scispace - formally typeset
Search or ask a question

Showing papers on "Static routing published in 1984"


Journal ArticleDOI
TL;DR: The algorithms are based on Gallager's method and provide methods for iteratively updating the routing table entries of each node in a manner that guarantees convergence to a minimum delay routing and utilize second derivatives of the objective function.
Abstract: We propose a class of algorithms for finding an optimal quasi-static routing in a communication network. The algorithms are based on Gallager's method [1] and provide methods for iteratively updating the routing table entries of each node in a manner that guarantees convergence to a minimum delay routing. Their main feature is that they utilize second derivatives of the objective function and may be viewed as approximations to a constrained version of Newton's method. The use of second derivatives results in improved speed of convergence and automatic stepsize scaling with respect to level of traffic input. These advantages are of crucial importance for the practical implementation of the algorithm using distributed computation in an environment where input traffic statistics gradually change.

162 citations


Journal ArticleDOI
01 Jun 1984
TL;DR: This work intends to use the datagram routing algorithm as an option of the DoD Internet Protocol with only minor changes to IP while preserving interoperability with IP modules not supporting multidestination.
Abstract: We present a solution to the problem of multidestination routing in internetworks. The component subnets of these internets share a common datagram internet layer, and the gateways and hosts can determine the next gateway en route to a foreign net. Our datagram routing offers high resilience to network failures, major reductions in network traffic, and no changes whatsoever to the subnetwork routing. The routing follows “shortest” paths as defined by the distance criteria of an internet. We intend to use the algorithm as an option of the DoD Internet Protocol with only minor changes to IP while preserving interoperability with IP modules not supporting multidestination.

126 citations


Journal ArticleDOI
TL;DR: A heuristic algorithm, based upon the daily vehicle routing algorithm of Fisher and Jaikumar, is presented and computational results are given for test problems drawn from the literature.
Abstract: In this paper we consider the period vehicle routing problem, which is the problem of designing routes for delivery vehicles to meet customer service level requirements (not all customers require delivery on every day in the period). A heuristic algorithm, based upon the daily vehicle routing algorithm of Fisher and Jaikumar, is presented and computational results are given for test problems drawn from the literature.

121 citations


Journal ArticleDOI
TL;DR: Two special types of three-layer channel routing, VHV and HVH, are introduced in this paper, and the merging algorithm and the left edge algorithm used in two-layer routing can be extended to three layers.
Abstract: With the advent of VLSI technology, multiple-layer routing becomes feasible. Two special types of three-layer channel routing, VHV and HVH, are introduced in this paper. The merging algorithm and the left edge algorithm used in two-layer routing can be extended to three layers. Attempts are made to compare the lower bounds of channel width of three types of routing--two-layer, VHV, and HVH. The algorithms were coded in PASCAL and implemented on VAX 11/780 computer. The computational results are satisfactory, since all the results lead to a further reduction in routing area.

105 citations


Patent
11 May 1984
TL;DR: In this article, the authors propose a packet switching architecture in which switching network nodes (200-206) automatically determine alternate routes through a switching network (107) so as to increase reliability and distribute traffic.
Abstract: A packet switching architecture in which switching network nodes (200-206) automatically determine alternate routes through a switching network (107) so as to increase reliability and distribute traffic. The switching network comprises stages of distribution (D) and routing nodes (R). The routing nodes are responsive to physical addresses associated with the packets to communicate those packets to address designated downstream nodes. The distribution switching nodes statistically communicate packets to downstream switching nodes on the basis of an alternate routing algorithm and availability of downstream nodes. The initial network stages are alternate distribution and routing stages followed by only routing stages for the remaining stages of the network. Both the routing and the distributing nodes are identical in design and are responsive to an input signal from the network to assume either routing or distributing functions.

83 citations


Patent
Alan Huang1, Scott C Knauer1
08 Mar 1984
TL;DR: In this article, a self-routing switching network consisting of a conventional self routing sorting network (20) followed by a self routing expander (40) is described. But the authors do not specify the destination address of the input signal.
Abstract: A full access, non-blocking, wide band switching network that is self routing. Responsive to input signals having embedded destination addresses, the self routing switching network comprises a conventional self routing sorting network (20) followed by a self routing expander (40). Incoming signals are ordered, and simultaneously appear at the output of the sorting network (20) in ascending order of destination addresses. The signals incoming to the expander network (40) are also processed simultaneously. The processing consists of relating the intermediate address of the expander network input line (1300) at which each signal appears with the destination address of the signal, and steering the signal to the appropriate output line (1400).

