{"title":"XTP versus TCP/IP in a UNIX/Streams environment","authors":"V. Roca, C. Diot","doi":"10.1109/FTDCS.1993.344173","DOIUrl":"https://doi.org/10.1109/FTDCS.1993.344173","url":null,"abstract":"Discusses whether XTP is a good replacement for TCP. The two are compared based on standard data transfers between UNIX systems using a Streams environment. The authors first modified the KRM implementation of XTP to port it to the Streams. Performance comparisons with TCP showed no significant differences. They modified XTP's acknowledgement strategy to reduce the number of control packets meant to acknowledge received data. It yielded an average 15% speedup. Beyond the protocol comparison, the authors also intend to analyze Streams environment support for communication system implementation.<<ETX>>","PeriodicalId":251095,"journal":{"name":"1993 4th Workshop on Future Trends of Distributed Computing Systems","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1993-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128989483","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Mobility support in open bridged networks","authors":"M. Zitterbart, A. Tantawy","doi":"10.1109/FTDCS.1993.344213","DOIUrl":"https://doi.org/10.1109/FTDCS.1993.344213","url":null,"abstract":"The proliferation of powerful portable computers and the need for uninterruptible access to information impose new requirements on communication services, particularly regarding the provision of mobility support. The authors focus on mobility in an open bridged networking environment, i.e., bridged LANs interconnected by open bridges (OBs) across a public switched networks (such as SMDS and B-ISDN). A simple timer-based mechanism is applied to the management of location information databases in order to efficiently support mobility within such an environment.<<ETX>>","PeriodicalId":251095,"journal":{"name":"1993 4th Workshop on Future Trends of Distributed Computing Systems","volume":"72 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1993-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126522733","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"The network-memory in a global distributed processing system","authors":"S. Covaci, R. Popescu-Zeletin","doi":"10.1109/FTDCS.1993.344203","DOIUrl":"https://doi.org/10.1109/FTDCS.1993.344203","url":null,"abstract":"The authors present two scenarios of a target data processing/communication architecture that alleviate the main identified bottlenecks and pave the way towards a global distributed system supporting multimedia-based applications. The proposed processing/communication paradigm is based on the shared-memory capabilities offered by the optical network. A new structure, referred to as the network-memory, is introduced and used as a building block in the system architecture. There is a need to reconsider the roles played by the network in this new environment, as well as the notion of end-system. Possible solutions for the network-memory and for the end-system architecture are given, and related issues are analyzed.<<ETX>>","PeriodicalId":251095,"journal":{"name":"1993 4th Workshop on Future Trends of Distributed Computing Systems","volume":"11 11","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1993-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121008692","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"A functional approach for expressing temporal properties of inputs and outputs of real-time applications","authors":"P. Raja, G. Noubir","doi":"10.1109/FTDCS.1993.344140","DOIUrl":"https://doi.org/10.1109/FTDCS.1993.344140","url":null,"abstract":"Many real-time applications are different from ordinary computer applications because they react to the state or change of state of the external environment within a specified time limit. The authors concentrate on the interface between such applications and the external environment. The interface is represented as real-time objects, on which operations that accomplish correct data flow from the external environment to the application are introduced. The main aim of this work is to characterize the data flow by specifying the semantics of the operations. A functional approach is adopted for specifying and expressing the semantics of the operations.<<ETX>>","PeriodicalId":251095,"journal":{"name":"1993 4th Workshop on Future Trends of Distributed Computing Systems","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1993-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133923787","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Harmonic network access architecture for ATM","authors":"D. Elias, A. Gavras","doi":"10.1109/FTDCS.1993.344214","DOIUrl":"https://doi.org/10.1109/FTDCS.1993.344214","url":null,"abstract":"With the development of an affordable technology for broadband networks, it becomes interesting to connect terminals directly to these networks, instead of using the broadband network only as backbone. Due to the high bandwidth of future networks, new network access architectures have to be considered. The authors give an idea concerning a strategy for a scalable and efficient B-ISDN access. They call that access harmonic, since the network and the architecture of common workstations are taken into account.<<ETX>>","PeriodicalId":251095,"journal":{"name":"1993 4th Workshop on Future Trends of Distributed Computing Systems","volume":"25 3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1993-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129194931","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"A reactive flow control mechanism in ATM network with end-to-end transport","authors":"H. Kwon, G. Pujolle","doi":"10.1109/FTDCS.1993.344148","DOIUrl":"https://doi.org/10.1109/FTDCS.1993.344148","url":null,"abstract":"A source rate control mechanism called RCT (rate control for end-to-end transport), which reacts on instantaneous internal congestion, is developed. As the propagation delay is relatively long in ATM (asynchronous transfer mode) networks in comparison with transmission time, the feedback information on congestion can be late. To avoid low utilization of communication resources caused by the wrong information, the control scheme uses measured overload period, and it uses only two adjustable transmission rates. One is the allowed peak rate and the other a rate that is for enforcing the transmission rate when there is a congestion in the network. The RCT is a simple flow control mechanism. It does not require any further real time information except the measured overload period and shows acceptable performance when the average overload period is bounded.