{"title":"XTP/KRM implementation on a transputer network","authors":"C. Diot, V. Roca","doi":"10.1109/LCN.1991.208080","DOIUrl":null,"url":null,"abstract":"XTP is a real time transfer protocol designed to be implemented in a dedicated hardware environment. To obtain a transputer based parallel implementation, the authors modified a standard implementation (based on UNIX BSD system) on three aspects: the structure of the implementation, interfaces with host, and system primitive references. This new implementation was modelled. Performance was analyzed. This paper tries, in two sections to distinguish what in the performance is linked to the protocol specification and what is the consequence of the implementation. For this purpose, the KRM is compared to a high performance implementation of OSI TP4 the authors had previously modelled. XTP high performance conditions of implementation (including software techniques and multi-processor host architectures) are discussed.<<ETX>>","PeriodicalId":163373,"journal":{"name":"[1991] Proceedings 16th Conference on Local Computer Networks","volume":"23 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1991-10-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"6","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"[1991] Proceedings 16th Conference on Local Computer Networks","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/LCN.1991.208080","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 6
Abstract
XTP is a real time transfer protocol designed to be implemented in a dedicated hardware environment. To obtain a transputer based parallel implementation, the authors modified a standard implementation (based on UNIX BSD system) on three aspects: the structure of the implementation, interfaces with host, and system primitive references. This new implementation was modelled. Performance was analyzed. This paper tries, in two sections to distinguish what in the performance is linked to the protocol specification and what is the consequence of the implementation. For this purpose, the KRM is compared to a high performance implementation of OSI TP4 the authors had previously modelled. XTP high performance conditions of implementation (including software techniques and multi-processor host architectures) are discussed.<>