Showing papers on "Distributed algorithm published in 1978"
TL;DR: In this article, the concept of one event happening before another in a distributed system is examined, and a distributed algorithm is given for synchronizing a system of logical clocks which can be used to totally order the events.
Abstract: The concept of one event happening before another in a distributed system is examined, and is shown to define a partial ordering of the events. A distributed algorithm is given for synchronizing a system of logical clocks which can be used to totally order the events. The use of the total ordering is illustrated with a method for solving synchronization problems. The algorithm is then specialized for synchronizing physical clocks, and a bound is derived on how far out of synchrony the clocks can become.
6,804 citations
TL;DR: This paper is an attempt to reverse the trend of words in the lexicon of the computer professional becoming cliches through over-use, losing much of their original meaning in the process.
Abstract: Words have only one purpose in a technical context–the transmission of information. When they fail to do that, they lead to confusion and misunderstanding. "Distributed data processing" and "distributed processing" are two phrases which illustrate that axiom. Like many other words in the lexicon of the computer professional, these have become cliches through over-use, losing much of their original meaning in the process. This paper is an attempt to reverse that trend.
260 citations
31 May 1978
TL;DR: A new algorithm for retrieving and updating data from a distributed relational data base that can optimize separately for two models of a communication network representing respectively ARPANET and ETHERNET like networks is presented.
Abstract: In this paper we present a new algorithm for retrieving and updating data from a distributed relational data base. Within such a data base, any number of relations can be distributed over any number of sites. Moreover, a user supplied distribution criteria can optionally be used to specify what site a tuple belongs to.The algorithm is an efficient way to process any query by "breaking" the qualification into separate "pieces" using a few simple heuristics. The cost criteria considered are minimum response time and minimum communications traffic. In addition, the algorithm can optimize separately for two models of a communication network representing respectively ARPANET and ETHERNET like networks. This algorithm is being implemented as part of the INGRES data base system.
198 citations
TL;DR: Cost-effective control of distributed computing systems requires practical models for reassigning program modules among the processors in real time.
Abstract: Cost-effective control of distributed computing systems requires practical models for reassigning program modules among the processors in real time.
159 citations
TL;DR: Though attention is focused mainly on the design and operation of the network itself, the formation of a unified distributed processing system from a network of separate computer systems always remains the ultimate objective.
Abstract: Publisher Summary The field of distributed processing and computer networking is growing at a very rapid rate within the industry, government, and university communities While distributed processing systems may differ from computer networks both in perspective and in environment, they do have some common characteristics This chapter analyzes both an expository survey and research results on local computer networks in general and loop computer networks in particular Different types of local loop computer networks are surveyed along with typical design problem areas, such as the loop interface design, message transmission protocols, the network operating system, the network command language, the distributed programming system, the distributed data base system, and performance studies Though attention is focused mainly on the design and operation of the network itself, the formation of a unified distributed processing system from a network of separate computer systems always remains the ultimate objective Since the loop topology is adopted for the communication subnetwork and control is fully distributed among the nodes of the network, the resulting system is called a distributed loop computer network
152 citations
Proceedings Article•
01 Jan 1978
71 citations
TL;DR: A fundamental thesis of the HXDP project is that the benefits and cost-effectiveness of distributed computer systems depend on the judicious use of hardware to control software costs.
Abstract: The Honeywell Experimental Distributed Processor (HXDP) is a vehicle for research in the science and engineering of processor interconnection, executive control, and user software for a certain class of multiple-processor computers which we call "distributed computer" systems. Such systems are very unconventional in that they accomplish total system-wide executive control in the absence of any centralized procedure, data, or hardware. The primary benefits sought by this research are improvements over more conventional architectures (such as multiprocessors and computer networks) in extensibility, integrity, and performance. A fundamental thesis of the HXDP project is that the benefits and cost-effectiveness of distributed computer systems depend on the judicious use of hardware to control software costs.
57 citations
TL;DR: This paper focuses on solutions to the problems of data mangement in distributed systems, some of which arise directly from the nature of distributed architecture, while others carry over from centralized systems, acquiring new importance in their broadened environment.
Abstract: Successful implementation of most distributed processing systems hinges on solutions to the problems of data mangement, some of which arise directly from the nature of distributed architecture, while others carry over from centralized systems, acquiring new importance in their broadened environment. Numerous solutions have been proposed for the most important of these problems.
33 citations
Proceedings Article•
01 Jan 1978
28 citations
TL;DR: Several major problem areas that are critical for its success in non-trivial applications are discussed and several rules-of-thumb that a manager considering a distributed approach will want to consider are discussed.
TL;DR: This paper outlines the requirements of distributed database systems and attempts to survey the present level of support, as well as outlining the needs of implementors and vendors.
Proceedings Article•
01 Jan 1978
01 Jan 1978
TL;DR: Algo~ithms that embody simple distribution tactics are shown to be optimal in the sense of minimizing response time and total time for a special class of queries and a method is proposed to extend the optimal algorithms to derive efficient distribution strategies for queries.