74 citations



Proceedings ArticleDOI
25 Jun 1984
TL;DR: It is shown that the Lee-Moore grid-based approach is actually a special case of the general search algorithm presented, which accomplishes the global routing for a building block or general cell routing problem.
Abstract: An algorithm is presented which accomplishes the global routing for a building block or general cell routing problem. A line search technique is employed and therefore no grid is assumed either for the module placements or the pin locations. Instead of breaking the routing surface up into channels, a maze search finds acceptable global routes while avoiding the blocks. Both multi-pin terminals and multi-terminal nets are accomodated. It is shown that the Lee-Moore grid-based approach is actually a special case of the general search algorithm presented. This algorithm is borrowed from the field of artificial intelligence where it has been applied to many state-space search problems.

54 citations


Journal ArticleDOI
TL;DR: A new approach to the global routing of gate arrays that can handle any channel capacities and pin distributions on the chip is proposed, to avoid congestion in the center of the chip.
Abstract: We propose a new approach to the global routing of gate arrays. The method can handle any channel capacities and pin distributions on the chip. The global router first finds unique routes, then pushes connections to the periphery. As outer wiring capacity is consumed, the routing continues inward, connecting pins and making global cell assignments for nets by a centrifugal layering process. The goal is to avoid congestion in the center of the chip, a common problem with conventional methods.

47 citations


Proceedings ArticleDOI
Takeshi Yoshimura1
25 Jun 1984
TL;DR: In this paper, an improved version of the classical "left edge algorithm" is presented, which uses a row-by-row approach, calculating an optimum net assignment to each row.
Abstract: In the LSI chip layout design, channel routing is one of the key problems. The problem is to route a spcified net list between two rows of terminals across a two layer channel. This paper presents a new routing algorithm, which is an improved version of the classical "left edge algorithm". The new algorithm uses a row by row approach, calculating an optimum net assignment to each row. The algorithm was implemented for examples in previously published papers. Experimental results show that the new algorithm produces optimum solutions in most cases.

45 citations


Journal ArticleDOI
TL;DR: To use network resources efficiently, selected routes must involve minimum cost, according to some specific cost criteria, such as the number and type of communications channels on the route, the utilization levels of the channels, and the average delay incurred in transporting a data unit along the route.
Abstract: A principal function of distributed computer networks is message routing in which a data unit is moved along a network path from where it is first accepted (source node) to where it is to be delivered (destination node), through possibly one or more network nodes. In the OS1 protocol architecture, routing is the principal function of Level 3, the Network Layer. 1 In most operational networks, the network automatically generates the routes used to forward data units. To use network resources efficiently, selected routes must involve minimum cost, according to some specific cost criteria, such as the number and type of communications channels on the route, the utilization levels of the channels, and the average delay incurred in transporting a data unit along the route. The cost of a route is the sum of the costs of the constituent links. Once we know the cost of the network links, we can calculate the minimum-cost path for any source-destination pair using algorithms developed in operations research. Major issues related to routing include

Proceedings ArticleDOI
24 Apr 1984
TL;DR: Heuristic strategies to solve multirelational queries in local area networks using broadcast routing are presented to minimize a linear combination of query response and total time.
Abstract: Heuristic strategies to solve multirelational queries in local area networks using broadcast routing are presented. In broadcast routing (used in broadcast and ring networks), data transmitted by one site are received by all the other network sites. This fact is exploited to derive new strategies to minimize a linear combination of query response and total time.

Journal ArticleDOI
TL;DR: The usefulness of nonhierarchical alternate routing in reducing the blocking probability in already existing networks, and in decreasing the number of trunks required to maintain a prespecified grade of service, both in real and artificially generated networks is given.
Abstract: Nonhierarchical alternate routing is the simplest extension to the hierarchical routing currently used in telephone networks. In order to evaluate the merits of such a technique, it is necessary to be able to optimize some performance measure over the routing variables. In a first step, a short discussion is presented of the evaluation algorithms used for computing the end-to-end blocking probabilities in networks operating under nonhierarchicai alternate routings. Then the results of a statistical sampling of the routing space for a small network are presented, to illustrate important characteristics of the routing problem. Next, two heuristic optimization techniques are discussed, one based on classical steepest descent and the other inspired from classical routing optimization of traffic flows in transportation and communication networks. Acceleration techniques are discussed in both cases, and the convergences rates of the alorgithms are presented for a number of networks. A brief discussion is finally given of the usefulness of nonhierarchical alternate routing in reducing the blocking probability in already existing networks, and in decreasing the number of trunks required to maintain a prespecified grade of service, both in real and artificially generated networks.

