Showing papers on "Node (networking) published in 1982"

Elections in a Distributed Computing System

Abstract: After a failure occurs in a distributed computing system, it is often necessary to reorganize the active nodes so that they can continue to perform a useful task. The first step in such a reorganization or reconfiguration is to elect a coordinator node to manage the operation. This paper discusses such elections and reorganizations. Two types of reasonable failure environments are studied. For each environment assertions which define the meaning of an election are presented. An election algorithm which satisfies the assertions is presented for each environment.

Echo Algorithms: Depth Parallel Operations on General Graphs

Abstract: This paper describes a method for the detection of properties of general graphs in an environment in which each node can be considered an autonomous processor, interacting with its neighbors by passing messages.

A Fault-Tolerant Communication Architecture for Distributed Systems

Abstract: A communication architecture for distributed processors is presented here. This architecture is based on a new topolgy we have developed, one which interconnects n nodes by using rn links where the maximum internode distance is logrn, and where each node has, at most, 2r, I/O ports. It is also shown that this network is fault-tolerant, being able to tolerate up to (r − 1) node failures.

Multicast communication tree creation and control method and apparatus

Abstract: In a multicast network communication system, administration of the communication path making up the multicast tree itself has been separated from control and administration of the network. Creation of a multicast distribution tree and control over the membership thereof, is separately controlled independently from the creation and use of the tree transmission path used to communicate among the members of a multicast set. Transmission distribution trees are set up when a transmission request is received and the properties of the transmission path that is required are known. Transmission paths are created and controlled by all nodes in the communications system, each node having necessary control code and processors for responding to requests from set members to transmit a message to groups of users by creating and activating the necessary tree communication path distribution linkages. A distribution tree is created by the Tree Leader by generating a tree address using a random number generator. A tree address correlator is generated utilizing network and node identifiers unique for the network, and a list of subnodes or users connected for each member of the multicast tree set is generated. Using this information, a tree distribution path is computed to cover all of the subnodes required and a tree set up request message is sent by the Tree Leader along a computed path to each involved subnode. Each subnode returns a message indicating whether the tree address is already in use or is available for use. Successfully negotiated tree addresses are marked at the path link initiation and termination points at each node through the network.

Sections des fibrés vectoriels sur une courbe

Abstract: Let X be a smooth and proper algebraic curve of genus g ̂ 2. If E is a semistable vector bundie on X ofrank r and degrce r{g-1) (thus xW^O), does H°(X, £®L)=0 for L a generic line bundie of degree O? Thé answer is no in général, but is positive for smaiï values of r and (in characteristic p > 0) when £ is thé sheaf of locally exact differentials. Thé latter case implies that thé /?-Sylow's subgroups of îii (X} are rather big. 0. Soit X une courbe algébrique, propre, lisse, connexe, de genre g, définie sur un corps k algébriquement clos. Si E est un fibre vectoriel sur X, E a un rang r=r(£) et un degré rf=rf(£). Pour £^0, on définit la pente de E : X=Â,(£)=rf/r. La caractéristique d'Euler-Poincaré de £, X(£)=A°(£)-A (£), (où h(E)=dimkH(X, £)), est donnée par la formule de Riemann-Roch : X(£)=rf-r(^~l)=r(?L-(^-l)). Dans la suite, L désigne un faisceau inversible variable sur X, de degré 0. Les divers faisceaux L sont paramétrés par les points de la jacobienne Jde X. On a îc (£) = X (E®L) == h° (£®L) h (£®L). L'entier h° (£®L) peut varier (*) Texte reçu le 19 mars 1981. révisé le 15 juin 1981. Michel RAYNAUD, Université de Paris-Sud, Département de Mathématique, Bâtiment 425, 91405 Orsay Cedex, E.R.A. 653. BULLETIN DE LA SOCIÉTÉ MATHÉMATIQUE DE FRANCE 0037-9484/1982/103/S 5.00 © Gauthier-Villars

