Showing papers on "Geographic routing published in 1988"

Inter-computer message routing system with each computer having separate routinng automata for each dimension of the network

Abstract: In a multicomputer, concurrent computing system having a plurality of computing nodes, this is a method and apparatus for routing message packets between the nodes. The method comprises providing a routing circuit at each node and interconnecting the routing circuits to define communications paths interconnecting the nodes along which message packets can be routed; at each routing circuit, forming routes to other nodes as a sequence of direction changing and relative address indicators for each node between the starting node and each destination node; receiving a message packet to be transmitted to another node and an associated destination node designator therefor; retrieving the route to the destination node from a memory map; adding the route to the destination node to the beginning of the message packet as part of a header; transmitting the message packet to the routing circuit of the next adjacent node on the route to the destination node; and at each intermediate node, receiving the message packet; reading the header; directing the message packet to one of two outputs thereof as a function of routing directions in the header, updating the header to reflect passage through the routing circuit; and at the destination node, stripping remaining portions of the header from the message packet; storing the message packet; and, informing the node that the message packet has arrived.

Neural networks for routing communication traffic

Abstract: The use of neural network computational algorithms to determine optimal traffic routing for communication networks is introduced. The routing problem requires choosing multilink paths for node-to-node traffic to minimize loss, which is represented by expected delay or some other function of traffic. The minimization procedure is implemented using a modification of the neural network traveling-salesman algorithm. Illustrative simulation results on a minicomputer show reasonable convergence in 250 iterations for a 16-node network with up to four links from origin to destination. >

Designing networks with compact routing tables

Abstract: Classes of network topologies are identified in which shortest-path information can be succinctly stored at the nodes, if they are assigned suitable names. The naming allows each edge at a node to be labeled with zero or more intervals of integers, representing all nodes reachable by a shortest path via that edge. Starting with the class of outerplanar networks, a natural hierarchy of networks is established, based on the number of intervals required. The outerplanar networks are shown to be precisely the networks requiring just one interval per edge. An optimal algorithm is given for determining the labels for edges in outerplanar networks.

Hypercube message routing in the presence of faults

Abstract: We discuss the problem of routing messages on hypercubes which have faulty processors and/or communication links. We are motivated by the belief that simple algorithms, operating under simple assumptions, can ensure high probabilities of successful message routing. In this paper, we consider the basic problem of routing a single message from an arbitrary source to an arbitrary destination. In our study, a fault is assumed to render the processor or link non-functional for purposes of communicating messages. As such, we may also consider communications hot spots as node faults, and our results also apply to routing in congested hypercubes.A framework for the analysis of fault tolerant routing schemes on a hypercube is presented. This framework includes differing routing schemes, routing information models and fault distribution models. The a priori probabilities of successful routing of a single, indivisible message under each of our possible sets of assumptions are calculated. Using random routing, under the one-step local information routing model, we show that the a priori probability of successful message routing is high even for an exceedingly large number of faults. We also analyze the behavior of sidetracking, a routing method which combines the concepts of local information and randomization. Using sidetracking, and in the one-step local information routing model, a message will be routed forward using random routing. If the message reaches a blocked processor (no non-faulty neighbors along a minimal path to the destination) it will be sent to a non-faulty neighbor, chosen uniformly at random from the set of non-faulty neighbors. We use simulation experiments to determine the performance of this routing scheme, analyzing the probability of successful routing and the expected path length of a routed message. The empirical performance of the sidetracking algorithms indicates strongly that, in the limit as the cube dimension grows larger and for a fixed probability of node failure, the probability of successful message routing is 100%.

Hyperswitch network for the hypercube computer

Abstract: The performance of a parallel algorithm depends in a large part on the interconnection topology of the multicomputer system. The method presented in this paper realizes a kind of interconnection network, called a hyperswitch network, that is achieved using a mixture of static and dynamic topologies. Here, available or fault free paths need not be specified by a source because the routing header can be modified in response to congestion or faults encountered as a path is established. This method can be accomplished in a static topology such as the hypercube network if the nodes have switching elements which are capable of performing the necessary routing header revisions dynamically. Detailed simulation results show that the hyperswitch network is consistently more efficient than fixed path routing for large message traffic conditions. The simulation results also show that the hyperswitch network has equivalent latency overhead for messages with localized and antilocal destinations (i.e., less then a 25% difference between diameter 1 and 5).

