Showing papers on "Routing table published in 1990"

Addressing, routing, and broadcasting in hexagonal mesh multiprocessors

Abstract: A family of six-regular graphs, called hexagonal meshes or H-meshes, is considered as a multiprocessor interconnection network. Processing nodes on the periphery of an H-mesh are first wrapped around to achieve regularity and homogeneity. The diameter of a wrapped H-mesh is shown to be of O(p/sup 1/2/), where p is the number of nodes in the H-mesh. An elegant, distributed routing scheme is developed for wrapped H-meshes so that each node in an H-mesh can compute shortest paths from itself to any other node with a straightforward algorithm of O(1) using the addresses of the source-destination pair only, i.e. independent of the network's size. This is in sharp contrast with those previously known algorithms that rely on using routing tables. Furthermore, the authors also develop an efficient point-to-point broadcasting algorithm for the H-meshes which is proved to be optimal in the number of required communication steps. The wrapped H-meshes are compared against some other existing multiprocessor interconnection networks, such as hypercubes, trees, and square meshes. The comparison reinforces the attractiveness of the H-mesh architecture. >

Method and system for routing packets in a packet communication network

Abstract: A packet network routing method and system based on geographic coordinate identifiers is described. Each node in a network is uniquely identified by absolute geographical coordinates or by a code indicating absolute location in an external coordinate-based reference system (node coordinates), and such absolute geographical coordinates or the equivalent are employed as part of a packet identifier for each packet generated for use in making routing decisions. The node coordinates of a local node and its neighboring nodes and the packet coordinates are used by means at each node through which a packet is routed for determining a desired forwarding route of a data packet. The routing may be prioritized according to preselected criteria, preferably achieving maximum forward progress using the least amount of power. The packet routing protocol according to the invention requires no routing directory or table to perform data routing.

Dynamic routing system for a multinode communications network

Abstract: A communication network having a multiplicity of nodes provides efficient exchange of messages between the nodes. The messages may be originated and received by the computers of a parallel computer system, the processors and associated memories of which are connected to each node. Each node includes a routing system which results in efficient system performance for the parallel computer system associated with the nodes. The messages have control information (a message tag) to which the routing system at each node is responsive. The tag contains data identifying the destination node of the message in the network and prioritization data which, when the message is generated, is initialized to have a value (weight) corresponding to the length (number of links) of the minimal path from the source node where the message originates to the destination node of the message. The routing system utilizes the weights to establish message priority. The routing system receives and sends messages not exceeding the number of links connected thereto on each cycle such that messages flow in and flow out of each routing node on each cycle without being held or stored in queues in the node. Messages of lesser priority are switched by the router to alternate links in accordance with their weights thereby dynamically routing and resolving conflicts among messages.

A detailed router for field-programmable gate arrays

Abstract: A detailed routing algorithm, called the coarse graph expander (CGE), that has been designed specifically for field-programmable gate arrays (FPGAs) is described. The algorithm approaches this problem in a general way, allowing it to be used over a wide range of different FPGA routing architectures. It addresses the issue of scarce routing resources by considering the side effects that the routing of one connection has on another, and also has the ability to optimize the routing delays of time-critical connections. CGE has been used to obtain excellent routing results for several industrial circuits implemented in FPGAs with various routing architectures. The results show that CGE can route relatively large FPGAs in very close to the minimum number of tracks as determined by global routing, and it can successfully optimize the routing delays of time-critical connections. CGE has a linear run time over circuit size. >

Semi-dynamic load balancer for periodically reassigning new transactions of a transaction type from an overload processor to an under-utilized processor based on the predicted load thereof

Abstract: A semi-dynamic load balancer for a transaction processing system reallocates transaction types among computers in the system as a group rather than as individual transactions. Statistical data is accumulated in an affinity matrix which records the number of times a transaction type i was blocked by transaction type j was holding a lock on a data item that transaction type i wanted to access. When an overloaded computer is detected, transaction types are reallocated on the computers of the system according to a transaction type routing table which is updated based on data from the affinity matrix.