Apparatus for recovery from failures in a multiprocessing system

Abstract: A number of intelligent nodes (bus interface units-BIUs and memory control units-MCUs) are provided in a matrix composed of processor buses (105) with corresponding error-reporting and control lines (106); and memory buses (107) with corresponding error-reporting and control lines (108). Error-detection mechanisms deal with information flow occuring across area boundaries. Each node (100, 101, 102, 103) has means for logging errors and reporting errors on the error report lines (106, 108). If an error recurs the node at which the error exists initiates an error message which is received and repropagated on the error report lines by all nodes. The error message identifies the type of error and the node ID at which the error was detected. Confinement area isolation logic in a node isolates a faulty confinement area of which the node is a part, upon the condition that the node ID in an error report message identifies the node as a node which is a part of a faulty confinement area. Logic in the node reconfigures at least part of the system upon the condition that the node ID in the error report message identifies the node as a node which is part of a confinement area which should be recofigured to recover from the error reported in the error report message.

All digital IDMA dynamic channel allocated satellite communications system and method

Abstract: An all digital TDMA dynamic channel allocated satellite communications system and method provides a satellite communications network between geographically separated nodes. The topology of the network is controlled by an on-line master node in accordance with user demands and operator commands received from the nodes in the network. A frame communication format is employed where each node in the system is allocated a burst within each frame. The length of the bursts is controlled by the on-line master node to provide the communications channels in the network. Control information to and from the on-line master node is conveyed in the network via the satellite link in selected frame bursts. In this way, the on-line master node dynamically controls the topology of the satellite communications network.

Network control center call trace

Abstract: A telecommunication network has a plurality of tandem node switches interconnected by trunks. Each switch generates message detail records (MDR) which are transmitted by data links to a network control center. In response to complaints, an NCC operator can start a mechanized search of MDR to retrace the path a call had earlier taken. The retraced path may be graphically illustrated on a display network map. Data from each call trace may be stored in a call trace log for subsequent retrieval.

Digital data processing system

Abstract: A data processing system having a flexible internal structure, protected from and effecitvely invisible to users, with multilevel control and stack mechanisms and capability of performing multiple, concurrent operations, and providing a flexible, simplified interface to users. The system is internally comprised of a plurality of separate, independent processors, each having a separate microinstruction control and at least one separate, independent port to a central communications and memory node. The communications and memory node is an independent processor having separate, independent microinstruction control and comprised of a plurality of independently operating, microinstruction controlled processors capable of performing multiple, concurrent memory and communications operations. Addressing mechanisms allow permanent, unique identification of information and an extremely large address space accessible and common to all such systems. Addresses are independent of system physical configuration. Information is identified to bit granular level and to information type and format. Protection mechanisms provide variable access rights associated with individual bodies of information. User language instructions are transformed into dialect coded, uniform, intermediate level instructions to provide equal facility of execution for all user languages. Operands are referred to by uniform format names which are transformed, by internal mechanisms transparent to users, into addresses.

Capacity Consideration in Reliability Analysis of Communication Systems

Abstract: This paper presents a simple method for deriving a symbolic reliability expression of some practical systems such as a communication system having fixed channel capacities of its links, a power distribution system having limited power ratings of its power lines, a transport system which might not allow traffic more than a particular value, or a chemical system in which oil or gas flow through pipes is permissible only up to some safe limits. A system is good if and only if it is possible to transmit successfully the required capacity from source node to the sink node. This paper defines a group as a set of branches such that success of these branches ensures system success, as defined above. All such groups are obtained from a knowledge of the minimal paths of the system graph. The method is computerized and implemented on DEC-20 computer. Two examples are considered and their solutions presented to illustrate the technique.

Sacrificing serializability to attain high availability of data in an unreliable network

Abstract: We present a simple algorithm for maintaining a replicated distributed dictionary which achieves high availability of data, rapid processing of atomic actions, efficient utilization of storage, and tolerance to node or network failures including lost or duplicated messages. It does not require transaction logs, synchronized clocks, or other complicated mechanisms for its operation. It achieves consistency contraints which are considerably weaker than serial consistency but nonetheless are adequate for many dictionary applications such as electronic appointment calendars and mail systems. The degree of consistency achieved depends on the particular history of operation of the system in a way that is intuitive and easily understood. The algorithm implements a "best effort" approximation to full serial consistency, relative to whatever internode communication has successfully taken place, so the semantics are fully specified even under partial failure of the system. Both the correctness of the algorithm and the utility of such weak semantics depend heavily on special properties of the dictionary operations.

