scispace - formally typeset
Search or ask a question

Showing papers on "Fast packet switching published in 1976"


Journal ArticleDOI
Robert M. Metcalfe1, David R. Boggs1
TL;DR: The design principles and implementation are described, based on experience with an operating Ethernet of 100 nodes along a kilometer of coaxial cable, of a model for estimating performance under heavy loads and a packet protocol for error controlled communication.
Abstract: Ethernet is a branching broadcast communication system for carrying digital data packets among locally distributed computing stations. The packet transport mechanism provided by Ethernet has been used to build systems which can be viewed as either local computer networks or loosely coupled multiprocessors. An Ethernet's shared communication facility, its Ether, is a passive broadcast medium with no central control. Coordination of access to the Ether for packet broadcasts is distributed among the contending transmitting stations using controlled statistical arbitration. Switching of packets to their destinations on the Ether is distributed among the receiving stations using packet address recognition. Design principles and implementation are described, based on experience with an operating Ethernet of 100 nodes along a kilometer of coaxial cable. A model for estimating performance under heavy loads and a packet protocol for error controlled communication are included for completeness.

1,701 citations


Proceedings Article
01 Jan 1976

525 citations



Journal ArticleDOI
TL;DR: This paper derives exact analytical expressions for the key system perfomance measures, the probability of loss for voice calls, and the expected waiting time for data packets from a multiplexing structure for mixing voice and data traffic in an integrated telecommunications system.
Abstract: Recent papers have introduced a multiplexing structure for mixing voice and data traffic in an integrated telecommunications system. This structure utilizes a master frame format of a time division statistical multiplex facility. A certain portion of the frame is allocated to voice calls, and data traffic is assigned to the remaining frame capacity. To achieve a high transmission utilization, data are allowed to use any residual voice capacity momentarily available due to statistical variations in the voice traffic. The voice traffic is treated as a loss system and data packets are buffered. In this paper we derive exact analytical expressions for the key system perfomance measures, the probability of loss for voice calls, and the expected waiting time for data packets. Actually, two cases are considered, the one discussed above, called the movable boundary case, and one where the boundary is fixed; i.e., data are not allowed to utilize the residual voice capacity. The computational aspects of calculating actual numbers are discussed in some detail, and results are presented for typical cases.

134 citations


Journal ArticleDOI
Simon S. Lam1
TL;DR: A model for a packet switching network in which each node has a finite pool of S/F buffers is presented and a heuristic algorithm for determining a balanced assignment of nodal S/f buffer capacities is proposed.
Abstract: Previous analytic models for packet switching networks have always assumed infinite storage capacity in store-store-and-forward (S/F) nodes. In this paper, we relax this assumption and present a model for a packet switching network in which each node has a finite pool of S/F buffers. A packet arriving at a node in which all S/F buffers are temporarily filled is discarded. The channel transmission control mechanisms of positive acknowledgment and time-out of packets are included in this model. Individual S/F nodes are analyzed separately as queueing networks with different classes of packets. The single node results are interfaced by imposing a continuity of flow constraint. A heuristic algorithm for determining a balanced assignment of nodal S/F buffer capacities is proposed. Numerical results for the performance of a 19 node network are illustrated.

70 citations




Patent
02 Mar 1976
TL;DR: In this article, a character oriented transmission control procedure for communicating with its associated terminals is used in the packet switching network, where data terminals each having a single communication link connected with the packet switch network, whereby the data sending and receiving system is effectively effected and a flow control for the switching network can be achieved without using any special procedure in the control.
Abstract: A packet switching network in which the data transmitted from a data sending terminal is temporarily stored in a packet switching exchange connected to a data sender where the data is divided into a plurality of units each of which is called a packet and the data in turn is transferred with each packet unit to a packet switching exchange connected to a data receiver, through a number of transmit switching exchanges, and then those packets are reassembled into the original format of the data which finally is transmitted to a data receiving terminal, the above-mentioned packet switching network employing a data sending and receiving system in which a character oriented transmission control procedure for communicating with its associated terminals is used in the packet switching network, and the packet switching network involves data terminals each having a single communication link connected with the packet switching network, whereby the data sending and receiving system is effectively effected and a flow control for the switching network can be achieved without using any special procedure in the control.