Proceedings ArticleDOI
25 Jun 1984
TL;DR: A topological data bus router has been implemented in the Hughes Automated Layout system and is the subject of this paper.
Abstract: In LSI and VLSI layout, certain categories of signal nets, e.g. critical nets, power buses and data buses, require special attention during routing. In data bus routing the goal is to maximize the commonality of the routing paths comprising the bus, without unnecessarily lengthening individual routing paths. A topological data bus router has been implemented in the Hughes Automated Layout system and is the subject of this paper.

Journal ArticleDOI
TL;DR: The main emphasis in the design of the COSIE Communication Subsystem was on presenting simple mechanisms that allow more general policies and protocols to be explored.
Abstract: Contemporary distributed office systems rely heavily on communication between workstations, file servers, printers, and other computerized components. The COSIE Communication Subsystem has been developed for an office system internetwork consisting of local networks of varying technologies. The communication facilities provide for the transmission of self-contained messages to mailboxes, which are referenced by capabilities and may be shared. A generalized method for pairing requests with responses supports a variety of communication paradigms, while a flexible routing algorithm permits diverse network topologies. The main emphasis in the design of the COSIE Communication Subsystem was on presenting simple mechanisms that allow more general policies and protocols to be explored.

Proceedings ArticleDOI
01 Dec 1984
TL;DR: This work proves convergence of a distributed gradient projection method for optimal routing in a data communication network and takes into account the possibility of transients caused by updates in the routing strategy being used.
Abstract: We prove convergence of a distributed gradient projection method for optimal routing in a data communication network. The analysis is carried out without any synchronization assumptions and takes into account the possibility of transients caused by updates in the routing strategy being used.

Journal ArticleDOI
TL;DR: This paper first models the switching machine which affects the network overload performance and clarifies the traffic characteristics of three-node triangle network model when out-of-chain routing is permitted unconditionally.
Abstract: When part of a telephone network is congested because a different traffic pattern from that designed is offered, attempts are made to improve the network throughput by utilizing other vacant facilities. One such control method is out-of-chain routing, which permits calls which overflow from the final in-chain route under the existing routing scheme (far to near rotation) to be offered to out-of-chain routes. Out-of-chain routing is expected to improve the network throughput but, on the contrary, it could degrade the network throughput by increasing the traffic toward the alternate route. Therefore a control algorithm for out-of-chain routing becomes necessary. This paper first models the switching machine which affects the network overload performance and clarifies the traffic characteristics of three-node triangle network model when out-of-chain routing is permitted unconditionally. As a result, it is clarified that the algorithm which stops routing to the out-of-chain routes with office congestion is important. Also, the algorithm which permits out-of-chain routing when there is no trunk congestion on out-of-chain routes may decrease the throughput compared with the case of no out-of-chain routing. the results presented here are confirmed by computer simulation of a laeger network.



01 Jan 1984
TL;DR: In this article, it was shown that local routing problems can be solved in O(n(log n)2) time, where n is the number of nodes in the planer grid.
Abstract: A local routing problem is given by a routing region (a subgraph of the planer grid) and a set of nets. For each net a global routing is also given. The problem is to find a local routing which is consistent with the global routing (if there is one). In this paper we show that local routing problems can be sloved in time O(n(log n)2).

Proceedings ArticleDOI
01 Dec 1984
TL;DR: In the decentralized case infrequent updates of relative costs are allowed to give aggregate information about congestion downstream, and optimal controls are re-calculated with the horizon pushed-back at each update time.
Abstract: The computational problems associated with routing of traffic in congested networks is analyzed. An optimal and a suboptimal, but partially decentralized approach is presented. In the decentralized case infrequent updates of relative costs are allowed to give aggregate information about congestion downstream, and optimal controls are re-calculated with the horizon pushed-back at each update time.