Showing papers on "Node (networking) published in 1980"

The New Routing Algorithm for the ARPANET

Abstract: The new ARPANET routing algorithm is an improvement over the old procedure in that it uses fewer network resources, operates on more realistic estimates of network conditions, reacts faster to important network changes, and does not suffer from long-term loops or oscillations. In the new procedure, each node in the network maintains a database describing the complete network topology and the delays on all lines, and uses the database describing the network to generate a tree representing the minimum delay paths from a given root node to every other network node. Because the traffic in the network can be quite variable, each node periodically measures the delays along its outgoing lines and forwards this information to all other nodes. The delay information propagates quickly through the network so that all nodes can update their databases and continue to route traffic in a consistent and efficient manner. An extensive series of tests were conducted on the ARPANET, showing that line overhead and CPU overhead are both less than two percent, most nodes learn of an update within 100 ms, and the algorithm detects congestion and routes packets around congested areas.

Cash accounting and surveillance system for games

Abstract: A cash accounting and surveillance system for games, whereby operation of a number of player operated gaming devices may be monitored for purposes of detecting abnormal operation and/or cheating and for providing automatic accounting information for record keeping and pilferage detection purposes. The system utilizes a node concept with each node having a non-volatile data storage capability and a communications capability for communicating with each of a plurality of gaming devices coupled to the node. The exemplary embodiment disclosed operates in conjunction with slot machines having mechanically rotatable reels and a microprocessor control system for randomizing the reel stopping payouts and other machine functions. Alternate embodiments are disclosed.

Analysis of Shared Finite Storage in a Computer Network Node Environment Under General Traffic Conditions

Abstract: Nodal storage limitations in a store and forward computer network lead to blocking; this results in degradation of network performance due to the loss or retransmission of blocked messages. In this paper, we consider several schemes for sharing a pool of buffers among a set of communication channels emanating from a given node in a network environment so as to make effective use of storage in a variety of applications. Five sharing schemes are examined, analyzed, and displayed in a fashion which permits one to establish the tradeoffs among blocking probability, utilization, throughput, and delay. The key to the analysis lies in the observation that the equilibrium joint probability distribution of the buffer occupancy obeys the well-known product form solution for networks of queues. The study indicates advantages and pitfalls of each of the sharing schemes. We observe, in general, that sharing with appropriate restrictions on the contention for space is very much desirable.

Complexity of network reliability computations

Abstract: This paper considers the difficulty of computing several measures of network reliability on directed and undirected networks. Results concerning the NP-difficulty of several network reliability analysis problems are unified and in several cases generalized to wider classes of measures. Reductions are also given that relate network reliability problems on directed and undirected networks and problems with and without node failures.

Parallel pattern verifier with dynamic time warping

Abstract: A speech recognition system is disclosed which employs a network of elementary local decision modules for matching an observed time-varying speech pattern against all possible time warpings of the stored prototype patterns. For each elementary speech segment, an elementary recognizer provides a score indicating the degree of correlation of the input speech segment with stored spectral patterns. Each local decision module receives the results of the elementary recognizer and, at the same time, receives an input from selected ones of the other local decision modules. Each local decision module specializes in a particular node in the network wherein each node matches the probability of how well the input segment of speech matches the particular sound segments in the sounds of the words spoken. Each local decision module takes the prior decisions of all preceding sound segments which are input from the other local decision modules and makes a selection of the locally optimum time warping to be permitted. By this selection technique, each speech segment is stretched or compressed by an arbitrary, nonlinear function based on the control of the interconnections of the other local decision modules to a particular local decision module. Each local decision module includes an accumulator memory which stores the logarithmic probabilities of the current observation which is conditional upon the internal event specified by a word to be matched or identifier of the particular pattern that corresponds to the subject node for that particular pattern. For each observation, these probabilities are computed and loaded into the accumulator memory of all the modules and, the result of the locally optimum time warping representing the accumulated score or network path to a node for the word with the highest probability is chosen.