Routing and broadcasting in faulty hypercube computers

Abstract: This paper examines routing and broadcasting algorithms for hypercube computers subject to node failures. First some simple message-passing algorithms are described which perform well with certain fault patterns, but poorly with others. The concept of an unsafe node is introduced to identify fault-free nodes that may cause communication difficulties in faulty hypercubes. It is then shown that by only using “feasible” paths that try to avoid unsafe nodes, routing and broadcasting can be substantially simplified. It is assumed that each active node is supplied with the fault status of all neighboring nodes within a specified radius k. A computationally efficient routing algorithm is presented which can route a message via a path of length no greater than p+2, where p is the minimum feasible distance from the source to the destination, provided that not all non-faulty nodes in the hypercube are unsafe, and k = 1. We further show that broadcasting can be achieved under the same fault conditions with only one more time unit than the fault-free case.

Source routing bridge implementation (LANs)

Abstract: Source routing is a bridging technique where the bridge makes a routing decision based on the contents of the media access control (MAC) frame header of the frame. This header consists of a routing information field, which immediately follows the source address field in the frame. The routing information field describes the path across one or more bridges to the ring containing the destination station. With source routing bridges, the source and destination addresses of the frame on the media are the MAC addresses of the originating and target stations, respectively. The operation of a source routing bridge is examined. An implementation example using the TMS380 token-ring-adapter chipset is presented. >

Decentralized adaptive routing for virtual circuit networks using stochastic learning automata

Abstract: The problem of routing virtual circuits according to dynamical probabilities in virtual-circuit packet-switched networks is considered. Queueing network models are introduced and performance measures are defined. A decentralized asynchronous adaptive routing methodology based on learning automata theory is presented. Every node in the network has a stochastic learning automaton as a router for every destination node. The routing probabilities that are assigned to the network paths are updated asynchronously on the basis of current network conditions. A learning algorithm suitable for routing is used. Some initial simulation experiments, for a simple network, show convergence to optimal routing. >

Packet forwarding

Abstract: A packet network is viewed as consisting of two major interacting layers: a lower layer responsible for the determination of a set of paths that can be used to carry packet flows, and an upper layer responsible for actually sending the flows over these paths, on a per-packet basis. A systematic review is provided of the various techniques that can be used to perform the actual packet-by-packet routing operation at the upper layer, called the packet forwarding function. The function components of a routing system and its characteristics (responsiveness to changes in network state and degree of centralization) are discussed. Five routing techniques are then examined, namely source, directory, destination, global-path, and channel-link-path routing. >

Analysis of a 3D toroidal network for a shared memory architecture

Abstract: Horizon is a shared-memory multiple-instruction-stream-multiple-data-stream architecture currently under development. A synchronized network model suitable for this architecture is described. The model is defined in terms of a topology and routing policy. A three-dimensional toroidal topology is investigated for its multiple redundant paths, memory locality, and simplicity in both routing and construction. The routing policy is based on a desperation routing scheme in which it is not guaranteed that a message will make progress on a given network cycle. This scheme requires no complex deadlock avoidance algorithms or node-to-node flow control, allowing a very simple and efficient implementation. Design considerations and preliminary performance results are discussed. >

Distributed routing using topology database in large computer networks

Abstract: The authors develop distributed routing algorithms for large computer networks that provide virtual circuit connections to end users. One routing scheme in computer networks is to use a topology database maintained at each node that contains information about nodes and link cost. Using this information, each node finds shortest routes to other nodes. This scheme may be inefficient in large networks since the overhead associated with topology database update and shortest path computation may become large. One way of solving this problem is to partition the network into subnetworks or clusters. After partitioning, each node in a cluster only maintains a topology database of nodes and links in its own cluster and possibly marginal information about the entire network. For intracluster communication, shortest path routes are provided. For intercluster communication the source node collaborates with other nodes to find a path to the destination node. The authors propose two collaboration schemes that use simple protocols among network nodes. In one scheme, a broadcast search is used to find a path to the destination node in other clusters. In the other scheme each node maintains information about intercluster links on an up/down basis as well as complete topology database of its own cluster. Using this information, it finds a path to the destination node with the help of border nodes (nodes sharing intercluster links) in the intermediate clusters. >