<<ETX>>","PeriodicalId":251095,"journal":{"name":"1993 4th Workshop on Future Trends of Distributed Computing Systems","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1993-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133103085","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Distributed programming for neural networks","authors":"N. Serbedzija, G. Kock, S. Jahnichen","doi":"10.1109/FTDCS.1993.344166","DOIUrl":"https://doi.org/10.1109/FTDCS.1993.344166","url":null,"abstract":"Presents a high-level approach for parallel and distributed programming of connectionist models. A generic description of an abstract connectionist model is given, providing means for necessary modifications and extensions. A concurrency model supports asynchronous communication among massively interconnected units, and distributed implementation provides a truly parallel and robust execution environment. This presentation covers the design rationales, programming model and implementation details, and is illustrated with concrete examples.<<ETX>>","PeriodicalId":251095,"journal":{"name":"1993 4th Workshop on Future Trends of Distributed Computing Systems","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1993-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116150934","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Completeness of the equational semantics for Basic LOTOS","authors":"M. Massink, L. Rooijakkers","doi":"10.1109/FTDCS.1993.344208","DOIUrl":"https://doi.org/10.1109/FTDCS.1993.344208","url":null,"abstract":"The logical correspondence between the equational semantics of Basic LOTOS and is standard, derivational one is proven. A derivational semantics is traditionally given by means of a set of axioms and deduction rules which define a deduction system. With such semantics, some difficulties arise when dealing with deduction rules with negative premises; also, the proof that a transition cannot take place cannot be carried out within the formal system. On the other hand, in the equational semantics approach, a transition is viewed as the application of a triadic predicate. Such a function is defined by a set of equations, and this, in a natural way, allows for the use of negative information within the system. It is shown that for Basic LOTOS, when restricted to guarded recursion, both formal reasoning systems strongly correspond.<<ETX>>","PeriodicalId":251095,"journal":{"name":"1993 4th Workshop on Future Trends of Distributed Computing Systems","volume":"84 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1993-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124800821","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Distributed computing with parallel networking","authors":"K. Maly, M. Zubair, S. Kelbar","doi":"10.1109/FTDCS.1993.344204","DOIUrl":"https://doi.org/10.1109/FTDCS.1993.344204","url":null,"abstract":"For many large scientific applications, computing on a cluster is a viable, economical alternative to a dedicated parallel machine. Application performance on a cluster is largely determined by the speed of the underlying communication network. The authors use a parallel network approach to improve the communication network performance. More specifically, they use multiple networks based on Ethernet to improve the performance. The authors modified the parallel virtual machine (PVM) environment to take advantage of multiple Ethernets by using round robin scheduling strategy. The modified PVM, referred as PPVM, showed an improvement of 35% over PVM under no background load, and an improvement of 100% under background load on two channels.<<ETX>>","PeriodicalId":251095,"journal":{"name":"1993 4th Workshop on Future Trends of Distributed Computing Systems","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1993-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114938603","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"A distributed object-oriented system with multi-threads of services","authors":"F.-J. Wang, J.-L. Chen, Chung-Hua Hu","doi":"10.1109/FTDCS.1993.344149","DOIUrl":"https://doi.org/10.1109/FTDCS.1993.344149","url":null,"abstract":"Distributed computing systems are becoming more and more popular. Most developing tools and environments for distributed systems were developed with structured programming languages and remote procedure calls (RPCs). The authors present an approach to construct a network-based distributed system, of which each subsystem is object-oriented and supports multiple threads of control. The remote services in the system are based on a client/server paradigm: a server site consists of a group of serving objects and a proxy is generated in a client site to represent a distinct remote serving object. Each serving object may serve (remote) requests concurrently, since it may have one serving thread for each request. With a library of (remote) serving classes, the system can be built in popular object-oriented languages and is allowed to reuse existing components directly. A prototype system is constructed using C++ language and Sun RPCs.<<ETX>>","PeriodicalId":251095,"journal":{"name":"1993 4th Workshop on Future Trends of Distributed Computing Systems","volume":"140 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1993-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115081923","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}