Routing method and routing system for switching system having a plurality of paths

Abstract: A routing method routes cells which are transferred through one of a plurality of paths within an asynchronous transfer mode (ATM) switching system (50, 10) selected by routing information. The routing method includes the steps of adding first routing information to incoming cells at an input stage of the ATM switching system when supplying the cells to the ATM switching system, generating a route switching confirmation cell in response to a route switching instruction and for adding second routing information to incoming cells thereafter so as to supply the route switching confirmation cell and the cells added with the second routing information to the ATM switching system, where the second routing information is different from the first routing information and is determined by the route switching instruction, comparing routing information of the cells with the second routing information at an output stage of the ATM switching system in response to the route switching instruction, outputting each cell from the ATM switching system having routing information different from the second routing information as it is, and temporarily storing each cell from the ATM switching system having routing information identical to the second routing information and reading out and outputting the stored cell after the route switching confirmation cell is output from the ATM switching system.

Apparatus and method for detecting and eliminating call looping in a node-by-node routing network

Abstract: An apparatus and method for effectively managing a switched virtual circuit network by detecting and resolving call routing problems occurring in the routing pattern of the network is provided. Detecting the call routing problems is advantageously achieved by a first process contained in a network management center which has access to all nodes in the network and performs routing table verification for each node whenever there are changes in the routing tables caused by changes in the network, such as an addition of nodes, links and the like. Otherwise, call looping may occur during call set-up as a result of any undiscovered or uncorrected errors in one or more routing tables in the network. Detection of call loops is achieved by the first process through use of a plurality of processing modules that recognizes and excludes in their analysis of the network certain network components that can not be involved in call looping. Identified routing problems are resolved by a second process which provides alternative recommendations as to how best to modify the routing tables in the nodes of the network. A selected one of the alternative recommendations is coupled from the network management center to the nodes in the network for changing the routing tables located at the respective nodes.

Locally-optimal deflection routing in the Bidirectional Manhattan Network

Abstract: A class of shortest-path distributed-routing techniques is presented, and the authors show a locally optimal solution under uniform traffic assumptions. It is argued that, as the number of nodes grows to infinity, the throughput tends to the limit represented by the maximum throughput attainable with the store-and-forward technique and infinite queueing storage. The results obtained show the suitability, as far as the throughput is concerned, of meshed networks using packet switching without storing packets at nodes. The routing techniques investigated perform similarly. At least one of these is simple enough to be implemented with very-fast circuitry, able to cope with the speed foreseen in future metropolitan-area networks (MANs). >

Space-efficient message routing in c -decomposable networks

Abstract: The problem of routing messages along near-shortest paths in a distributed network without using complete routing tables is considered. It is assumed that the nodes of the network can be assigned suitable short names at the time the network is established. Two space-efficient near- shortest path routing schemes are given for any class of networks whose members can be decomposed recursively by a separator of size at most a constant c, where $c \geqslant 2$. For an n-node network, the first scheme uses a total of $O(cn \log n)$ items of routing information, each $O(\log n)$ bits long, and $O(\log n)$-bit names, generated from a separator-based decomposition of the network, to achieve routings that are at most three times longer than shortest routings in worst case.l The second scheme augments the node names with $O(c \log c \log n)$ additional bits and uses this to reduce the bound on the routings to $(2 / \alpha) + 1$, where $\alpha, 1 < \alpha \leq 2$, is the root of the equation $\alpha ^{\lceil (c+1) / ...

An analysis of 'hot-potato' routing in a fiber optic packet switched hypercube

Abstract: Two implementations of a fiber-optic packet-switched hypercube are proposed. In the first, each directed link is implemented with a fixed wavelength laser and photodetector, and all optical transmissions are wavelength multiplexed onto one or more fibers. In the second, the electronic crosspoint matrices within the nodes are eliminated by allowing each laser to be tunable over a range of log N wavelengths. Assume that a hot potato, or deflection, routing algorithm is used; as soon as a packet is received at a node, a routing decision is made and the packet is sent out. The node attempts to send the packet towards its destination. The analysis indicates that a hypercube, hot-potato routing offers essentially optimal performance for random traffic, regardless of how large the hypercube grows, and it significantly outperforms traditional shortest-path routing with buffering and flow control. A few variations, including an algorithm which gives priority to packets closer to their destinations and one which gives priority to various classes of traffic, are also proposed and analyzed. >