The Design of a Reliable Remote Procedure Call Mechanism

Abstract: In this correspondence we describe the design of a reliable Remote Procedure Call mechanism intended for use in local area networks. Starting from the hardware level that provides primitive facilities for data transmission, we describe how such a mechanism can be constructed. We discuss various design issues involved, including the choice of a message passing system over which the remote call mechanism is to be constructed and the treatment of various abnormal situations such as lost messages and node crashes. We also investigate what the reliability requirements of the Remote Procedure Call mechanism should be with respect to both the application programs using it and the message passing system on which it itself is based.

Routing Schemes for the Augmented Data Manipulator Network in an MIMD System

Abstract: There have been many multistage interconnection networks proposed in the literature for interconnecting the processors that comprise large parallel processing systems. In this paper, the use of the Augmented Data Manipulator and Inverse Augmented Data Manipulator multistage networks in the MIMD mode of operation is considered. A tag based routing scheme which allows distributed control of either network is proposed. Rerouting schemes that allow a message blocked by a busy or known faulty node in its present path to dynamically make use of a nonbusy node and continue, when possible, are described for both networks. Finally, a tag based broadcasting scheme for the networks is introduced that allows one processor to send messages to a subset of the other processors.

Computer system with tasking

Abstract: An integrated data processing and communication service is provided by a system which comprises a plurality of nodes interconnected via a packet-switched transport network. Each node includes a database processor and at least one node processor, with a number of terminals and hosts being connected to the latter via at least one front-end processor. Processes existing with each node processor are two types--application layer processes which perform "useful work" on behalf of the customers and the vendor of the service, and control layer processes, each of which manages, or provides some part of, the service itself. The processors are virtual memory processors and the program region of the virtual address space of each process includes both a process-specific image and shared image. The former is weakly linked to the latter and provides it, via a set of primitives, with a number of communications processing services including file/database management, application program control, interprocess communications, station access and message services. Such services are also available for use within the shared image itself. The shared image also provides a number of "supervisor" services including enforcement of a system of privileges for the primitives; memory allocation; condition handling; event services; and establishment and management of the tasking system.

Minimum cost flow with set‐constraints

Abstract: The minimum cost network flow problem with set-constraints is a generalization of the well-known minimum cost network flow problem, in which bounds on the sum of flows through sets of arcs exist. This paper investigates some variations of this problem, including the polymatroid intersection problem, where for each node two polymatroids are given; one polymatroid constrains flows entering the node, and the other constrains flows leaving it.

Internodal conference call administrator

Abstract: A telecommunication network has a plurality of tandem node switches interconnected by trunk circuits. A network control center controller mechanically administers internodal conference calls by storing data pertaining to a scheduled conference call. Prior to the scheduled time, the data is downloaded to the tandem node switches which set an internodal conference bridge. Participants in a conference call first dial a bridge access number and then a password which assures security.

Multiple Control Stations in Packet Radio Networks

Abstract: This paper presents a design recently implemented by the Packet Radio project for control of large networks. The network is built on a carrier sense, multiple access broadcast channel and is populated with mobile store and forward nodes known as packet radio units, or PRs. Until recently, the packet radio network operated with centralized routing controlled by one node, the station. To increase the size of the network and to provide redundant control a multiple station design, "multistation", was necessary. The PRs gather local connectivity information which is then forwarded to the controlling nodes (stations). Each station is responsible for up to 40 PRs located near it in connectivity and uses these PRs to communicate with other stations. Questions important to the development of multistation include: How are packet radios matched with controlling stations? How do stations find and communicate with each other? How are long routes crossing the borders of many stations determined, and how is their successful setup insured? How is control traffic minimized?

