Showing papers on "Node (networking) published in 1981"

The Architectural Organization of a Mobile Radio Network via a Distributed Algorithm

Abstract: In this paper we consider the problem of organizing a set of mobile, radio-equipped nodes into a connected network. We require that a reliable structure be acquired and maintained in the face of arbitrary topological changes due to node motion and/or failure. We also require that such a structure be achieved without the use of a central controller. We propose and develop a self-starting, distributed algorithm that establishes and maintains such a connected architecture. This algorithm is especially suited to the needs of the HF Intra-Task Force (ITF) communication network, which is discussed in the paper.

Functionally Accurate, Cooperative Distributed Systems

Abstract: A new approach for structuring distributed processing systems, called functionally accurate, cooperative (FA/C), is proposed. The approach differs from conventional ones in its emphasis on handling distribution-caused uncertainty and errors as an integral part of the network problem-solving process. In this approach nodes cooperatively problem-solve by exchanging partial tentative results (at various levels of abstraction) within the context of common goals. The approach is especially suited to applications in which the data necessary to achieve a solution cannot be partitioned in such a way that a node can complete a task without seeing the intermediate state of task processing at other nodes. Much of the inspiration for the FA/C approach comes from the mechanisms used in knowledge-based artificial intelligence (Al) systems for resolving uncertainty caused by noisy input data and the use of approximate knowledge. The appropriateness of the FA/C approach is explored in three application domains: distributed interpretation, distributed network traffic-light control, and distributed planning. Additionally, the relationship between the approach and the structure of management organizations is developed. Finally, a number of current research directions necessary to more fully develop the FA/C approach are outlined. These research directions include distributed search, the integration of implicit and explicit forms of control, and distributed planning and organizational self-design.

Communication Structures for Large Networks of Microcomputers

Abstract: This paper compares nine network interconnection schemes and introduces "dual-bus hypercubes," a cost-effective method of connecting thousands of dual-port single-chip microcomputers into a room-sized information processing system, a "network computer." Each network node is a chip containing memory and a pair of processors for tasks and input/output. Nodes are linked by shared communication buses, each conceptually spanning a D-dimensional, W-wide hypercube of N = WD nodes. Each node shares two buses. Each bus is shared by up to W nodes. The number of bus connections per node is fixed to satisfy chip pin limitations.

The Distribution of Queuing Network States at Input and Output Instants

Abstract: Queuing networks are studied at selected points in the steady state, namely, at the moments when jobs of a given class arrive into a given node (either from the outside or from other nodes) and at the moments when jobs of a given class leave a given node (either for the outside or for other nodes). The processes defined by these points are known to be, in general, non-Potsson, interdependent, and serially correlated; therefore the relation between the distribution of the system state embedded at those moments and the steady-state (or random point) distribution is not obvious a priori. For a large class of networks having product-form equihbrium distribnttons it is shown that (a) if the given job class belongs to an open subchain, the state distributions at input pomts, output points, and random points are identical, and (b) if the job class belongs to a closed subchain, the distribution at input and output points ts the same as the steady-state distribution of a network with one less job in that subchain.

Method for the dynamic replication of data under distributed system control to control utilization of resources in a multiprocessing, distributed data base system

Abstract: A method for dynamic replication of data under distributed system control to control the utilization of resources in a multiprocessing, distributed data base system. Previously, systems providing for data replication at nodes of a multiprocessing, distributed data base system required that a central node maintain control, or that replicated data be synchronized by immediately conforming all copies of an updated data item. By this invention, requests for access to data of a specified currency are permitted and conformation of updated data is selectively deferred by use of a control procedure implemented at each node and utilizing a status and control (SAC) filed at each node which describes that node's view of the status for shared data items at other nodes.

Distributed dynamic programming

Abstract: We consider distributed algorithms for solving dynamic programming problems whereby several processors participate simultaneously in the computation while maintaining coordination by information exchange via communication links. A model of asynchronous distributed computation is developed which requires very weak assumptions on the ordering of computations, the timing of information exchange, the amount of local information needed at each computation node, and the initial conditions for the algorithm. The class of problems considered is very broad and includes shortest path problems, and finite and infinite horizon stochastic optimal control problems. When specialized to a shortest path problem the algorithm reduces to the algorithm originally implemented for routing of messages in the ARPANET.