The chaos router: a practical application of randomization in network routing

Abstract: We present the chaos router, an asynchronous adaptive router, which under certain circumstances can send messages farther from their destinations. The chaos router greatly simplifies the routing logic by removing the livelock protection of previous schemes. Through an effective use of randomness, whose sources include that due to the adaptively processed load, the natural timing differences of selftimed circuitry and explicitly injected randomization, the chaos router avoids long message routes with high probability. In this paper the router is described, it is argued that the chaos router is deadlock free and probabilistically live-lock and starvation free, and simulation results are presented showing that the chaos router performs well.

General models and algorithms for over-the-cell routing in standard cell design

Abstract: When an over-the-cell routing layer is available for standard cell layout, efficient utilization of routing space over the cells can significantly reduce layout area. In this paper, we present three physical models to utilize the area over the cells for routing in standard cell designs. We also present efficient algorithms to choose and to route a planar subset of nets over the cells so that the resulting channel density is reduced as much as possible. For each of the physical models, we show how to arrange inter-cell routing, over-the-cell routing and power/ground busses to achieve valid routing solutions. Each algorithm exploits the particular arrangement in the corresponding physical model and produces provably good results in polynomial time. We tested our algorithms on several industrial standard cell designs. In our tests, this method reduces total channel density as much as 21%.

Distributed algorithms for shortest-path, deadlock-free routing and broadcasting in arbitrarily faulty hypercubes

Abstract: The authors present a distributed table-filling algorithm for point-to-point routing in a degraded hypercube system. This algorithm finds the shortest length existing path from each source to each destination in the faulty hypercube and fills the routing tables so that messages are routed along these paths. A novel scheme for broadcast routing with tables is proposed, and the algorithm required to fill the broadcast tables, given the point-to-point routing tables, is presented. In addition, the modifications necessary to make these algorithms ensure deadlock-free routing are given. A quantitative and equalitative comparison of previously proposed reroute strategies with table routing, where the tables are filled by the authors' algorithms, are presented. >

Hybrid routing

Abstract: A general-purpose routing algorithm for very-large-scale integrated (VLSI) circuits and printed circuit board (PCB) designs is proposed. Ideas behind the maze-running algorithm and the hierarchical routing algorithm are combined into a powerful algorithm called hybrid routing. The new algorithm demonstrates a speed compatible to a hierarchical router and produces routings with quality equivalent to that obtained by a maze router. Hybrid routing is based on the maze-running method with a third search dimension added. The extra search space is built by recursively constructing a hierarchy of coarser grid meshes. By means of a parameter-controlled expansion into the coarser meshes, the hybrid router is able to find the preferred search region very quickly and will not miss local information as a hierarchical router does. A user-given parameter can turn the algorithm into a pure maze router, a pure hierarchical router, or a wide spectrum of hybrid routers with different speed/quality characteristics between the extremes. The algorithm has been implemented and integrated into a global router that can handle large-scale routing, such as that encountered in the sea-of-gates layout. >

Maximum free circuit routing in circuit-switched networks

Abstract: An analysis is made of an alternate-path routing rule called maximum free circuit routing (MFCR). In the use of MFCR, a call is routed to the alternate path that has the maximum number of free circuits when the direct path is blocked. Analytical results show that in conjunction with trunk reservation, this routing rule can offer a stable throughput at high traffic conditions and can increase the call carrying capacity by about 20% (compared to direct path routing) under a blocking requirement of 10/sup -2/ on a fully connected symmetrical nonhierarchical network. >

Distributed dynamic routing schemes

Abstract: Schemes that do not explicitly use much information about the state of networks are briefly surveyed, with the focus on dynamic alternative routing (DAR), a simple but highly effective routing method currently planned for the British Telecom Network. State-dependent routing and how some of the methodology also has bearing on the control issue are discussed. The problem of dimensioning a network that uses dynamic routing (i.e. how much capacity is needed and where it should be put to provide an acceptable performance) is addressed. A practical example, which refers to routing in an international access network, is discussed. Some conclusions are drawn on the benefits and drawbacks of distributed routing. >