Spatial reuse through dynamic power and routing control in common-channel random-access packet radio networks

Abstract: Topologies of common-channel packet radio networks (PRNETs) are difficult to optimize because some of the links between multiple pairs of packet radio units are not independent. Previous analysis has shown that designing the topology to provide spatial reuse of the common-channel will improve the network throughput and delay performance in general. Unfortunately, the complexity of the link interactions has impeded the design of protocols that can be implemented in operational networks. This dissertation discusses how to optimize the topologies of common-channel random-acess PRNETs through dynamic power control at the link layer and routing at the network layer. Methods of implementing dynamic power control at the link layer on an individual packet-by-packet transmission basis are presented. These methods should be implementable at the link layer of any packet radio with dynamic per-packet power control capability. A new routing protocol, called Least Interference Routing (LIR), is defined which is designed specifically to operate in common-channel random-access PRNETs. The goal of LIR is to minimize the destructive interference caused along each route within the network, thus improving the spatial reuse of the common-channel. The LIR protocol calculates the potential destructive interference along each link, creates the network routing tables that minimize the potential destructive interference along an entire route, and specifies the per-packet transmission power. The implementation flexibility of each of these operations allows LIR to be implemented in a variety of radios and radio networks. Myopic one-hop and network multiple-hop simulations indicate that dynamic power control and/or LIR improve end-to-end PRNET performance over no power control or other routing strategies, such as minimum hop routing.

Connectivity biased channel construction and ordering for building-block layout

Abstract: A number of techniques are presented for the construction and ordering of routing channels for building-block layout. First, before the routing channels are defined the placement is modified such that proper routing space is assigned between the circuit blocks. Second, a channel graph is constructed on which the global routing will be performed. Finally, after the global routing a feasible routing order is assigned to the channels. In contrast to other works, the algorithms use both the geometrical data (the placement) and the topological data (the connectivity) to decide which channel structure should be chosen from the feasible set.

The routing and cost of the information flow in a system

Abstract: A study of information flows in different types of procedures used for the management of the agricultural credit in the Agricultural Bank of Greece is discussed in this paper. Systems and subsystems are structured using new diagrammatical notation in order to identify explicitly the system. The communication network of the information flow with certain nodes (individuals) and types of links (types of communication) is also considered. In addition to the normal link cost of such diagrams, a node cost, which depends on the types of links involved in the network, is included. An optimal routing from a given node to its destination node can thus be sought. The conditions of stability and change of the system are examined from the information flow cost point of view through the procedures associated with short-term agricultural loans.

Hierarchical clustering with topology databases

Abstract: Dynamic routing in networks helps to provide efficient utilization of network resources. However, when networks are too large, the overhead introduced by dynamic routing consumes too much of network resources (processing, transmission capacity, and storage). To overcome this, researchers have investigated techniques to hierarchically organize networks and thereby reduce overhead. This paper introduces a new technique to organize routing information in large networks. A “radius database” provides optimal routing between frequent communicators. The combination of a “cluster database”, and a backbone, allows reasonable routing to all nodes. The paper details the information maintained at each node, describes distributed update algorithms to allow dynamic maintenance of the information, and indicates how to cluster networks to take advantage of this techniques. Simulation results support the efficacy of the technique.

On routing and performance comparison of techniques for packet-switched networks using learning automata

Abstract: Distributed adaptive routing algorithms with the ability to adaptively proportion traffic over several paths is proposed for packet-switched data networks. A learning automaton is situated at each node of the network where a routing decision must be made and directs traffic entering the node onto one of the outgoing links. Using network feedback, and automaton modifies its routing strategy to improve its link selections. This approach has the advantage over existing routing schemes of offering a simple and extremely practical feedback and updating policy. >

Message routing schemes in a hypercube machine

Abstract: The aim of this study is to determine efficient routing schemes for message passing in a hypercube machine. Two different algorithms are considered, namely static routing in which the path of a message is predetermined by the addresses of the source and destination nodes, and dynamic routing where the decision as to the next node in the path is made by the current node on the basis of local information regarding queue lengths. In addition, various different prioritization schemes are compared for both static and dynamic routing. The results show that dynamic routing can be up to twice as efficient as static routing, provided priority is given to messages which have only a few hops to traverse or were transmitted early in the computation sequence.