Mechanisms for broadcast and selective broadcast

Abstract: This thesis deals with a problem in the effective use of a loosely-coupled store-and-forward network like the ARPANET. Many applications assume the existence of a mechanism to send a broadcast message to every node in the network, or a selective broadcast message to a specified group of nodes. Previous work in this field has resulted in several techniques for broadcasting to the entire network. These include an algorithm for imposing a minimum spanning tree on the network and maintaining it in the face of failures and changing costs, so that a broadcast can be routed along the branches of the MST at a minimum cost to the network as a whole. I extend this work in two directions. First of all, I discuss ways of adapting existing broadcast mechanisms to the more general problem of selective broadcast. In one case this involves generalizing the MST problem to the problem of finding a minimum Steiner tree, a problem which is NP-complete. I examine an existing approximation algorithm and show how to modify it for use in a distributed environment. In the process I show that if we make a reasonable assumption about the way ties between edges are broken then we can considerably simplify this algorithm. Second, I propose an original broadcasting technique that forwards along the branches of a single tree in the same manner as the MST-based algorithm. The tree used is a shortest-path tree for a particular node that is, in some sense, "in the center" of the network; hence this method is called center-based forwarding. Using such a tree allows us to provide a broadcast facility with a small delay rather than a small cost; unfortunately it is impossible in general to minimize the delay from every source if we use only a single tree. To evaluate center-based forwarding, I define four measures of the delay associated with a given broadcast mechanism, and then propose three ways of selecting the center node. For each of the three forms of center-based forwarding, I compare the delay to the minimum delay for any broadcasting mechanism and also to the minimum delay for any single tree. In most cases, the delay on the centered tree is bounded by a small constant factor multiplied by either of these two minimum delays. When I can, I give a strict bound on the ratio between the center-based delay and the minimum delay; otherwise I demonstrate that no bound is possible. These results immediately imply bounds on how bad the three centered trees can be with respect to each other; most of these bounds are strict, and I improve the rest to bounds that are also strict. Finally I present an algorithm for maintaining the centered tree in the face of changes in traffic conditions and failures in nodes and links. I also discuss the extension of center-based forwarding to the problem of selective broadcast.

Transparent intelligent network for data and voice

Abstract: A transparent intelligent communication network having both terrestrial and satellite links between nodes and providing improved accuracy and speed in transmission by distinguishing between data requiring rapid receipt at the receiving node and other data, and dispatching data accordingly, while transmitting system protocol and error-correcting information (e.g., retransmissions) over terrestrial links.

Modelling of rail networks: Toward a routing/makeup model

Abstract: Freight flow management in rail systems involves multicommodity flows on a network complicated by node activities (queueing and classification of cars at marshalling yards). Routing in these systems should account for technology requirements of motive power and traction as well as resource allocation (cars to blocks, blocks to trains). In this paper, we propose a hierarchial taxonomy of modelling issues and describe a class of models dealing with car routing and train makeup from the viewpoint of network flows and combinatorial optimization. We compare our model with two previous rail network models and discuss possibilities for algorithmic development.

Distributed digital data communications network

Abstract: A distributed digital data communication network having widely dispursed nodes provides for effective transmission of messages between each node through the use of flood routing. A response message based upon stored data link status obtained when a message is received through the flood routing protocol selects the shortest time delay path for responding to that message. This process eliminates network instability.

Isolation Method in a Network of Queues

Abstract: In this paper a new method for analyzing complex queueing networks is proposed: the isolation method. As an example, we study packet switching networks with finite buffer size at each node.

Approximate analysis of exponential queueing systems with blocking

Abstract: A network of service stations Q 0 Q 1,...,QM is studied. Requests arrive at the centers according to independent Poisson processes; they travel through (part of) the network demanding amounts of service, with independent and negative exponentially distributed lengths, from those centers which they enter, and finally depart from the network. The waiting rooms or buffers at each service station in this exponential service system are finite. When the capacity at Q i is reached, service at all nodes which are currently processing a request destined next for Q i is instantaneously interrupted. The interruption lasts until the service of the request in the saturated node Q i is. completed. This blocking phenomenon makes an exact analysis intractable and a numerical solution computationally infeasible for most exponential systems. We introduce an approximation procedure for a class of exponential systems with blocking and show that it leads to accurate approximations for the marginal equilibrium queue length distributions. The applicability of the approximation method may not be limited to blocking systems.

