Previous methods of making fault tolerant routing algorithm for mesh-connected multicomputers have been based on adding virtual channels to these networks or sacrificed some non-faulty *** this paper,we focus on fault...
详细信息
Previous methods of making fault tolerant routing algorithm for mesh-connected multicomputers have been based on adding virtual channels to these networks or sacrificed some non-faulty *** this paper,we focus on fault-tolerant routing algorithm for 2-D *** paper proposes a novel and simple fault tolerant routing algorithm based on the concept of k-submesh-connected without virtual channels and not sacrificed non-faulty nodes on mesh *** the one hand,the proposed algorithm is local information based in the sense,because each node knows only its neighbors' status and no global information of the networks is required by the algorithm in the course of routing,on the other hand,for a given pair of source node and destination node,when the routing path extends in each k-submesh,this k-submesh can independently finish routing operations and the algorithm does not consider the status of operations in other k-submeshes,so the algorithm is highly *** running time of the routing algorithm is *** results show that the length of the routing path constructed by this algorithm is very close to the optimal length.
The new Path Length Control (PLC) algorithm establishes and maintains multicast trees which maximize the bandwidth to be shared by multiple receivers and which satisfy the maximum path length bounds for each receiver....
详细信息
The new Path Length Control (PLC) algorithm establishes and maintains multicast trees which maximize the bandwidth to be shared by multiple receivers and which satisfy the maximum path length bounds for each receiver. The PLC algorithm can be implemented as a distributed algorithm, can tradeoff end-to-end delay and bandwidth consumption, and can be implemented for polynomial time execution. Analysis and simulation show that (a) the PLC algorithm generates multicast trees which consume less bandwidth than chose generated by the SPT algorithm while guaranteeing the same shortest path length and (b) consume less bandwidth than trees generated by the Greedy algorithm with only a moderate increase in path length. The PLC algorithm is more flexible and has a lower cost than a combined SPT and Greedy algorithm. (C) 2000 Elsevier Science B.V. All rights reserved.
A k-ary n-cube direct interconnection structure is originally introduced and developed in the context of massively parallel computer(MPC) architectures. In recent years, with the development of VLSI(Very Large Scale I...
详细信息
ISBN:
(纸本)0780378407
A k-ary n-cube direct interconnection structure is originally introduced and developed in the context of massively parallel computer(MPC) architectures. In recent years, with the development of VLSI(Very Large Scale Integration) technology and requirements of high performance Internet routers/switches, the k-ary n-cube interconnection network is considered to build Terabit routers. In this paper, analysis model for three-dimensional torus is presented as switch cores of Internet ***/throughput characteristics are measured with four different routing algorithms, buffer length and scale of fabric. Also a new algorithm Ladder is presented. All simulation experiments are done on OPNET. Results show that this architecture can build high performance *** to other routing algorithms, Ladder has better performance.
In the past years, various network architectures for parallel computers have been proposed, for instance, hyper cubes or star graphs. These classes of networks are known as Cayley graphs. In recent years, there have b...
详细信息
ISBN:
(纸本)9783540405238
In the past years, various network architectures for parallel computers have been proposed, for instance, hyper cubes or star graphs. These classes of networks are known as Cayley graphs. In recent years, there have been some proposals of new families of interconnection networks, namely, constant degree networks. In this paper, we propose a new interconnection network named extended star graphs, and we prove the extended star graphs have hypercube's structure. We also propose routing algorithms for node-to-node communication on extended star graphs. Based on the algorithms, we obtain an upper bound 2n - 1 on the diameter for the n-th order extended star graph.
A k-ary n-cube direct interconnection structure is originally introduced and developed in the context of massively parallel computer(MPC) *** recent years,with the development of VLSI(Very Large Scale Integration) tec...