On routing and performance comparison of techniques for

Abstract: Distributed adaptive routing algorithms with the ability to adaotively proportion traffic over several paths is proposed for packet-switched data networks. A learning automaton is situated at each node of the network where a routing decision must be made and directs traffic entering the node onto one of the outgoing links. Using network feedback, an automaton modifies its routing strategy to improve its link selections. r'his awproach has the advantage over existing routing schemes of offering a simple and extremely practical feedback and Updating policy.

Relay approach message routing in hypercube multiprocessors

Abstract: A relay approach message routing for hypercube multiprocessors is proposed which is able to forward a received message immediately after the address fields of the message are received if the outgoing channels are not blocked. The relay message routing can significantly reduce the communication latency compared with the store-and-forward message routing. Based on the relay approach, a distributed routing mechanism which is able to handle unicast, multicast, and broadcast interprocessor communications is proposed. These three types of interprocessor communication are modeled as an Optimal Multicast Tree problem which can guarantee a minimum message delivery time. Network traffic is minimized for the cases of unicast and broadcast and close to minimum for the case of multicast. A hardware router design based on the proposed distributed relay routing is discussed.

A survey of hierarchical routing algorithms and a new hierarchical hybrid adaptive routing algorithm for large scale computer communication networks

Abstract: A survey is presented of present algorithms, with emphasis of a proposed scheme, the hierarchical hybrid adaptive routing algorithm (HHARA). HHARA reduces the size of routing database by dynamically organizing nodes into hierarchically structured clusters so that only partial information is stored and maintained in each site. The responsibility for routing is shared by the routing hierarchy so that the algorithm can maintain the global routing optimality and the local adaptivity at the same time. In this way, the storage, maintenance, communication, and computation overheads can be reduced while the response time to local status changes is kept small. HHARA can be used with either diagram routing or virtual circuit routing. Due to its ability to adapt to the network changes, it is specially useful for dynamic networks such as large military computer communication networks. >

Application of AI techniques to adaptive routing in wide-area networks

Abstract: Adaptive decentralized routing in wide-area networks is characterized by all nodes making routing decisions based on their current view of the status of remote nodes. Unfortunately, each node's view can be, and usually is, different, especially when the network is very large. The authors introduce the idea of cooperating intelligent agents, embedded within nodes, which infer the states of remote nodes using all relevant local information including knowledge of past behavior, to reduce explicit communication of remote status information. >

Fault tolerant message routing on large parallel systems

Abstract: The problem of designing massively fault-tolerant message routing schemes for large parallel systems is considered. The notion of faults is extremely flexible and applies to all situations where a component is unavailable to participate in message communications. Attention is focused on the performance of schemes which use only local information to make local decisions. A framework for the analysis of fault-tolerant routing schemes is presented and used to analyze the efficacy of minimal path routing methods. Fault-tolerant routing schemes are derived by application of a technique called sidetracking. Viewed as making local decisions, a sidetracking scheme attempts to decrease the distance to the destination: if this is not possible, then the packet is routed randomly so as to increase the distance as little as possible. For single-message routing on a hypercube, it is shown that the performance of a sidetracking scheme is near optimal, successfully routing with high probability and low average excess delay. Applications of the sidetracking technique to single-message routing on a two-dimensional mesh and to multiple-message permutation routing on a hypercube are presented. >

Networks of queues with service anticipating routing

Abstract: Queueing networks are studied with general service requirements and routing depending upon the next required service time. A non-standard product form expression is obtained that averages out dependencies of routing and service times.

Application of AI Techniques to Adaptive Routing

Abstract: Adaptive decentralized routing in wide-area networks is characterized by all nodes making routing decisions based on their current view of the status of remote nodes. Unfortunately, each node's view can be, and usually is, different, especially when the network is very large. We introduce the idea of cooperating intelligent agents, embedded within nodes, which infer the states of remote nodes using all relevant local information including knowledge of past behavior, to reduce explicit communication of remote status information.