scispace - formally typeset
Search or ask a question

Showing papers on "Hazy Sighted Link State Routing Protocol published in 1996"


Journal ArticleDOI
TL;DR: This paper gives theoretical analyses of simple wormhole routing algorithms, showing them to be nearly optimal for butterfly and mesh connected networks, and reports simulation results suggesting that the idea of random initial delays may have an impact beyond theoretical analysis.
Abstract: Virtually all theoretical work on message routing in parallel computers has dwelt on packet routing: messages are conveyed as packets, an entire packet can reside at a node of the network, and a packet is sent from the queue of one node to the queue of another node until its reaches its destination. A trend in multicomputer architecture, however, is to use wormhole routing. In wormhole routing a message is transmitted as a contiguous stream of bits, physically occupying a sequence of nodes/edges in the network. Thus, a message resembles a worm burrowing through the network. In this paper we give theoretical analyses of simple wormhole routing algorithms, showing them to be nearly optimal for butterfly and mesh connected networks. Our analysis requires initial random delays in injecting messages to the network. We report simulation results suggesting that the idea of random initial delays may have an impact beyond theoretical analysis.

27 citations


Proceedings ArticleDOI
24 Mar 1996
TL;DR: It is shown how distributed route selection algorithms embedded within network nodes may be managed by a TMN system, by virtue of a management service called load balancing, to increase their adaptivity to network wide conditions and therefore their effectiveness.
Abstract: This paper proposes a hierarchical approach to the routing issue in ATM-based broadband multi-service networks. We show how distributed route selection algorithms embedded within network nodes may be managed by a TMN system, by virtue of a management service called load balancing, to increase their adaptivity to network wide conditions and therefore their effectiveness. The paper elaborates on the issues involved, presents specific algorithms and a TMN-compliant management architecture. The proposed management service influences the routing decisions by conveying network-wide information and contributes to the network load balancing by directing the routing algorithms to route traffic to the least congested network areas. Results regarding the effect of the proposed management system on the network operation are also presented.

26 citations


Journal ArticleDOI
TL;DR: This paper considers two variations of the mesh model: 1) a mesh with fixed buses, and 2) aMesh with reconfigurable buses which have been the subject matter of extensive previous research and solves numerous important problems related to packet routing and sorting on these models.
Abstract: Mesh connected computers have become attractive models of computing because of their varied special features. In this paper we consider two variations of the mesh model: 1) a mesh with fixed buses, and 2) a mesh with reconfigurable buses. Both these models have been the subject matter of extensive previous research. We solve numerous important problems related to packet routing and sorting on these models. In particular, we provide lower bounds and very nearly matching upper bounds for the following problems on both these models: 1) Routing on a linear array; and 2) k-k routing and k-k sorting on a 2D mesh for any k/spl ges/12. We provide an improved algorithm for 1-1 routing and a matching sorting algorithm. In addition we present greedy algorithms for 1-1 routing, k-k routing, and k-k sorting that are better on average and supply matching lower bounds. We also show that sorting can be performed in logarithmic time on a mesh with fixed buses. Most of our algorithms have considerably better time bounds than known algorithms for the same problems.

16 citations


Proceedings ArticleDOI
15 Apr 1996
TL;DR: This paper introduces a graph-partitioning generic methodology for developing deadlock-free wormhole routing in an arbitrary network and shows that the methodology can be applied to store-and-forward and virtual cut-through routings as well.
Abstract: This paper introduces a graph-partitioning generic methodology for developing deadlock-free wormhole routing in an arbitrary network. Further extension allows partial cyclic dependencies among virtual channels. A novel fully adaptive nonminimal deadlock-free routing algorithm has been developed for k-ary n-cube torus network. Since our technique is based on decomposing a network into several subdigraphs, it simplifies and generalizes the development of both static and adaptive deadlock-free routing algorithms for arbitrary network topologies. We show that our methodology can be applied to store-and-forward and virtual cut-through routings as well.

14 citations


Proceedings ArticleDOI
21 Oct 1996
TL;DR: A new method is described for routing multimedia traffic in a frequency-hop (FH) store-and-forward packet radio network based on least-resistance routing with different link and path resistance metrics for different message types.
Abstract: A new method is described for routing multimedia traffic in a frequency-hop (FH) store-and-forward packet radio network. The method is illustrated for traffic of two types, each type having its own throughput, delay, and error-rate requirements. A typical application is the routing of voice and data packets in a multi-hop network. In such an application, the voice packets cannot tolerate much delay, but they are allowed to contain a small number of errors, while the data packets must be delivered error-free even if a moderate delay is required to do so. The routing protocol presented in the paper takes into account the type of service required for each type of traffic. In addition, the routing protocol adapts to the interference as seen by the FH radios in the network. The approach is based on least-resistance routing with different link and path resistance metrics for different message types. The link resistance metrics reflect the ability of a radio to receive a packet on the link and the ability of the link to accommodate the service required by the type of packet that is being transmitted. The route chosen for a particular type of packet depends on the ability of the radios along the route to receive and forward the packet within the constraints imposed by its service requirements.

