View in EDS

A task-uncoordinated distributed dataflow model for scalable high performance parallel program execution.

Saved in:
Bibliographic Details
Title: A task-uncoordinated distributed dataflow model for scalable high performance parallel program execution.
Authors: Wilson, Lucas A.1,2 lucaswilson@acm.org, von Ronne, Jeffery1 vonronne@acm.org
Source: Parallel Computing. Jan2016, Vol. 51, p79-87. 9p.
Subject Terms: *PARALLEL programming, *MATHEMATICAL optimization, *PROGRAM transformation, *COMPUTER simulation, *DATA flow computing
Abstract: We propose a distributed dataflow execution model which utilizes a distributed dictionary for data memoization, allowing each parallel task to schedule instructions without direct inter-task coordination. We provide a description of the proposed model, including autonomous dataflow task selection. We also describe a set of optimization strategies which improve overall throughput of stencil programs executed using this model on modern multi-core and vectorized architectures. [ABSTRACT FROM AUTHOR]
Database: Academic Search Index
Description
Abstract:We propose a distributed dataflow execution model which utilizes a distributed dictionary for data memoization, allowing each parallel task to schedule instructions without direct inter-task coordination. We provide a description of the proposed model, including autonomous dataflow task selection. We also describe a set of optimization strategies which improve overall throughput of stencil programs executed using this model on modern multi-core and vectorized architectures. [ABSTRACT FROM AUTHOR]
ISSN:01678191
DOI:10.1016/j.parco.2015.10.013