{"title":"The STL integrated computer operating system","authors":"Manfred Kory, Paul t Berning","doi":"10.1145/800257.808908","DOIUrl":null,"url":null,"abstract":"The separation of conventional unit record input/output processing from a large general purpose computer system has been and still is a common and accepted practice among many large-scale computer installations. This kind of processing has been called off-line or peripheral processing and seeks to reduce costs by removing the relatively slow I/O functions such as card reading, card punching, and printing from the large computer and having them performed on smaller less expensive computers. This practice is not without its inherent penalties, all of which tend to reduce the operational efficiency of the system by introducing large throughput times and equivalently large setup times. This paper describes a general-purpose computer system currently implemented at STL that returns the conventional data input/output functions to on-line processes. The paper not only includes a general description of the hardware and software employed but also makes comparisons with prior systems, describes simulaion experiments performed, and evaluates current performance.","PeriodicalId":167902,"journal":{"name":"Proceedings of the 1964 19th ACM national conference","volume":"124 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1964-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of the 1964 19th ACM national conference","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/800257.808908","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 2
Abstract
The separation of conventional unit record input/output processing from a large general purpose computer system has been and still is a common and accepted practice among many large-scale computer installations. This kind of processing has been called off-line or peripheral processing and seeks to reduce costs by removing the relatively slow I/O functions such as card reading, card punching, and printing from the large computer and having them performed on smaller less expensive computers. This practice is not without its inherent penalties, all of which tend to reduce the operational efficiency of the system by introducing large throughput times and equivalently large setup times. This paper describes a general-purpose computer system currently implemented at STL that returns the conventional data input/output functions to on-line processes. The paper not only includes a general description of the hardware and software employed but also makes comparisons with prior systems, describes simulaion experiments performed, and evaluates current performance.