详细信息
A k-ary n-cube direct interconnection structure is originally introduced and developed in the context of massively parallel computer(MPC) *** recent years,with the development of VLSI(Very Large Scale Integration) technology and requirements of high performance Internet routers/switches,the k-ary n-cube interconnection network is considered to build Terabit *** this paper,analysis model for three-dimensional torus is presented as switch cores of Internet routes. Delay/throughput characteristics are measured with four different routing algorithms,buffer length and scale of *** a new algorithm Ladder is *** simulation experiments are done on *** show that this architecture can build high performance routers. Compared to other routing algorithms,Ladder has better performance.
Wu, Lakshmivarahan and Dhall[5] recently described a deterministic, distributed routing scheme for some special classes of metacyclic graphs. However they have no proof of correctness that the scheme is a shortest pat...
详细信息
Wu, Lakshmivarahan and Dhall[5] recently described a deterministic, distributed routing scheme for some special classes of metacyclic graphs. However they have no proof of correctness that the scheme is a shortest path routing algorithm. In the note we give a suboptimal, deterministic routing algorithm.
We study hypercube networks with a very large number of faulty nodes. A simple and natural condition, the local subcube-connectivity, is identified under which hypercube networks with a very large number of faulty nod...
详细信息
We study hypercube networks with a very large number of faulty nodes. A simple and natural condition, the local subcube-connectivity, is identified under which hypercube networks with a very large number of faulty nodes still remain connected. The condition of local subcube-connectivity can be detected and maintained in a distributed manner based on localized management. Efficient routing algorithms on locally subcube-connected hypercube networks are developed. Our algorithms are distributed and local-information-based in the sense that each node in the network knows only its neighbors' status and no global information of the network is required by the algorithms. For a locally subcube-connected hypercube network that may contain up to 37.5 percent faulty nodes, our algorithms run in linear time and, for any two given nonfaulty nodes, find a routing path of length bounded by four times the Hamming distance between the two nodes. Theoretical analysis and experimental results are presented which show that, under a variety of probability distributions of node failures, hypercube networks are locally subcube-connected with a very high probability and our routing algorithms run in linear time and construct routing paths of nearly optimal length.
Multicast communication involves transmitting information from a single source node to multiple destination nodes, and is becoming an important requirement in high-performance networks. In this paper, we study multica...
详细信息
Multicast communication involves transmitting information from a single source node to multiple destination nodes, and is becoming an important requirement in high-performance networks. In this paper, we study multicast communication in a class of optical WDM networks with regular topologies such as linear arrays, rings, meshes, tori and hypercubes. For each type of network, we derive the necessary and sufficient conditions on the minimum number of wavelengths required for a WDM network to be wide-sense nonblocking for multicast communication under some commonly used routing algorithms.
This paper describes a complete simulation model of distributed direct-sequence, multi-hop high-speed packet radio networks, which is applicable in a highly dynamic environment. In its design, the key part lies in des...
详细信息
ISBN:
(纸本)0780375475
This paper describes a complete simulation model of distributed direct-sequence, multi-hop high-speed packet radio networks, which is applicable in a highly dynamic environment. In its design, the key part lies in designing the appropriate protocols and algorithms to provide reliable network transport and datagram service by dynamically determining optimal routes, effectively controlling traffic and fairly competing the channel in the face of challenging link conditions, mobility and varying traffic loads. Give the practical backgrounds of network, this model truly emulates the whole run process of network and generates many numerical results on network performance, which is very helpful to network design, management and optimization as well as person training.
Divisor-Skip Wavelength Division Multiplexing (DS-WDM) ring is an optical interconnection network for workstation clusters or parallel machines which can connect various number of nodes easily using wavelength divisio...
详细信息
Divisor-Skip Wavelength Division Multiplexing (DS-WDM) ring is an optical interconnection network for workstation clusters or parallel machines which can connect various number of nodes easily using wavelength division multiplexing techniques. However, the wavelength-ordered routine algorithm proposed for the DS-WDM ring requires complicated processes in each router. Here, a new routing algorithm called the comparing dimensional number routing algorithm for the DS-WDM ring is proposed and evaluated. Although the diameter and average distance are almost same as traditional wavelength-ordered routing, the cost and latency are much reduced.
暂无评论