A Queueing Analysis of Two ARQ Protocols

Abstract: In every data communication system, a procedure must be provided to allow for the retransmission of data when errors are detected. The receiving node is required to make a (positive/negative) acknowledgment (ACK/NACK) to the sending node. Until an acknowledgment is received, a "copy" of the message must be retained at the sending node. If an ACK is received, the space assigned to the "copy" is released. If either a NACK or no acknowledgment is received in a suitable interval of time, retransmission is required. Different protocols specifying how the nodes recover from a transmission error can be defined. In this paper, we present a queueing analysis of the two ARQ (automatic repeat request) protocols-block and select ARQ-for a (slotted) concentrator network node.

Partitionable parallel processor

Abstract: In a parallel processing array wherein each processor therein issues a ready signal to signify that it is ready to begin a parallel processing task and initiates the task upon receipt of an initiate signal the parallel processing array is rendered partitionable into parallel processing subarrays by a control node tree having a plurality of control nodes connected to the plurality of processors and in decreasing levels to each other in a tree-like fashion down to a single root node. Each node is controlled to function as a non-root wherein it receives a ready signal from its processor side and passes it along toward the single root node or as a root node whereupon receiving a ready signal it issues back an initiate signal toward the plurality of processors.

An abstract mode–abstract (destination) node model of foreign holiday demand

Abstract: (1980). An abstract mode–abstract (destination) node model of foreign holiday demand. Applied Economics: Vol. 12, No. 2, pp. 163-180.

Method of and system for routing in a packet-switched communication network

Abstract: A node in a packet-switching data-transmission network has a processor distributing incoming messages or packets from a receiving buffer to transmitting buffers selected according to routing data established by an updating circuit which algebraically combines incremental delays with respective path delays to obtain total delays assigned to message transmission from the processor over respective transmitting buffers and associated outgoing transmission paths. The increment delays assigned to the respective buffers are calculated by a delay estimator utilizing data from the processor including packet-service times and arrival and departure times of the packets in the various buffers, the path delays assigned to routes extending from the respective transmitting buffers to a terminal node being communicated to the updating circuit via the processor from nodes connected downstream of the transmitting buffers.

Fast Detection and Identification of Islands in Power Networks

Abstract: Two algorithms fur the determination of network connectivity and detection of islands in a power system are presented. Both algorithms are tested on a number of typical power systems. Effects of node ordering and matrix storage methods on the computation time are studied for both algorithms and the results are compared. Each algorithm proved to be more suitable under certain conditions. The results show both algorithms to be very fast. The connectivity of a 115 substation power system is determined in less than 25 milliseconds on a small industrial computer using the new algorithms.

Synchronization of distributed simulation using broadcast algorithms

Abstract: Simulation, particularly of networks of queues, is an application with a high degree of inherent parallelism, and is of considerable practical interest. We continue the analysis of synchronization methods for distributed simulation, defined by the taxonomy in our previous paper. Specifically, we develop algorithms for time-driven simulation using a network of processors. For most of the synchronization methods considered, each node k of an n-node network simulation cannot proceed directly with its part of a simulation. Rather, it must compute some function B k ( ν 1 , ν 2 , …, ν n ), where ν i is some value which must be obtained from node i . The value of ν i at each node changes as the simulation progresses, and must be broadcast to every other node for the recomputation of the B -functions. In some cases, it is advantageous to compute the B -function in a distributed manner. Broadcast algorithms for such distributed computation are presented. Since the performance of a broadcast algorithm depends on the properties of the inter-process communication facility, we characterize some particular cases and give algorithms for each of them.

A class of optimal routing algorithms for communication networks

Abstract: : This report describes an algorithm for minimum delay routing in a communication network. During the algorithm each node maintains a list of paths along which it sends traffic to each destination together with a list of the fractions of total traffic that are sent along these paths. At each iteration a minimum marginal delay path to each destination is computed and added to the current list if not already there. Simultaneously the corresponding fractions are updated in a way that reduces average delay per message. The algorithm is capable of employing second derivatives of link delay functions thereby providing automatic scaling with respect to traffic input level. It can be implemented in both a distributed and a centralized manner, and it can be shown to converge to an optimal routing at a linear rate.

