,MultiGPU systems,CUDA,Heterogeneous computing
Computer science,CUDA,Real-time computing,Partitioned global address space,Distributed computing,Runtime system,Computer architecture,Operator overloading,Unified Parallel C,Programming paradigm,Shared memory,Parallel computing,Execution model
Conference
ISSN
ISBN
Citations 
2379-5352
978-1-4673-7147-6
2
PageRank 
References 
Authors
0.39
6
2
Name
Order
Citations
PageRank
Lei Zhou1253.91
Karl Fürlinger2151.61