Title
A Design Methodology for Data-Parallel Applications
Abstract
A methodology for the design and development of data parallel applications and components is presented. Data-parallelism is a well-understood form of parallel computation, yet developing simple applications can involve substantial efforts to express the problem in low-level notations. We describe a process of software development for data-parallel applications starting from high-level specifications, generating repeated refinements of designs to match different architectural models and performance constraints, enabling a development activity with cost-benefit analysis. Primary issues are algorithm choice, correctness, and efficiency, followed by data decomposition, load balancing, and message-passing coordination. Development of a data-parallel multitarget tracking application is used as a case study, showing the progression from high to low-level refinements. We conclude by describing tool support for the process.
Year
DOI
Venue
2000
10.1109/32.844491
IEEE Trans. Software Eng.
Keywords
Field
DocType
data-parallel multitarget tracking application,software development,development activity,parallel computation,index terms keywords software design,multi-target tracking algorithms.,data decomposition,prototyping,software templates,data-parallel application,design methodology,low-level refinement,low-level notation,high-level programming languages,algorithm choice,data-parallel applications,parallel algorithms,data parallel application,concurrent computing,application software,parallel algorithm,process design,software design,cost benefit analysis,indexing terms,message passing,load balance,data parallelism,algorithm design and analysis,parallel programming,load balancing,programming language,parallel computer,resource allocation,parallel processing,high level programming languages
Software design,Parallel algorithm,Computer science,Load balancing (computing),Correctness,Theoretical computer science,Design methods,High-level programming language,Message passing,Software development,Distributed computing
Journal
Volume
Issue
ISSN
26
4
0098-5589
Citations 
PageRank 
References 
6
0.53
23
Authors
4
Name
Order
Citations
PageRank
Lars S. Nyland1596.26
Jan Prins226723.14
Allen Goldberg355878.36
Peter H. Mills4162.78