Passage times for overtake-free paths in gordon-newell networks

Abstract: Consider a path in a multiclass Gordon-Newell network such that a customer present in a node of this path cannot be overtaken by any other customer behind him in a node of this path or by probabilistic influences created by such customers. The passage time through such a path is a mixture of Erlangian distributions, where the mixing distribution is given by the steady

Synchronization in a communication network of interconnected rings

Abstract: In a network of communication rings (R1 . . . R7, BB) interconnected by bridges (B1 . . . B4), access to each ring is regulated by a circulating token which is normally issued at irregular intervals depending on occurrence and length of messages. To enable communication of messages which have to be transmitted synchronously, i.e., at regular intervals, each ring has a synchronous bandwidth manager SBM which periodically issues a token for only synchronous information. Each SBM is located in a bridge node (13, 15) but only one of them which is located in a particular bridge node (15) is the master SBM from which all other SBM's are synchronized, using a synchronous token and/or special synchronization circuitry in each bridge. This allows a common period in the whole network for handling synchronous information, even if all rings have different bit rates.

Controlling the complexity of menu networks

Abstract: A common approach to the design of user interfaces for computer systems is the menu selection technique. Each menu frame can be considered a node in an information/action network. The set of nodes and the permissible transitions between them (menu selections) form a directed graph which, in a system of substantial size, can be large and enormously complex. The solution to this problem of unmanageable complexity is the same for menu networks as for programs: the disciplined use of a set of well-defined one-in-one-out structures. This paper defines a set of such structures and offers some guidelines for their use.

Apparatus for redundant operation of modules in a multiprocessing system

Abstract: A number of intelligent nodes (bus-interface units-BIUs and memory-control units-MCUs) are provided in a matrix composed of processor buses (105) with corresponding error-reporting and control lines (106); and memory buses (107) with corresponding error-reporting and control lines (108). Each node (100, 101, 102, 103) has means for logging errors and reporting errors on the error-report lines (106, 108). Processor modules (110) and memory modules (112) are each connected to a node which controls access to a common memory bus (107). Each node includes means (a married bit-170 and a shadow bit-172) for marrying modules in pairs such that each module in the pair tracks the operations directed to the module pair, and each module in the pair alternates with the other module in the handling of requests or replies. Each node registers the ID of the other node in a spouse ID register. Comparison logic (162, 164) in each node resets the married bit upon the condition that the node ID (identifying the node at which the error occurred) in an error-report message is equal to the ID stored in the spouse ID register, thus identifying the spouse node (the partner of the node in which the comparison logic is located) as the source of the error. Resetting the married bit splits apart the primary/shadow pair, so that the error-free module takes over and ceases to alternate with its partner.

A Cutset Approach to Reliability Evaluation in Communication Networks

Abstract: A global reliability measure called 'network reliability' (NR) is the probability that a call entering a probabilistic network at any originating node can reach every other node. This concept is quite useful in multiterminal networks such as computer networks and parallel processors, etc. A simple technique is presented for evaluating NR in symbolic form. The method is based on cutsets and is computationally advantageous with respect to the spanning tree approach. It requires fewer cutsets to be manipulated in the process of determining the NR expression. The number of cutsets is approximately half that of spanning trees even for a small sized computer communication network and there is a further improvement in the situation for larger networks.

Reliability Evaluation by Network Decomposition

Abstract: The method evaluates the reliability of large computer communication systems by systematic decomposition of the probabilistic graph of the system into two parts using an appropriate cutset. A technique is evolved for determining the conditional success events using both the node removal and connection multiplication methods for path enumeration. An example is solved to show the versatility of the method. The results of the example are verified by using an existing algorithm. The suggested method is general and computationally economical.

The throughput of a series of buffers

Abstract: Messages are to be transmitted through a series of n nodes linked by communication channels. The lengths of successive messages are independent identically distributed random variables, and the time taken to transmit a message through a channel is determined by its length. Each node has a finite buffer, and when the number of messages at a node reaches the buffer size transmission from the preceding node is interrupted. This paper is concerned with the maximum rate at which messages can pass through the system, called the throughput. We investigate the asymptotic behaviour of throughput as the series length increases, and determine the rate at which buffer sizes should grow to ensure that throughput does not decline to 0.

Method and apparatus for controlling access to a communication network

Abstract: A communication system comprises several transmission bus segments (A . . . G) to which data stations (ST) are attached. Node units (1 . . . 7) are provided for interconnecting the bus segments which are selectively activated in a predetermined sequence for authorizing message transmission by attached data stations according to any given access protocol. Thus, only small groups of stations contend for access, but all messages are distributed over the whole network. Activation of a bus segment is effected by an authorization message from an assigned node unit. A token message is used to pass the status enabling a node to activate its assigned bus segment, sequentially through all nodes of the network. Each node unit comprises a control unit (17) for sending and receiving token messages and segment activation messages, and further comprises unidirectional amplifier means and switches (11) for establishing a connection between two associated bus segments selectively in either direction.

System for resolving collision in local network

Abstract: The invention provides a method for resolving a data packet collision in a local network where a plurality of nodes which respectively comprise processors and terminal equipment utilize a single transmission line in accordance with a contention system. A contention phase is initiated unless a collision occurs. When a collision occurs, a resolution phase is initiated. The nodes which have been performing transmission immediately stop transmission. They discriminate their positions on the transmission line before a time interval (2×maximum transmission delay time interval) elapses from the start of transmission. If no signal is received by a given node from the left, the given node is regarded as the leftmost node among the conflicting nodes. The leftmost node immediately starts transmitting a packet. Other nodes except for the leftmost node stop retransmitting the packets. The nodes which have performed retransmission are stopped until the resolution phase is completed. The nodes except for the leftmost node receive retransmitted packets and immediately send out short reservation signals to the right. A node which does not receive the reservation signal from any other node during the reservation signal transmission is the leftmost node among the nodes which may not perform retransmission, and performs retransmission immediately after the reservation signal is sent out. Retransmission of the packet and transmission of the reservation signal are repeated. When the final conflicting node finishes retransmission, all of the nodes complete transmission. Each node continuously monitors the transmission line. When a nonsignalling period continues for more than a given time period (2×maximum propagating delay time interval), another contention phase is initiated.

Messages as active agents

Abstract: Network algorithms are usually stated from the viewpoint of the network nodes, but they can often be stated more clearly from the viewpoint of an active message, a process that intentionally moves from node to node. This paper gives some examples of this notion, and then discusses a means of implementing it. This implementation applied in both directions also demonstrates the logical equivalence of the two viewpoints.

A connectionist approach to visual recognition

Abstract: Strictly sequential approaches to computer vision are at best slow and cumbersome, at worst impossible. In this thesis, a massively parallel, connectionist approach is brought to bear on the problem of visual recognition. Computing with connections is a synthesis of results from Neuroscience, Computer Science, and Psychology. The fundamental assumption of connectionism is that individual computing units do not transmit large amounts of symbolic information. Instead, these units compute by being appropriately connected in a network of similar units. Using the communication pathways (connections) defined by the arcs of the network, the units cooperate and compete towards a globally consistent interpretation of the input scene. The problem, visual recognition, is defined as matching instances of predefined objects in the input with a fixed set of internal models. Predefined objects come from Kanade's Origami World{Kanade78}. The program represents and recognizes such pre-defined objects from line drawing input. To organize these networks, conceptual hierarchies are defined. A conceptual hierarchy is a semantic network hierarchically arranged according to abstraction levels. Levels represent the extraction of progressively more complex features. A node on a level, a computing unit, represents an instantiation of a feature defined on that level. Connections represent the composition and competition relations between feature units. Iterative relaxation is the form of control in the network. Each unit iteratively computes activation levels, a reflection of current confidence in the associated feature. Numerous testcases illustrate network behavior in presence of perfect, noised, incomplete and occluded input. The greatest benefits of this approach are seen in the systems ability to cope with incomplete and occluded input. Another advantage is an inherently parallel approach that eliminates the need for establishing a search order on exploring interpretations. The system exhibits many similarities with the structure and behavior of animal vision systems.