Prasant Mohapatra. Wormhole routing techniques for directly connected multicomputer systems. ACM Computing Surveys, 30(3):374--410, 1998.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....switches, 26 computational nodes, and 12 inter switch links. In this example, the problem is to synchronize the 14 member nodes in a process group (the dark circles in Figure 1) at a barrier point, which are selected for running a parallel application. Cut through switching and wormhole routing [34, 35, 36, 37] have been employed in most new generation networks to minimize the communication latency and buffer requirement. As 4 explained earlier, routing in interconnection networks is restricted to prevent deadlock. Most recently, as it is feasible to implement a single chip switch that can accommodate ....

.... with current VLSI technology, virtual cut through routing 3 is considered in a cluster environment, while still achieving the advantages of virtual channels and adaptive routing [38, 39] Without loss of generality, in this paper, we consider cut through switches implementing wormhole routing [34, 35, 36, 37] with input and output buffers of one flit wide each. A k port switch typically provides a k Theta k crossbar connectivity in order to make a message from any input port to be transferred to any output port. As in most system configurations, we assume a simple node switch connection with ....

P. Mohapatra, "Wormhole Routing Techniques for Directly Connected Multicomputer Systems," ACM Computing Surveys, Vol. 30, No. 3, pp. 374-410, Sep. 1998.

.... mechanism with backtracking for fault tolerant routing systems [5] Virtual channels have been employed for adaptive routing [4] A number of adaptive and faulttolerant wormhole routing techniques have been developed for mesh topologies [2, 3, 6, 11 and 12] that have been reviewed recently [13]. We present a new fault tolerant wormhole routing technique that is based on distributed recovery block approach [9] The method can route messages for various network topologies with a node connectivity of three or more. 1.1 Preliminaries and Assumptions A Q n hypercube contains 2 n nodes ....

.... Some of the existing techniques including the backtracking determine the fault free path only and assume that no more faults will occur during the message transfer [5] Some previous fault tolerant wormhole routing techniques only consider a limited number of node or link failures for evaluation [13] while other deals with only mesh topologies [2, 6, 11 and 12] The DRB based routing suits to all regular and irregular topologies with a minimum node connectivity of three. It also caters for nodes or links failures and the failures can be static or dynamic. The prototype wormhole routing system ....

MOHAPATRA, P.: `Wormhole routing techniques for directly connected multicomputer systems', ACM Computing Surveys, 1998, 30 (3), pp. 374-410

....characteristics of the BTM scheme and compares it with conventional schemes. Using simulation, the performance of the BTM scheme is presented in Section 4. Conclusions and future works are discussed in Section 5. 2 Barrier Tree for Meshes (BTM) The proposed BTM scheme assumes a wormhole capable [7, 8, 9] 2 D mesh network, where a router is connected to the local node via a pair of input and output channels [6] Hardware support for barrier synchronization is provided using barrier registers within the routers. Similar concept has been assumed in most hardware supported synchronization schemes [2, ....

P. Mohapatra, "Wormhole Routing Techniques for Directly Connected Multicomputer Systems, " ACM Computing Surveys, Vol. 30, No. 3, pp. 374-410, Sep. 1998.

....following section. Section 3 analyzes the characteristics of the BTM scheme. Using simulation, the performance of the BTM scheme is presented in Section 4. Conclusions and future works are discussed in Section 5. 2 Barrier Tree for Meshes (BTM) The proposed BTM scheme assumes a wormholecapable [7, 8, 9] 2 D mesh network, where a router is connected to the local node via a pair of input and output channels [6] Hardware support for barrier synchronization is provided using barrier registers within the routers. Similar concept has been assumed in most hardware supported synchronization schemes [2, ....

P. Mohapatra, "Wormhole Routing Techniques for Directly Connected Multicomputer Systems," ACM Computing Surveys, Vol. 30, No. 3, pp. 374410, Sep. 1998.

....support multiple paths between a source and destination. Adaptive routing is supported on the nCUBE3 and Cray T3E. Adaptive routing algorithms can be differentiated by the number of shortest paths allowed. Fully adaptive routing algorithms allow all messages to use any shortest path. Mohapatra [13] provides a recent survey of adaptive routing algorithms. When an adaptive routing algorithm is used, many messages can choose from multiple output channels. The output selection function chooses one of these output channels for the message. 1 When multiple output channels are available, one of ....

P. Mohapatra. Wormhole Routing Techniques for Directly Connected Multicomputer Systems. ACM Computing Surveys, 30(3):374--410, September 1998.

No context found.

Prasant Mohapatra. Wormhole routing techniques for directly connected multicomputer systems. ACM Computing Surveys, 30(3):374--410, 1998.

No context found.

P. Mohapatra. Wormhole routing techniques for directly connected multicomputer systems. ACM Computing Surveys, 30(3):374--410, 1998.

No context found.

Mohapatra, P.; Wormhole routing techniques for directly connected multicomputer systems.ACM Computing Surveys, 30(3), Sep. 1998.

No context found.

P.Mohapatra, "Wormhole Routing Techniques for Directly Connected Multicomputer Systems.," ACM Computing Surveys, Vol. 30(3), pp. 375-411,Sep., 1998.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC