Showing papers on "Shared resource published in 1972"

A system for interprocess communication in a resource sharing computer network

Abstract: A system of communication between processes in a time-sharing system is described and the communication system is extended so that it may be used between processes distributed throughout a computer network. The hypothetical application of the system to an existing network is discussed.

Flow Control in a Resource-Sharing Computer Network

Abstract: In this paper, we discuss flow control in a resourcesharing computer network. The resources consist of a set of inhomogeneous computers called hosts that are geographically distributed and are interconnected by a store-and-forward communications subnet. In the communication process, messages pass between hosts via the subnet. A protocol is used to control the flow of messages in such a way as to efficiently utilize the subnet and the host resources. In this paper, we examine in some detail the nature of the flow control required in the subnet and its relation to the host flow control and subnet performance.

Resource-sharing computer communications networks

Abstract: The development of resource-sharing networks can facilitate the provision of a wide range of economic and reliable computer services. Computer-communication networks allow the sharing of specialized computer resources such as data bases, programs, and hardware. Such a network consists of both the computer resources and a communications system interconnecting them and allowing their full utilization to be achieved. In addition, a resource-sharing network provides the means whereby increased cooperation and interaction can be achieved between individuals. An introduction to computer-to-computer networks and resource sharing is provided and some aspects of distributed computation are discussed.

Improvements in the design and performance of the ARPA network

Abstract: In late 1968 the Advanced Research Projects Agency of the Department of Defense (ARPA) embarked on the implementation of a new type of computer network which would interconnect, via common-carrier circuits, a number of dissimilar computers at widely separated, ARPA-sponsored research centers. The primary purpose of this interconnection was resource sharing, whereby persons and programs at one research center might access data and interactively use programs that exist and run in other computers of the network. The interconnection was to be realized using wideband leased lines and the technique of message switching, wherein a dedicated path is not set up between computers desiring to communicate, but instead the communication takes place through a sequence of messages each of which carries an address. A message generally traverses several network nodes in going from source to destination, and at each node a copy of the message is stored until it is safely received at the following node.

Shared Resource Multiprocessing

Abstract: A good design is the Elderado in any systems area. The elements of good design include efficiency, simplicity, and flexibility for use. While we make no pretense that the ideas described in this paper are essential to good computer design, it seems to us that it does meet many of the criteria and avoids a number of the pitfalls associated with high performance computers.

Network computer modeling

Abstract: Network computers are becoming a reality in the seventies. While systems such as the ARPA network, Carnegie Mellon's PLN, the Collins C-System, CDC's Cybernet, and GE's Time Share Net are coming to fruition, even more grandiose systems are being discussed. These networks all offer the designer the potential of combining the advantages of data base sharing, resource sharing, and load leveling with those of message switching. From a postulation of the essential characteristics of a network computer currently under development, we formulate a queueing theory model for a multiserver system with a finite length priority queue. Then, under the assumptions of Poisson input and exponentially distributed processing times, we use this idealized mathematical model to investigate the steady state stochastic behavior of jobs in the network. We are particularly interested in the efficiency of computer utilization and the average waiting time for jobs of different priority classes.

Aspects of Large-Scale Resource Sharing through Networks of Computers

Abstract: : The paper discusses some advantages of resource sharing through a network of computers as compared to resource sharing through other forms of timesharing services. The ARPA network of computers, sponsored by the Defense Advanced Research Projects Agency of the Department of Defense, is used as an example of large-scale resource sharing in a computer network. The paper discusses the technical and economic aspects of computer networks, touching only briefly on legal and social implications. The paper describes some difficulties encountered in the use of computer networks and possible ways to address these difficulties. The problem areas discussed are those with application to a broad group of users, and with high potential for solution in the near future.

Background reading in microprogramming and emulation

Abstract: "Microprogrammed computers basically differ from conventional computers in that the equivalent of a single instruction in a conventional computer is often a subroutine (interpreted instruction) in a microprogrammed computer." In such computers a program can be considered a calling sequence. The calling sequences of the machines listed above are described to show how the control sequences are simplified and more flexibility is allowed through the use of microprogramming. Tlle cost of additional execution time is compared with the flexibility provided by microprogramming and the time saved in executing some complex instructions.

Design automation in network computers

Abstract: Network computers are becoming a reality in the seventies. While systems such as the ARPA Network, Carnegie Mellon's PLN, the Collins C-System, CDC's Cybernet, and GE's Time Share Net are coming to fruition, even more grandiose systems are being discussed. These networks all offer the designer the potential of combining the advantages of resource sharing, data base sharing, and message switching with those of design automation. In this paper we present a brief description of a network computer currently under development and then propose an interactive design automation system for this geographically distributed network.The control mechanism for our design automation system is provided by an operating system, resident in each computer, whose primary function is to segment, disseminate, and regulate the jobs generated by the design automation system. When a job enters the system it may be assigned to anyone of the computers in a center. The job then runs to completion under control of that computer, but the segments that make up the job may be executed in any available computer. Additionally, we may transmit jobs between centers to improve the network response time. Thus we see that by detecting when segments are ready to be executed and by queueing this available work uniformly among all computers in the network, our design automation system will give each designer access to the full parallel processing power of the network.