9 citations


Journal ArticleDOI
30 Oct 1996
TL;DR: The results show that the heuristic class of search techniques are potentially suitable for adaptation as fault-tolerant routing strategies on multiprocessor networks.
Abstract: Most fault-tolerant routing strategies are topology dependent or constrained to tolerate specific classes of faults. Artificial Intelligence search techniques, on the other hand, are potentially adaptive and can be applied to a generic problem space. We investigate the performance of such search techniques modified for use as fault-tolerant routing strategies on a problem space simulating a multiprocessor network of generic topology. This virtual network comprises processors individually equipped with a communications router designed to support message routing. The performances of these modified search techniques are compared for up to 20% of randomly generated link faults (with a processor failure being simulated by total link failure). The results show that the heuristic class of search techniques are potentially suitable for adaptation as fault-tolerant routing strategies on multiprocessor networks.

7 citations


Journal ArticleDOI
TL;DR: This work presents a new switching format for multiprocessor networks, which is a hybrid of packet and circuit switching, and combines advantages of both, and presents an implementation of this protocol for a hypercube computer and a particular routing algorithm.
Abstract: We propose a new switching format for multiprocessor networks, which we call conflict sense routing protocol. This switching format is a hybrid of packet and circuit switching, and combines advantages of both. We initially present the protocol in a way applicable to a general topology. We then present an implementation of this protocol for a hypercube computer and a particular routing algorithm. We also analyze the steady-state throughput of the hypercube implementation for random node-to-node communications.

7 citations


Proceedings ArticleDOI
30 Apr 1996
TL;DR: A simulation of a frequency hopping packet radio network using the SINCGARS radio is utilized to examine the performance of LRR in a realistic tactical environment and LRR algorithms using both node and link resistance measures are investigated as well as a number of different routing metrics.
Abstract: Given the mobility requirements, the need for multihop (i.e., store-and-forward) operation, and the variability of the link conditions, distributed network protocols are necessary in order to make a reliable network out of the collection of unreliable links. Mobile, multiple-hop packet radio networks must route packets through several radios in order to deliver packets to their destinations. Least resistance routing (LRR) is specifically designed for routing packets in a tactical packet radio network. A unique feature of LRR is that it incorporates side information into the routing protocol for the purpose of detecting the presence of interference. The routing tables are adjusted to avoid particular radios and regions of the network that are subjected to high levels of partial-band or multiple-access interference. A simulation of a frequency hopping packet radio network using the SINCGARS radio is utilized to examine the performance of LRR in a realistic tactical environment. The model of the SINCGARS radio allows for greater detail in accounting for mobile partial-band jamming and FH multiple-access interference than has been possible in previous investigations. Both link and transport layer acknowledgments along with retransmission of dropped packets are modeled. The performance of LRR is compared to a conventional shortest path routing algorithm. LRR algorithms using both node and link resistance measures are investigated as well as a number of different routing metrics. Networks with both static and mobile user topologies are simulated.

6 citations


Proceedings ArticleDOI
21 Oct 1996
TL;DR: In this article, the authors compare two routing algorithms and examine their tradeoffs in time efficiency and memory usage in a mixed ground/satellite tactical communication network simulation, and analyze their performance under a wide range of varying parameters (e.g., bit error rate, packet size, etc).
Abstract: We describe and analyze several results from a mixed ground/satellite tactical communication network simulation On its own, the simulation is important as it provides a testbed for a wide range of simulated experiments Our main results to date are in the areas of routing algorithms and packet retransmission protocols We compare two routing algorithms and examine their tradeoffs in time efficiency and memory usage A flood-search routing (FSR) algorithm is first examined, and then improved upon in an augmented flood-search routing algorithm that we call FSR+ FSR+ locally caches address information, providing greater message throughput performance than FSR For two packet retransmission protocols ("go-back-N" and "selective repeat"), we analyze their performance under a wide range of varying parameters (eg, bit error rate, packet size, etc) Results include the effect of buffer size on throughput Large buffer sizes hurt "go-back-N"'s performance but not "selective repeat"'s performance as bit error rate increases

4 citations


Journal ArticleDOI
TL;DR: In this paper, the authors present a group routing protocol for a network of processes, which is based on the local unicast routing tables of each process in the network and relies on a tree of processes.