Reliability of Algorithms for Pipe Network Analysis

Abstract: Algorithms for analyzing steady-state flow conditions in pipe networks are described for general applications. The algorithms are based on both loop equations expressed in terms of unknown flowrates and node equations expressed in terms of unknown heads. Five methods, which represent those in significant use today, are included. Using an extensive data base, describing a variety of pipe networks, the reliabilities of these algorithms for pipe network analysis were investigated. Numerous convergence and reliability problems were documented. Two methods based on loop equations have superior convergence characteristics. Methods based on node equations were less reliable; single adjustment methods must be employed with caution.

Digital data processing system having an I/O means using unique address providing and access priority control techniques

Abstract: A data processing system having a flexible internal structure, protected from and effectively 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 indentification of information as objects 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.

A DAG-Based Algorithm for Prevention of Store-and-Forward Deadlock in Packet Networks

Abstract: Store-and-forward deadlock (SFD) occurs in packet- switched computer networks when, among some cycle of packets buffered by the communication system, each packet in the cycle waits for the use of the buffer currently occupied by the next packet in the cycle. Several techniques for the prevention of SFD are known, but all exact some cost in terms of efficient and flexible packet handling. An ideal SFD-prevention technique is as unobtrusive as possible; it imposes no routing restrictions on packets, does not require that the buffer pool on each node grow with network size, and imposes no buffer-pool partitioning. All SFD-prevention techniques described so far lack some or all of these desirable properties. The new algorithm here described has all of them; in return, it imposes other unconventional costs. Under certain circumstances it requires that packets be rerouted around areas of potential deadlock, and one arbitrarily chosen node is required to accept within finite time any packet seeking entrance to its buffer pool, even if this requires erasing some packet. It is argued nonetheless that these costs are imposed infrequently enough and are sufficiently well manageable by heuristic techniques to make this new algorithm an attractive and practical alternative to the older techniques. An implementation designed for a microprocessor network now under construction is described.

Serial data bus communication system

Abstract: A node device for use in a digital data processing and communications system of the type utilizing a bus organization for facilitating the interconnection of a large plurality of digital data processing devices (user devices) in which redundant cables are employed. The node devices are interposed between the user devices and the redundant cables to permit automatic reconfiguration of the interconnection of the user devices in the event of malfunctioning or severing of one or more of the cable sets within a minimum period of time. The node devices provide the user devices with the structure needed to detect and diagnose system problems and to effect recovery procedures. In accordance with the invention, one of the plurality of nodes functions as the Bus Controller and by sampling the remaining nodes in the system, it determines the priority with which user devices may transmit or receive data over the bus. Each of the nodes employed is substantially identical and any one may be selected to function as the Bus Controller. Each includes a Microprogrammed Controller and necessary firmware to permit the controller to function in conjunction with special purpose hardware including a Programmable Logic Array (PLA) configured to perform a pre-processing function on control line inputs so that poll/response traffic on the active cables can continue with a minimum of delay.

Cryptographic application for interbank verification

Abstract: In a data communication network which includes terminals interconnected via a central switch, a process for verifying the identity of a terminal user who is provided with secret data associated with his identity. In carrying out the verification process, the secret data is first encrypted at the terminal under a transfer-in key for transmission to an associated data processing system. When it is determined that the terminal user maintains an account at the associated data processing system, a first translate operation is performed to translate the data from encryption under the transfer-in key to encryption under an authentication key, both of which keys are protected under other keys which are different from each other, thereby providing an authentication parameter which may be used to verify the identity of the terminal user. When it is determined that the terminal user does not maintain an account at the associated data processing system, a second translate operation is performed to translate the data from encryption under the transfer-in key to encryption under a transfer-out key for transmission to the next associated host system, the switch or a remote host system. At each such node, except the switch, a determination is made as to whether a verification process can be performed otherwise, the encrypted data is translated for transmission to the next or a remote node of the network for such verification.

