{"title":"Performance of a multiprocessor for Applicative programs","authors":"D. Grit, R. Page","doi":"10.1145/800199.806162","DOIUrl":null,"url":null,"abstract":"Applicative programming Languages provide opportunities for parallel processing without requiring the programmer to be concerned with explicit synchronization of portions of the computation. We present a computational model of a multiprocessor which executes applicative programs, and we analyze the expected performance of the model via simulation. As the number of processors is doubled, elapsed execution time is nearly halved, until system bottlenecks occur. An alternative model is proposed which alleviates these bottlenecks. The basis of the second model is an interconnection switch which is characterized by log(n) access time and n log(n) cost.","PeriodicalId":32394,"journal":{"name":"Performance","volume":"16 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"1980-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Performance","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/800199.806162","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 3
Abstract
Applicative programming Languages provide opportunities for parallel processing without requiring the programmer to be concerned with explicit synchronization of portions of the computation. We present a computational model of a multiprocessor which executes applicative programs, and we analyze the expected performance of the model via simulation. As the number of processors is doubled, elapsed execution time is nearly halved, until system bottlenecks occur. An alternative model is proposed which alleviates these bottlenecks. The basis of the second model is an interconnection switch which is characterized by log(n) access time and n log(n) cost.