Use of nodes to uniquely identify processes

Abstract: A method for providing a node address uniquely identifies a process to be separately addressed and dispatched in a uniprocessor or multiprocessor environment. A single processor (14) identified by a unique primary node may include a single node of multiple processes (50), multiple subnodes of processes which do not replicate all of the processes in the subnode, or multiple subnodes with multiple processes replicated in each of said subnodes. Multiple processors (14), (64) and (66) may be joined by a communication data buss (60) to form a network of processors in which any processor may assign a work request to a process in its own or any other processor of the network to efficiently schedule work requests queued to any process in the network of processors.

Method of and device for reducing speech-reproduction irregularities in packet-switching telecommunication systems

Abstract: Two mutually independent clocks at a transmitting node and a receiving node of a routing network in a packet-switching telephone system are correlated by a synchronizing device associated with the receiving node whose processor, at the beginning of voice transmission between the two nodes, decodes a label of an incoming packet to determine its time of departure ts from the transmitting node whereupon a control unit at the receiving node adjusts the local clock in conformity with an estimated reading tx of the remote clock. This adjustment takes into account only fixed network parameters determining a minimum transit time for the packets by introducing a basic delay Td. After the correlation, an initial packet of a sequence representing a period of coherent speech --termed talkspurt--is stored for a time sufficient to augment the basic delay Td by a supplemental time lag T80 between its emission at the transmitting node and its insertion into an outgoing queue of the receiving node, time lag T80 being at least equal to an average random delay dt actually superimposed upon the basic delay by variable network parameters such as traffic density; this minimizes the incidence of gaps in a talkspurt and major variations of pauses between talkspurts as perceived by a receiving subscriber.

Deadlock Avoidance in Store-and-Forward Networks--II: Other Deadlock Types

Abstract: This paper describes the construction of loop-free buffer graphs which avoid four types of buffer deadlocks in store-and-forward networks. 1) Progeny deadlock, where original messages spawnother ones, and buffer contention occurs between the original and progeny messages. This occurs when positive or negative acknowledgments are created, e.g., if messages reverse direction after encountering a path failure. 2) Copy-release deadlock, where a message copy is stored at the source node and the buffer is not released until an acknowledgment is received from the destination node. Buffer contention may arise among the original messages, stored copies, and acknowledgments. 3) Pacing deadlock, where a local flow control protocol is used between a network node and attached terminals. Buffer contention may arise between the message flows into and out of the terminal, preventing the transmission of go-ahead commands. 4) Reassembly deadlock, whereby reassembly of packetized messages at the destination node cannot be completed. The solution presented here has the novel features of not requiring preallocation of reassembly buffers before transmission of multiple packets of a multipacket message, and not requiring dedication of buffer space at intermediate nodes for individual messages. These schemes are believed to have modest buffer requirements at each node, and if adequate buffer pools are provided, will incur negligible performance degradations under normal conditions, with overhead increasing under heavy buffer usage when deadlock is near.

Communication system for distributed control arrangement

Abstract: A communication system is disclosed for a distributed control arrangement comprising a plurality of communication nodes, each corresponding to a separate physically remote operating unit of a common installation. Each node includes a set of port buffers respectively associated with separate subsystems of the corresponding operating unit, a common node bus and a controller. Each port buffer is adapted to establish a 2-stage transfer link between its corresponding subsystem and the controller, in which the port buffer provides intermediate data storage. One port buffer of each node is coupled to a secondary station associated with that node which is slaved to a primary station. The primary station issues commands through a common data link to the secondary stations and communicates with selected subsystems associated with each node through the secondary station corresponding to that node.

Dynamic evolution in societal networks

Abstract: A societal network is defined to be a finite directed graph in which individuals are represented by nodes, and relations between individuals by labelled arcs. Each individual undergoes state transitions at discrete instants of time, so that the societal network may be thought of as a deterministic dynamic process. It is shown that the individuals of such a network may be divided into equivalance classes, so that the original network may be represented by a reduced network containing one node for each equivalence class. Similar results are obtained for a more general type of partition called a “class structure.” The application of these concepts is illustrated in the context of “balanced” networks, which have either positive or negative relations between nodes. The long‐run behavior of societal networks is then examined, and it is shown that any network will eventually reach either a stable state or a periodic pattern of state transitions. In the case of a probabilistic transition rule, it is proved that t...

