J. Kim and C.R. Das, Hypercube communication delay with wormhole routing, IEEE TC 43(7), pp. 806-814, 1994.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....for at most one other packet before acquiring a channel. This assumption is only reasonable for networks in which the routers havealownumber of ports, or if the utilization of the network is low. Kim and Das presented a method specially designed for hypercubes based on combinatorial arguments [18]. Draper and Gosh uses elements of queuing theory to generate a method for k ary n cubes with wraparound channels that exhibits an impressive accuracy [8] Other approaches focus on particular load conditions such as bimodal load [19] or other routing techniques such as virtual cut through [1] ....

....When we compare our method with other approaches we nd that ours is dioeerent from that of Dally [4] in that it takes into consideration that eachpacket may collide with more than one other packet on each link. We distinguish ourselves from the work of Draper and Ghosh [8] Kim and Das [18] and Agarwal [1] in that we are able to handle any topology. The model as it is presented here is unnished in the sense that the extension to networks is not worked out in all detail, and the extension has not been tested through simulations. 10 However, we do believe that the results wehave ....

J. Kim and C. R. Das. Hypercube communication delay with wormhole routing. IEEE Transactions on Computers, 43(7):806814, 1994.

....Deterministic routing has been widely used in practice [12,13,14,15,19] as a result of its simplicity and minimal requirement for virtual channels. Analytical models of deterministic routing in wormhole routed k ary n cubes, e.g. hypercubes and tori, have been widely reported in the literature [1,4,7,9,10]. More recently, a similar model for the 2 dimensional mesh has been proposed by Greenberg and Guan [8] As these models either have not considered virtual channels or have considered but assume one flit buffer for each virtual channels. So they can not show the effect of the number of virtual ....

J. Kim, C.R. Das, Hypercube communication delay with wormhole routing, IEEE Trans. Computers C43 (7) (1994) 806--814.

....The pertinent question is if traditional techniques for performance evaluation of buffered networks carry over to wormhole switched networks. To the knowledge of this author no systematic study of the subject exists. There are some attempts at analytical modeling of wormhole switched networks [32, 39, 87]. They are all characterized by imposing crude assumptions about stochastic independence. Otherwise, studies are based on simulation experiments [6, 13, 34, 35, 50, 88, 102, 103, 105, 106] Unfortunately, the random nature of output data is often neglected in these cases. Usually little or nothing ....

KIM, J., AND DAS, C. Hypercube communication delay with wormhole routing. IEEE Transactions on Computers 43, 7 (1994), 806--814.

.... in characterizing network delay even with the introduction of wormhole routing [11, 12] Wormhole routing does not reduce the time required for intermediate PNs to process small packets (flits) Larger diameter networks must therefore keep a larger number of PNs busy for a single packet transfer [13] which leads to a reduction in throughput for the entire system. Consequently, a small diameter should still be considered an important criterion in characterizing network delay. It is proving to be very difficult for flat interconnection networks to satisfy the above requirements, especially ....

J. Kim and C. R. Das, "Hypercube Communication Delay With Wormhole Routing," IEEE Transactions on Computers, vol. 43, pp. 806 -- 814, July 1994.

.... wormhole routing [5] 6] Wormhole routing does not reduce the time required for intermediate PEs to process small packets (flits) Larger diameter networks must therefore keep a larger number of PEs busy for a single packet transfer, which leads to a reduction in throughput for the entire system [7]. To this end, several topologies have been proposed to fit different styles of computation. Examples include crossbars, multiple buses, multistage interconnection networks, and hypercubes, to name a few. Unfortunately, most of the presently known network topologies do not satisfy all the above ....

# J. Kim and C.R. Das, "Hypercube Communication Delay with Wormhole Routing," IEEE Trans. Computers, vol. 43, no. 7, pp. 806-- 814, July 1994.

....Port i (b) An illustration of notation and various delays (a) A wormhole routing switch Figure 1: An illustration of wormhole routing. 1. 2 Wormhole Routing Analysis Many performance models for wormhole routing in a multi processor environment have been proposed and presented in the literature [7, 8, 9, 10, 11]. However, they all assumed a negligible size of input buffers. This buffer size must increase in a LAN environment to accommodate transit data that cannot be stopped immediately due to the longer link propagation delay than in a multiprocessor interconnection application. As an example, a 640 ....

J. Kim and C. R. Das. "Hypercube Communication Delay with Wormhole Routing". IEEE Trans. on Computers, 43(7), July 1994.

....[20] which has been adopted as the LAN infrastructure for the Supercomputer SuperNet (SSN) a research project being conducted at UCLA, JPL and Aerospace Corp. 14] Many performance studies for wormhole routing in a supercomputer environment have been carried out and presented in the literature [1, 2, 5, 6, 9, 16]. However, many performance studies do not focus on the LAN environment, which has irregular topology and usually consists of low cost non intelligent switches. Moreover, except for the simulation studies in [16] there is no analysis work evaluating the timeout reset mechanism, which is not only ....

J. Kim and C. R. Das, "Hypercube Communication Delay with Wormhole Routing", IEEE Trans. on Computers, vol. 43, no. 7, July 1994.

....and packets snake through the network one flit after another; only a constant number of flits may be stored in an intermediate node at any time. Most analysis of wormhole routed networks, however, has focused on the class of k ary n cubes networks (which includes the two dimensional torus) e.g. [1, 4, 6, 7, 8]. Figure 1 illustrates the mesh and torus networks in a 4 Theta 4 size. The mesh links are bidirectional. Unidirectional links in the torus suffice for complete reachability and lead to a fairer comparison with the mesh than would bidirectional links, by giving the two networks the same bisection ....

Jong Kim and Chita R. Das. Hypercube communication delay with wormhole routing. IEEE Trans. Computers, 43(7):806--814, July 1994.

No context found.

J. Kim and C. R. Das. Hypercube Communication Delay with Wormhole Routing. IEEE Trans. on Comp., 43(7):806--814, July 1994.

No context found.

J. Kim and C.R. Das, Hypercube communication delay with wormhole routing, IEEE TC 43(7), pp. 806-814, 1994.

No context found.

Kim, J. and Das, C. R. (1994) Hypercube communication delay with wormhole routing. IEEE Trans. Computers, 43, 806--814.

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