On Locating a Facility to Service Areas within a Network

Abstract: Problems involving the selection of a site in a network at which to locate a facility, such as a station for a security guard force, generally fall into one of two categories, depending upon the nature of the facility to be established. One may be interested in minimizing the maximum travel time from the facility to any node in the network, or one may wish to minimize the sum of the distances from the facility to all the nodes in the network. In this paper problems involving travel time from the facility to areas within the network (that is, collections of points as opposed to individual points) are considered. This generalization of the problems appears not to alter their nature, and similar techniques for solutions can be used.

Digital B-trees

Abstract: A new tree index organization for files, capable of efficiently supporting both random and sequential access, is introduced. The organization, called digital B-tree (DB-tree), is similar in many aspects to B-trees. Its advantage is that it permits much larger fanout per node, thus reducing the height of the tree for a given file size. The effect of this is to reduce the cost of a random access to the file. The fanout of DB-tree nodes is increased substantially by permitting multiple page nodes. The unique advantage of DB-trees is that only one page of the node need ever be examined for each data access. This is accomplished by using the bits of the key to compute which page of the node is desired, in a way similar to the technique used in extendible hashing, but without performing a hashing operation. The DB-tree organization is described and analyzed. Particular algorithms are suggested for searching, building, and maintaining DB-trees.

Optical communication network

Abstract: An optical communication network includes a plurality of nodes and optical fibers for interconnecting the nodes whereby an optical signal may be transmitted from a source node to a destination node which may be determined selectively. Each node includes input channels, output channels and a connection control device for controlling the connection between the input and output channels. Terminal devices are provided to be in operative association with selected nodes. Thus, the present invention provides a high speed network having high reliability. The present network is easy to expand and its reliability increases as it is expanded.

Loop communication system

Abstract: A loop communication system, interconnecting a number of nodes around the loop, in which two oppositely directed unidirectional loops service the two directions of transmission. Each of the loops comprises a time division multiplexed transmission facility. Each of the nodes has facilities for connecting the communications on each incoming channel to a utilization device connected to that node, for connecting the communication from a utilization device connected to that node to each outgoing channel, or for repeating the communication from any incoming channel to a corresponding outgoing channel. Assignment of a communication to a given channel is accomplished using a centralized control means which causes the communicating devices to be connected using corresponding channels in the two loops, and using that portion of the loop which has the smaller number of intermediate nodes.

Selected node-pair analysis in dial's assignment algorithm

Abstract: An algorithm is proposed to calculate the matrix of trips which use a particular link in a road network or alternatively pass through a particular set of nodes given the route-splitting assumptions inherent in the Dial assignment algorithm.

The architecture of Tandem's NonStop system

Abstract: A Tandem T16 computer system is a network of up to 256 nodes. Each node consists of two to sixteen processors. The system had three major design goals: (1) Continuous data availability. (2) Modular growth by adding processing elements to a node. (3) Support of a network of geographically distributed nodes for on-line transaction processing. This talk will sketch Tandem's approach to continuous data availability. At the hardware level, NonStop is achieved by designing a single-fault tolerant system. A Tandem system has two or more modules and paths for each function. In addition, the system design addresses issues such as power failure, on-line maintenance, and reconfiguration.

A Technique for Adaptive Routing in Networks

Abstract: A two-level adaptive routing scheme for packet-switched computer communication networks is proposed and investigated. The first level is quasi-static and based on the global network status. The second level is dynamic with decisions being made at each node in an attempt to obtain the savings in average delay predicted by a multiserver model of the node. Simulations confirm the predicted improvement.

An Evaluation of Local Path ID Swapping in Computer Networks

Abstract: This paper analyzes a method for identifying end-to-end connections in computer networks which is designed to provide reductions in the sizes of the packet headers and routing tables stored in the nodes. The method, known as Local Path ID Swapping, uses a shortened connection identifier, called the LPID, in the message headers and routing tables. In general, the LPID field is swapped in the message header from node to node along the path of the route. Some analytical results are presented for evaluating the important tradeoffs involved in LPID swapping. Most notable is the tradeoff between the size of the LPID field and the number of connections which can be defined in the network.

Repositioning of Two Distinguishable Service Vehicles on Networks

Abstract: The problem of repositioning urban emergency service vehicles on the transportation network is discussed. Repositioning problems deal with real-time movement of available servers to better anticipate short-term future requests for service. It is assumed that there are two distinguishable service units in the network and that repositioning can take place to any node of the network. The objective is to find the optimum repositioning policy that minimizes the long-term expected cost (total travel time) of operating the system.