43 citations


Journal ArticleDOI
TL;DR: This paper study and compare the reliability of broadcast radio networks for different routing algorithms and suggests that a routing algorithm becomes more efficient when the number of alternate paths are restricted; on the other hand, reliability considerations suggest increasing the numberof alternate paths.
Abstract: Packet switching broadcast radio networks are receiving considerable attention as a feasible solution for applications involving fast network deployment requirements, inaccessible physical environments, and mobile communication devices. Such networks also offer economic alternatives to traditional multiplexing schemes for local distribution. Routing algorithms which enable point-to-point packet transportation have been developed for broadcast radio networks. It appears that a routing algorithm becomes more efficient when the number of alternate paths are restricted; on the other hand, reliability considerations suggest increasing the number of alternate paths. In this paper, we study and compare the reliability of broadcast radio networks for different routing algorithms.

39 citations


Journal ArticleDOI
C. Jenny1, K. Kummerle
TL;DR: A novel node concept is presented, imposing PS on a CS structure, and systems throughput for fixed processor module utilization is nearly a linear function of the number of modules.
Abstract: Based on the premise that flexibility is a primary objective for future data networks, it is proposed to integrate circuit switching (CS) and packet switching (PS) in a single network. This paper addresses the switching node in such a network. A novel node concept is presented, imposing PS on a CS structure. Connection management and node control functions on the one hand and packet processing/storing on the other are performed by two groups of modules. The modularity requirement leads to very small, simple processors built into storage modules for packet handling. Considering the high throughput requirement of about 1200 packets/s/node, configurations may exceed 30 packet-handling modules. The wellknown saturation effect of multiprocessor systems, i.e., decreasing incremental throughput with increasing number of processor modules, is overcome by a new concept of intermodule communications, taking advantage of the particular data-communications environment. Intermodule communications-and therefore overhead-have been significantly reduced. As a result, systems throughput for fixed processor module utilization is nearly a linear function of the number of modules, the deviation from linearity being less than 10 percent over the range considered.

14 citations


Journal ArticleDOI
TL;DR: Under the assumptions of this study, packet switched implementations showed decided advantages over circuit switched implementation, as measured by transmission utilization efficiencies and overall network costs.




Proceedings ArticleDOI
07 Jun 1976
TL;DR: A number of tradeoffs which offer guidelines for the design and optimum utilization of mixed media networks are examined, which are called mixed media packet switching networks.
Abstract: In this paper we present some of the important design issues for packet switching networks with both satellite and terrestrial components---which we call mixed media packet switching networks. Satellite packet switching has considerable promise for low cost, high bandwidth data communications. However there is inherent high delay in satellite links which does not appear in ground links. Therefore a mix of the two communications media offers the advantages of low-cost/high bandwidth together with low-delay communications where required. In this paper we examine a number of tradeoffs which offer guidelines for the design and optimum utilization of mixed media networks.

Patent
16 Nov 1976
TL;DR: In this article, a multi-stage switching network in a telecommunication switching system includes a virtual link frame having fixedly closed virtual switches which is inserted according to the switching system size required by the application range of telephone traffic into a specified intermediate stage of the multistage switching network to permit the switching network of upper size to be degenerated to a switching network with smaller numbers of stages by degeneratively inducing a partial path structure from the path structure of the upper sized switching network having a great traffic capacity.
Abstract: of the Disclosure A multi-stage switching network in a telecommunication switching system includes a virtual link frame having fixedly closed virtual switches which is inserted according to the switching system size required by the application range of telephone traffic into a specified intermediate stage of the multi-stage switching network to permit the switching network to permit the switching network of upper size to be degenerated to a switching network with smaller numbers of stages by degeneratively inducing a partial path structure from the path structure of the upper sized switching network having a great traffic capacity. The switching network with smaller numbers of stages can be incremented to a switching network of upper size according to the switching system size. In either case, the multi-stage switching network can be controlled by the same control circuit and the same control program.


Journal ArticleDOI
TL;DR: It is concluded that, although packet switching has several significant merits, these are achieved at the expense of a substantial increase in overheads.
Abstract: Overhead factors for data networks using message-switching or packet-switching techniques are defined and analysed. The transmission lines are assumed to have only random errors and the acknowledgments are handled on a ‘stop-and-wait’ strategy. For packet switching, the packet size used is such that, in each case, the overhead factor is minimised. Results show that the overheads for packet switching tend to be much higher than for message switching when the line error rates are low, i.e. better than 10−5. It is concluded that, although packet switching has several significant merits, these are achieved at the expense of a substantial increase in overheads.