Abstract: The application of computer network technology ~o-database systems has produced much interest in distributed database systems. Query processing on a distributed system is seen to be quite a different problem from query processing on a centralized system. A query requiring data from two or more distinct nodes in the network must be solved by a distribution strategy that consists of a schedule of local data pro-ceasing and data transmissions. Two cost measures, total time and response time, are used to judge the quality of a given distribution st~ategy. Methods that find efficient distribution strategies for queries are proposed and snalyzed. Algo~ithms that embody simple distribution tactics are shown to be optimal in the sense of minimizing response time and total time for a special class of queries. A method fs proposed to extend the optimal algorithms to derive efficient distribution strategies fo~ general query processing in distributed database systems.
01 Jan 1978
TL;DR: By modeling the broadcast portion of a mixed media network as a fully connected point-to-point network with link capacities varying as functions of the traffic rate, it is possible to extend an adaptive distributed routing algorithm that was originally developed for point to point ground networks as mentioned in this paper.
Abstract: By modeling the broadcast portion of a mixed media network as a fully connected point-to-point network with link capacities varying as functions of the traffic rate it is possible to extend an adaptive distributed routing algorithm that was originally developed for point-to-point ground networks. Additional modifications for improved dynamic performance at the satellite interface message processors are also considered.
Proceedings Article•
13 Sep 1978
TL;DR: This paper presents an architecture based upon logical relational machines which are the components of the distributed data base which is based on the model proposed in this paper.
Abstract: Distributed data bases are currently an important field of research and development. Our paper concerns the design of general purpose distributed data base management systems. We present an architecture based upon logical relational machines which are the components of the distributed data base.
01 Dec 1978
TL;DR: Current generic research studies in the areas of distributed search and distributed planning as outlined in the paper will aid in the development of a general theory for functionally accurate cooperative distributed systems.
Abstract: : The authors feel methodologies can be developed for functionally accurate cooperative distributed systems in which the distributed algorithms and control structures function with both inconsistent and incomplete data. These methodologies are necessary in order to extend the range of applications that can be effectively implemented in distributed environments. Their current generic research studies in the areas of distributed search and distributed planning as outlined in the paper will aid in the development of a general theory for functionally accurate cooperative distributed systems.
13 Nov 1978
TL;DR: This paper describes an existing, real-time, signal-processing system redesigned as a distributed signal processing system that is intellectually manageable, has no synchronization or mutual exclusion problems, is deadlock free, and implements transparent application interfaces.
Abstract: This paper describes an existing, real-time, signal-processing system redesigned as a distributed signal processing system. The design approach results in a system that is intellectually manageable, has no synchronization or mutual exclusion problems, is deadlock free, and implements transparent application interfaces. Techniques for achieving these capabilities are described, and include the use of Concurrent Pascal language, the design of specialized interprocess communication schemes, the use of preventive deadlock strategies, and a systematic testing approach.
11 Apr 1978
TL;DR: Several organizational possibilities for managing distributed databases are outlined and evaluated and the advantages and the problems introduced by distributing databases are described.
Abstract: Several organizational possibilities for managing distributed databases are outlined and evaluated. The advantages and the problems introduced by distributed databases are described.
13 Nov 1978
TL;DR: This paper will take a systems-oriented view of computer communication systems and examine the concept and relationships of distributed systems and networks.
Abstract: This paper will take a systems-oriented view of computer communication systems. It will begin with the examination of the concept and relationships of distributed systems and networks. There are many and varied definitions of distributed systems. For purposes of this discussion we will use a simple definition with which a large number of researchers(1) tend to agree. A distributed system is defined to be a system in which a processing load is distributed across a number of processing elements (PES) which are physically interconnected, but allow varying amounts of software intercommunication. Research in the area of distributed processors is proceeding mainly in three areas: network systems(2), special-purpose systerns(3), and high-performance, general-purpose systems(4).
01 Jan 1978
TL;DR: It is advantageous to identify weakly coupled systems by taking account of their space structure, and identify the inputs and outputs belonging to the same subsystems one by one usingernary matrices.
Abstract: It is advantageous to identify weakly coupled systems by taking account of their space structure. The solution may be obtained in three steps: (1) Considering the impulse response of the system, recognize the inputs and outputs belonging to the same subsystems. Ternary matrices are introduced. (2) Identify the subsystems one by one. If the coupling is weak, the interaction terms may be considered as perturbation terms. (3) Identify the interactions either systematically or in the order where they bring the most information on the complex system. Considerable savings in memory space can be achieved when using this method.
10 Oct 1978
TL;DR: A simulation model has been built of a loop network with characteristics of flexibility, modularity and programmability and has been used also as a tool for documenting the evolving project and as a design aid.
Abstract: Moving from a practical problem, the preliminary definition of a loop network, a simulation model has been built and will be described in this paper. For its characteristics of flexibility, modularity and programmability, this model has been used also as a tool for documenting the evolving project and as a design aid.