Topology distribution cost vs. efficient routing in large networks

Abstract: Routing a message in a network is efficient (in terms of weight of the path used to carry the message) when nodes know the full topology of the network. This may not be the case in large networks since a network may be composed of smaller autonomous pieces by design or by requirements on performance, with each piece having less than complete information about other pieces. We present a trade-off between the amount of topology information exchanged among these pieces and the efficiency of routing in the network. The large network that we study is a collection of networks connected by boundary nodes. Each boundary node knows the topology of its network and the connectivity of networks to each other. The question addressed here is how much topology information about each network should be distributed to other networks in order to achieve reasonably efficient routing.

Design of inter-administrative domain routing protocols

Abstract: Policy Routing (PR) is a new area of development that attempts to incorporate policy related constraints on inter-Administrative Domain (AD) communication into the route computation and forwarding of inter-AD packets.Proposals for inter-AD routing mechanisms are discussed in the context of a design space defined by three design parameters: location of routing decision (i.e., source or hop-by-hop), algorithm used (i.e., link state or distance vector), and expression of policy in topology or in link status. We conclude that an architecture based upon source routing, a link state algorithm, and policy information in the link state advertisements, is best able to address the long-term policy requirements of inter-AD routing. However, such an architecture raises several new and challenging research issues related to scaling.

Design and control of networks with dynamic nonhierarchical routing

Abstract: Dynamic routing concepts are described, and the design and control of dynamic routing networks is discussed. The vastly improved performance of the networks is illustrated with examples from operational experience. The evolution of dynamic routing with respect to several future directions is highlighted. These directions are extension to new networks and services, robust design and real-time adaptivity, and extension to interconnecting networks, including the Worldwide Intelligent Network. >

Composite routing protocols

Abstract: A novel routing protocol for computer networks is designed by combining two routing protocols that adapt to changes in the network topology. The first routing protocol is expensive to operate but yields shortest-paths in the networks; therefore, it is best used when changes in the network topology are rare. The second routing protocol is not as costly to operate, but may yield longer paths; hence, it is best used when changes in the network topology are frequent. By combining these two protocols, the resulting protocol has a wide domain of applicability that does not depend on the frequency of changes in the network topology. >

An extended least-hop distributed routing algorithm

Abstract: A routing strategy called NELHNET has been developed for networks with multiprecedence traffic and operating under dynamic traffic and topological conditions. An adaptive distributed algorithm that uses least-hop and least-hop-plus-1 routes in a table of routing vectors, as opposed to the usual table of routing scalars, is described. Current delays are passed backward and forward with the packets to allow development of expected delays to each node via all acceptable routes. The route then selected is the acceptable route with the least expected delay. For speedier recovery, a node returning to service receives the current network status from an adjoining node as soon as the link connecting them is operational. The resultant algorithms show far greater than the marginal improvements originally expected over Arpanet simulations. NELHENET strategies also permit the network to function stably under more heavily loaded conditions than do the Arpanet strategies. >

Routing logic means for a communication switching element

Abstract: Routing logic (RL) for a communication switching element (ISE) of a self-routing multi-stage switching network and able to transfer cells or packets of information from any of its inlets (I1/32) to any of its outlets (O1/32). The outlets of the switching element are arranged in routing groups containing one or more of them and of which the identity is derived by the routing logic from an output-port-address (OPA) identifying an output of the switching network and contained in the self-routing-tag (SRT) associated to the cell. This cell is then transferred to one of the outlets belonging to the selected routing group. The routing logic (RL) is also able to control the transfer of a cell through the switching element according to the execution of a predetermined routing function selected amongst a plurality of routing functions (RS, DI, MC, BH, IS). This routing function to be executed is selected by the routing logic according to a routing-control-code (RCC) also contained in the self-routing-tag (SRT) and each value thereof identifies a specific transfer pattern constituted by a predetermined sequence of routing functions to be executed in the switching elements (ISE) through the switching network (SN).

