Abstract | ||
---|---|---|
Data-parallelism is a relatively well-understood form of parallel computation, yet developing simple applications can involve substantial efforts to express the problem in low-level data-parallel notations. We describe a process of software development for data-parallel applications starting from high-level specifications, generating repeated re- finements of designs to match different architectural models and performance constraints, supporting 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 appli- cation 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 |
---|---|---|
1996 | 10.1007/3-540-61626-8_18 | Euro-Par, Vol. I |
Keywords | Field | DocType |
data-parallel applications,refinement methodology,message passing,cost benefit analysis,parallel computer,load balance,software development | Notation,Load balancing (computing),Computer science,Correctness,Parallel computing,Data decomposition,Software development,Distributed computing | Conference |
ISBN | Citations | PageRank |
3-540-61626-8 | 2 | 0.36 |
References | Authors | |
9 | 6 |
Name | Order | Citations | PageRank |
---|---|---|---|
Lars S. Nyland | 1 | 59 | 6.26 |
Jan Prins | 2 | 267 | 23.14 |
Allen Goldberg | 3 | 558 | 78.36 |
Peter H. Mills | 4 | 16 | 2.78 |
John H. Reif | 5 | 4180 | 810.75 |
Robert A. Wagner | 6 | 1783 | 802.73 |