Fault diagnosis in computing networks

Abstract: The problem of diagnosis of failures in computing systems constructed as interconnections of discrete modules or units is considered. It is assumed that the modules in the network are capable of performing tests upon other modules, and also of being tested by other modules. Two classes of systems are considered, those which incorporate some central, or host facility, and those which are fully distributed. For the former class, extensions to existing system-level fault diagnosis concepts are given, to allow the central facility to diagnose failures based on results collected from tests performed by the nodes of the system. Specific problems considered include the problem of producing a correct diagnosis from test results which do not all reflect the condition of the system at a single point in time, and the problem of diagnosis of failures affecting the communication facilities over which tests are performed. New diagnostic models and measures are introduced to deal with these problems and a number of necessary and sufficient conditions for a system to achieve a given level of diagnosability under these models, are given. For the class of fully distributed systems, a new diagnostic process, known as self-diagnosability is proposed via which each node in a network can independently diagnose the condition of all other nodes in the system. The need for this type of diagnostic ability is argued via introduction of a notion called distributed fault-tolerance. A number of factors affecting the extent and efficiency of the self-diagnostic process are discussed.

A Survivable Network of Ground Relays for Tactical Data Communications

Abstract: The rationale and a preliminary design concept are presented for a new mobile ground-to-ground communications system to support tactical command and control. Survivability is attained through high connectivity, antenna directivity, and spread-spectrum signals. Each node of the proposed network is equipped with a circular cylindrical array antenna capable of steering a narrow beam in the direction of one of its several neighbors. A single transceiver operates half-duplex m a flexible time-division format among the multiple links at a node. Each link uses a unique pseudonoise code, and the network as a whole employs code-division multiple access. Mobility and survivability are enhanced because a single equipment supports multiple links while providing the benefits of high antenna gain and directivity. Network entry and synchronization schemes are considered. System parameters are defined and link power budgets for signal-to-jammer and signal-to-noise ratios are calculated. Implementation prospects are addressed drawing on technology developments in antenna arrays, surface-acoustic wave signal processing, and adaptive routing in packet networks.

Large Change Response Sensitivity of Linear Networks

Abstract: Absrmzcr-A general systematic method for determining lqe-hmge, muMiparameter sensitivity of linear networks is developed from the formal thmy of scattering in amdensed matter The method, based on matrix mdpulation, co- and unifies previous works on analog and digital networks Standard Basis Operators are intmduced as an expedient too4 wbicb aide3 in the solution of scattering matrix equations Explicit exact sensitivity formulas are developed for several types of multipammeter changes in both analog and digital networks I INTR~LXJCT~~N I N THE last decade many papers have dealt with the problem of linear network response function sensitivity to changes in network parameters Several recent works [l]-[7] have focused attention on response function change due to large-scale multiparameter changes In the area of digital networks, Temes and Cho [l] have developed a useful formula for calculating large-scale sensitivities Their method, utilizing Tellegen’s theorem for digital networks, extends the Taylor series single-parameter formula of Crochiere [5] to a multivariable sensitivity formula The motivation for the work presented here is severalfold It is shown that general linear network sensitivity functions are generated by matrix manipulation using techniques from scattering theory commonly applied in the quantum theory of condensed matter [8] The results are independent of any physical arguments, other than the assumption that the network topology is a constant No assumptions are made concerning the structure of the “shift” matrix characterizing the changed parameters of the network 4 systematic procedure, utilizing Standard Basis Operators (SBO) [9] is introduced to solve scattering matrix equations, which yields network response functions of the changed network in terms of response functions of the original network In calculating the new response functions, it is shown to be advantageous to distinguish three classes of network parameters: those associated with a single node (node parameters), those linking two nodes with one way information, or signal, transfer (unilateral bond parameters), and those linking two nodes with twoway signal transfer (bilateral bond parameters) Explicit exact sensitivity formulas are developed for parameter changes in each class, and a linear active two-port network is used as an example to illustrate the application of one of the multiparameter formulas