An IPC protocol and its hardware realization for a high-speed distributed multicomputer system

Abstract: Multicomputer systems with distributed control form an architecture that simultaneously satisfies such design goals as high performance through parallel operation of VLSI processors, modular extensibility, fault tolerance, and system software simplication. The nodes of the system may be locally concentrated or spatially dispersed as a local network. Applications range from data base-oriented transactional systems to “number crunching.” The system is service-oriented; that is, it appears to the user as one computer on which parallel processing takes place in the form of cooperating processes. Cooperation is regulated by the unique interprocess communication (IPC) protocol presented in this paper. The high-level protocol is based on the computer/producer model and satisfies all requirements for such a distributed multicomputer system. It is demonstrated that the protocol lends itself toward a straightforward mechanization by dedicated hardware consisting of a cooperation handler, an address transformation and memory guard unit, and bus connection logic. These special hardware resources, assisted by a “local operating system”, form the supervisor of a node. Nodes are connected by a high-speed bus (280M bit/sec). Programming aspects as implied by the IPC protocol are also described.

Filtering formulas and the ./M/1 queue in a quasireversible network

Abstract: This paper proposes an elementary justification of the filtering formulas for a Markov chain and an analysis of the arrival and departure processes at a ./M/1 queue in a quasireversible network. It is shown that the interarrival time distributions of the two above mentioned processes are always identical under equilibrium. This generalizes the corresponding result proved for Jackson networks in [1]. An example shows that the those two prrcesses do not necessarily have the same law even if the network outside of the o M,T node is reversible and if there is no immediate feedback on that node. This contradicts a conjecture made in [2].

Optimal constructions of event-node networks

Abstract: — There are two types of networks in the scheduling and planning which represent a project t, e., the activities together with their precedence relations, namely, the activity networks and theevent networks. For each project, there exists a unique activity network without redundant arcs but since there îs an infinité numher of different sized eventnetworks, the problem is tofind an event network with the minimum number of dummy activities, The motivation behind this problem is to minimize the time of the analysis of a network which is proportionai to the number of activities, induding the dummy ones. Krishnamoortky and Deo proved thatthis problem is NP-complete, In sections 2 and 3 we characterize activity networks for which there exist event networks without dummy activities and we show that the question whether a given activity network requires dummy activities in the event network can be answered in poiynomial time. We review some algorithmsforfinding an optimal event network and a new approach is presented which gives rise ta an approximate algorithm and can lead to an optimal branchand-bound method. Some generalizations of this reaï-world problem are also considérée.

A hardware support mechanism for scheduling resources in a parallel machine environment

Abstract: The allocation of resources in an array computer using either an SW-banyan or a tree interconnection network is implemented in hardware. A generalized solution for the resource allocation problem in such machines is presented. The solution reduces network node blockage and node distances and supports differing network topologies and decentralized scheduling. The mechanism presented is shown to reduce significantly the amount of computational work required in software allocation of resources.

Configuration-Dependent Performance of a Prioritized CSMA Broadcast Network

Abstract: The interaction of node placement and contention mechanism is crucial to network performance. These experiments suggest that elaborate contention schemes have but little advantage over simpler ones.

Optimal Placement of Identical Resources in a Distributed Network.

Abstract: : The problem is considered of locating a number of identical resources at nodes of a tree so as to minimize the total expected cost of servicing a set of random requests for the resources. The cost of servicing a request is the tree distance from the requesting node to the node at which the resource satisfying the request is located. An algorithm for finding an optimal placement of resources is presented.

A simulation model of the MICRONET computer system during join processing

Abstract: The MICRONET computer system was specifically designed for processing distributed relational databases. This paper describes a simulation study of the queueing and resource utilization of this system during processing of a relational join operation. Problems associated with representing a computer network in a simulation model are presented. In our model, one set of simulation facilities can represent any node of interest in the network. The processing at each node involves competition for resources by the data bus and the join algorithm. This was represented by processes which were synchronized by Wait Event and Post Event primitives. An algorithm for interrupting a non-shareable resource (e.g. the cpu) was also developed. Results for various system loadings and database sizes are presented.