Optimal routing in multihop packet radio networks

Abstract: An optimal routing algorithm for slotted Aloha PRNs (packet radio networks) which minimizes the average packet delay is developed. The network topology and traffic are arbitrary but fixed. The packet radio nodes serve as sources (and sinks) of traffic as well as repeaters which forward packets to other nodes. The optimal routing algorithm captures the important PRN features discussed and avoids routes that result in high levels of interference and delay. Numerical results which show that optimal routing can result in dramatic performance improvements are presented. >

Markov decision algorithms for dynamic routing (telephone networks)

Abstract: The nature of the traffic-routing problem is described, and early studies of state-dependent routing are noted. A state-dependent scheme seeks to route each call so as to minimize the risk of blocking future calls, and thus responds to the current state of the network on the basis of certain assumptions about future traffic demands. State-dependent routing is considered as a Markov decision process. How the relative costs can be determined for the case of direct routing is shown. >

On bit-serial packet routing for the mesh and the torus

Abstract: The bit-serial routing problem wherein each packet consists of a sequence of k flits and is thus called a snake, is considered. On the basis of the properties of the snake during the routing, a formal definition is given for three different packet routing models, namely, the store-and-forward model, the cut-through model, and the wormhole model. The wormhole model, which is most commonly used in practice, is studied. The first algorithms (deterministic and probabilistic) based on the wormhole model for the permutation routing problem on a chain, on a square mesh, and on a square torus are given. A new lower bound is derived for distance-limited permutation routing on a ring of processors, and an algorithm that matches this lower bound if the packets are routed independently is given. >

A multi-layer router utilizing over-cell areas

Abstract: A new methodology is presented for the solution of the multilayer routing problem that has the potential to outperform channel based multi-layer routing algorithms by expanding the routing regions to include areas over the cells. It is assumed that four routing layers are available. Routing is completed in two steps; a selected group of nets is routed in the between-cell areas using existing channel routing algorithms and the first two routing layers. Then the remaining nets are routed over the entire layout area, between-cell and over-cell areas, using a new two-dimensional router and the next two routing layers. The router used for over-cell routing recognizes arbitrarily sized obstacles, for example, due to power and ground routing or sensitive circuits in the underlying cells. The proposed router was tested on a number of macro-cell layout examples. Advantages of dedicating two routing layers for over-cell routing is illustrated. Results show a significant reduction in total layout area, wire length and number of via when compared to results obtained using two-layer or multi-layer channel routing methods.

Performance issues in the design of dynamically controlled circuit-switched networks

Abstract: The evolution of circuit-switched networks is analyzed with respect to the functional innovations that allowed for new traffic controls. The technical alternatives for routing techniques and their integration with congestion and flow control are considered. The performance of dynamically controlled networks is analyzed in the framework of research done at the Centre National d'Etudes des Telecommunications (CNET) together with studies carried out in other research laboratories. Comparisons are made of dynamic routing versus fixed hierarchical routing and time-dependent routing versus adaptive routing. For adaptive routing, centralized and isolated methods are compared. The benefits of advanced control methods are summarized. >

Network Routing Evolution

Abstract: In this paper we briefly examine some directions for evolution of routing in traffic networks of the future. General trends identified are the expected growth in ability to shift bandwidth both logically and physically. We identify four distinct stages in this evolution. The first stage we describe is the hierarchy that was the basis of virtually all networks just a few years ago. The next level of freedom, found in dynamic routing, allows logical shifts in routing to reallocate network bandwidth on, say, an hourly basis, or more rapidly, on a call-by-call basis. The third level we describe is robust routing for integrated networks, and this network implementation allows logical routing to shift network bandwidth rapidly among node pairs and services. Finally we describe integrated traffic/facility routing in which both physical and logical bandwidth are shifted in response to changing customer and network requirements.

Empirical evaluation of randomly-wire multistage networks

Abstract: Experimental data are presented indicating that multistage interconnection networks with randomly positioned wires are likely to be substantially better for message routing applications than traditional multistage networks, such as the butterfly. The data are presented for a variety of routing models, including store-and-forward routing, cut-through routing, and circuit switching, as well as for scenarios in which a potentially large number of switches are faulty. In most cases, the differences are dramatic, particularly when several switches in the network are faulty. The data provide empirical confirmation of recent theoretical work. >