4 citations


Journal Article
TL;DR: In this article, the authors consider 1-1 routing of m packets in a d-dimensional mesh with n/sup d/ processors and d/spl middot/n/supd-1/ buses (one per each row, column).
Abstract: Recently, the architecture of 'mesh of buses' is becoming quite popular in parallel computing. Its main advantage is the limited broadcast capability that is used to overcome the main disadvantage of the mesh, namely the relatively big diameter. We show that in such networks busses indeed accelerate the time for the fundamental problem of routing. Furthermore, unlike in the 'store and forward' model, the time becomes proportional to the network load. Namely, small number of packets is faster to route. We consider 1-1 routing of m packets in a d-dimensional mesh with n/sup d/ processors and d/spl middot/n/sup d-1/ buses (one per each row, column). The two standard models of accessing the buses are considered and compared. CREW in which only one processor may transmit at any given time, and the CRCW model in which several processors may attempt to transmit at the same time (getting a noise signal as a result). We design a routing algorithm that routes m packets in the CREW model in O(max(1/d, n 1/4+1)) steps. A matching lower bound is also proved. In the CRCW case we show an algorithm of O(m1/d log n) and a lower bound of /spl Omega/(1/d). It is shown that the difference between the models is essentially due to the improved capability of estimating threshold functions in the CRCW case.

Proceedings ArticleDOI
30 Apr 1996
TL;DR: The use of the robust design method in conjunction with the simulation model to optimize the OSPF protocol for dynamic routing to maintain acceptable performance under anticipated traffic loading and network scenarios is described.
Abstract: Dynamic adaptive routing is required to improve the survivability of a packet radio network A protocol called Open Shortest Path First (OSPF) Version 2 has been selected to meet the requirement The network connectivity status, which often changes rapidly in a highly mobile tactical environment (due to jamming and station outages/additions), must be updated dynamically and automatically by the protocol to maintain reliable network service With fast updates, the topology database will be more up-to-date and routing decisions will be more optimum However high update rate will consume more channel bandwidth and will have adverse impact on user traffic The objective is to optimize the OSPF protocol such that the system will maintain acceptable performance under anticipated traffic loading and network scenarios The timing parameters associated with the update process will be varied to determine the optimum settings The updates are transmitted in OSPF packets called link state updates (LSUs) The LSU packets will compete with user traffic on channel utilization To reduce the impact on user traffic, the OSPF protocol parameters will be optimized to reduce the overhead traffic A detailed simulation model has been developed for this effort We describe the use of the robust design method in conjunction with the simulation model to optimize the protocol for dynamic routing The main advantage of the robust design method is to provide a disciplined approach for experimentation (ie, with no need to run all possible combinations of the experiment) that is needed to determine the most influential control factors

Proceedings ArticleDOI
25 Feb 1996
TL;DR: In this article, a burst-switched all-optical network, TBONE, has been demonstrated using a new signaling protocol, CC, which proactively prevents looping by dynamically selecting a path that avoids a loop.
Abstract: In summary, a burst-switched all-optical network, TBONE, has been demonstrated using a new signaling protocol, CC. A routing scheme is used that pro-actively prevents looping by dynamically selecting a path that avoids a loop. Future work includes the implementation of Tell-and-Go on TBONE, and examination of its performance in comparison to CC.

Book ChapterDOI
26 Aug 1996
TL;DR: In this article, the authors describe a leveled optical network and routing algorithms for it, and show how to simulate this network on high-dimensional meshes of optical busses (MOBs).
Abstract: We describe routing algorithms on networks composed of optical busses. Using networks with short busses and small degree we are able to give very fast routing algorithms. First, we describe a leveled optical network and routing algorithms for it. Next, we show how to simulate this network on high-dimensional meshes of optical busses (MOBs). We present algorithms, e.g., for h-relations with runtime being double-logarithmic in the size of the mesh, linear in h, and polynomial in the dimension. Previous results are exponential in the dimension. We use a novel type of protocol and analysis inspired by hashing based shared memory simulations with redundant storage representation from [MSS95].

Proceedings ArticleDOI
S. Rahman1, Chita R. Das
27 May 1996
TL;DR: Design of a parallel routing algorithm simulator for n-dimensional mesh connected networks with wormhole switching and virtual channel flow control and experimental results show that the parallel simulator can provide significant speedup compared do a uniprocessor environment.
Abstract: Performance of a network depends primarily on the network topology, switching mechanism, flow control protocol and the underlying routing algorithm. While many routing algorithms have been proposed recently for direct networks, there is no time efficient technique to evaluate and compare all of them. A conventional routing algorithm simulation of a network on a uniprocessor takes unacceptably large computing time. The simulation can be made very time efficient by parallelizing it and running on a parallel test bed. This research is focussed on designing a parallel routing algorithm simulator for n-dimensional mesh connected networks with wormhole switching and virtual channel flow control. The research addresses partitioning mapping, synchronization issues, and implementation of various routing algorithms for 2-D and 3-D mesh architectures. Experimental results show that the parallel simulator can provide significant speedup compared do a uniprocessor environment.

Proceedings ArticleDOI
12 May 1996
TL;DR: The application of a mixed-signal VLSI realization of a recurrent spatiotemporal neural network to a computer network routing problem makes it particularly attractive for link-state network routing protocols.
Abstract: This paper discusses the application of a mixed-signal VLSI realization of a recurrent spatiotemporal neural network to a computer network routing. The neural network exhibits a regular interconnect structure and uses simple processing units in a combination which is well suited for VLSI implementation with a standard fabrication process. Its ability to rapidly arrive at solutions to the single-source multiple-destination routing problem makes it particularly attractive for link-state network routing protocols.

Journal ArticleDOI
TL;DR: A simple network layer protocol that integrates routing and connectionless transfer of data in a wireless environment based on a rooted tree topology that has the potential to be used also in other environments.
Abstract: This paper presents a simple network layer protocol that integrates routing and connectionless transfer of data in a wireless environment. The protocol is specifically geared towards supporting transfer of signalling in mobile networks based on a rooted tree topology. Exploiting the special characteristics of such a topology allows the specification of a very simple and processing efficient routing function. Using the routing function, a connectionless message transport service is implemented. The connec- tionless transport service is comparable to that of typical network layer protocols of existing data networks. The protocol has originally been specified to carry signalling messages in the control plane of mobile, cellular systems but has the potential to be used also in other environments.

Proceedings ArticleDOI
27 May 1996
TL;DR: A group routing protocol for a network of processes that has the property of adapting the group tree to changes in the unicast routing tables without compromising the integrity of thegroup tree, even in the presence of unicorn routing loops.
Abstract: We present a group routing protocol for a network of processes. The task of the protocol is to route data messages to each member of a process group. To this end, a tree of processes is constructed in the network, ensuring each group member is included in the tree. To build this tree, the group routing protocol relies upon the unicast routing tables of each process. Thus, group routing is a composition of a unicast routing protocol, whose detailed behavior is unknown but its basic properties are given, and a protocol that builds a group tree based upon the unicast routing tables. The design of the group routing protocol is presented in three steps. First, a basic group routing protocol is presented and proven correct. Then, the protocol is refined twice, strengthening its properties with each refinement. The final protocol has the property of adapting the group tree to changes in the unicast routing tables without compromising the integrity of the group tree, even in the presence of unicast routing loops.

Journal ArticleDOI
TL;DR: This paper studies the hypercrossbar network which has high flexibility and extensibility as a massively parallel computer network, and an implementation method of adaptive routing based on dynamic path determination is proposed, and its performance is evaluated by computer simulation.
Abstract: This paper studies the hypercrossbar network which has high flexibility and extensibility as a massively parallel computer network. Its performance is evaluated in nonroutine processing by computer simulation and theoretical analysis. To improve its performance, an implementation method of adaptive routing based on dynamic path determination is proposed, and its performance is evaluated by computer simulation. It is verified that the hypercrossbar network with adaptive routing has about 45 percent of the peak throughput at random transfer. Furthermore, its effectiveness is extremely high for applications with hot-spot. It can attain about 8 times throughput compared with fixed routing.

Book ChapterDOI
26 Aug 1996
TL;DR: A high-performance router approach to handle adaptive, deadlock-free, wormhole routing algorithms in a number of interconnection network topologies with programmable support for a wide range of networks and routing algorithms.
Abstract: In this paper, we propose a high-performance router approach to handle adaptive, deadlock-free, wormhole routing algorithms in a number of interconnection network topologies. The router allows programmable support for a wide range of networks and routing algorithms. Routing algorithms are mapped to a set of bit-patterns which are matched in parallel. The number of bit patterns required depends on the network topology and adaptive routing algorithm; in general this number is of O(network degree). To show the applicability of this adaptive router, we have studied a number of adaptive routing algorithms.

Journal ArticleDOI
TL;DR: This research note addresses the problem of efficiently updating routing tables in computer networks by presenting an efficient and practical solution for finding nearer-ancestors.