N. F. Maxemchuk. Comparison of de ection and store and forward techniuques in the Manhattan street and shue exchange networks. In Proc. IEEE INFOCOM, pages 800-809, 1989.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....vs. Delay, r = 8M, Buffer size: 100 1000, biased link length Log(B b) 0 Log(B b) 1 Log(B b) 2 Figure 8: Loss rate vs. end to end delay, biased link lengths increase in the drop rate when that small amount is reduced below a certain minimum, are consistent with the observations made in [11]. In that study, it is shown experimentally that a moderate amount of bu er space available to the routers tends to drastically improve the maximum throughput of a de ection network, bringing it quickly to a level comparable to that of a network with in nite bu ers. In confrontation with our ....

N. Maxemchuk. Comparison of de ection and store-and-forward techniques in manhattan-street network and shue-exchange networks. In Proceedings of IEEE INFOCOM'89, pages 800-809, 1989.

....since there is no need for bu ering or queuing. Moreover, hot potato routing algorithms have been observed to work well in practice. Hot potato algorithms have been used in the HEP multiprocessor [24] the Connection Machine [15] the CalTech Mosaic C [23] and in high speed communication networks [20]. Hot potato routing algorithm are also well suited for use in optical switching networks [1, 14] Our algorithm is greedy : whenever possible packets are sent closer to their destination. Greedy algorithms are adaptive: when loads are light, packets follow short paths. Non greedy algorithms ....

....networks [1, 14] Our algorithm is greedy : whenever possible packets are sent closer to their destination. Greedy algorithms are adaptive: when loads are light, packets follow short paths. Non greedy algorithms typically impose detours that delay packets unnecessarily. Simulation results [20] show that greedy hotpotato routing is e ective; packets almost never move away from their destinations. Our algorithm is also local : Nodes are stateless, making routing decisions exclusively on the basis of the packets received at that time step. 1.2 The Mesh Network We consider the n n ....

N. F. Maxemchuk. Comparison of de ection and store and forward techniuques in the Manhattan street and shue exchange networks. In Proc. IEEE INFOCOM, pages 800-809, 1989.

....since there is no need for bu ering or queuing. Moreover, hot potato routing algorithms have been observed to work well in practice. Hot potato algorithms have been used in the HEP multiprocessor [24] the Connection Machine [15] the CalTech Mosaic C [23] and in high speed communication networks [20]. Hot potato routing algorithm are also well suited for use in optical switching networks [1, 14] Our algorithm is ###### : whenever possible packets are sent closer to their destination. Greedy algorithms are adaptive: when loads are light, packets follow short paths. Non greedy algorithms ....

....networks [1, 14] Our algorithm is ###### : whenever possible packets are sent closer to their destination. Greedy algorithms are adaptive: when loads are light, packets follow short paths. Non greedy algorithms typically impose detours that delay packets unnecessarily. Simulation results [20] show that greedy hotpotato routing is e ective; packets almost never moveaway from their destinations. Our algorithm is also ##### : Nodes are stateless, making routing decisions exclusively on the basis of the packets received at that time step. 1.2 The Mesh Network We consider the # # # ....

N. F. Maxemchuk. Comparison of de ection and store and forward techniuques in the Manhattan street and shue exchange networks. In ##### #### #######, pages 800-809, 1989.

....can be extended to an N 1 node network without major changes to the existing structure. It may be noted that although the authors 3 in [13] refer to the network discussed in their paper as the Multi Mesh network, in the research community this network is known as the Manhattan Street Network [7]. The Multi Mesh network in our paper refers to the network proposed in [2, 3, 4, 5] The architecture proposed in this paper can be used both as a physical as well as a logical topology for optical networks. In terms of simplicity of interconnection and routing, the architecture is comparable to ....

M. F. Maxemchuk, \Comparison of de ection and store-and-forward techniques in Manhattan Street and Shue-Exchange networks", Proc. IEEE INFOCOM'89, pp. 800-809, 1989.

....immediately be forwarded to another node. Hot potato routing algorithms have been observed to work well in practice and have been used in parallel machines such as the HEP multiprocessor [17] the Connection machine [10] and the Caltech Mosaic C [16] as well as high speed communication networks [14]. Hot potato routing algorithms are well suited for optical networks [1; 9; 14; 19; 20] because it is dicult to bu er optical messages. The network we consider here is the two dimensional n n mesh, one of the simplest networks for multiprocessor architectures. The nodes in the network are ....

N. F. Maxemchuk. Comparison of de ection and store and forward techniuques in the Manhattan street and shue exchange networks. In Proc. IEEE INFOCOM, pages 800-809, 1989.

.... algorithms are interesting because they have been observed to work well in practice: hot potato routing algorithms have been used in parallel machines such as the HEP multiprocessor [20] the Connection machine [12] and the Caltech Mosaic C [19] as well as high speed communication networks [16]. Hot potato routing algorithms are well suited for optical networks [1, 10, 16, 22, 23] because it is dicult to bu er optical messages. A hot potato routing algorithm is greedy [3, 6] if each node forwards each packet closer to its destination whenever possible (that is, whenever the desired ....

.... well in practice: hot potato routing algorithms have been used in parallel machines such as the HEP multiprocessor [20] the Connection machine [12] and the Caltech Mosaic C [19] as well as high speed communication networks [16] Hot potato routing algorithms are well suited for optical networks [1, 10, 16, 22, 23] because it is dicult to bu er optical messages. A hot potato routing algorithm is greedy [3, 6] if each node forwards each packet closer to its destination whenever possible (that is, whenever the desired links are not already assigned to other advancing packets) A packet that does not advance ....

N. F. Maxemchuk. Comparison of de ection and store and forward techniuques in the Manhattan street and shue exchange networks. In Proc. IEEE INFOCOM, pages 800-809, 1989.

....immediately be forwarded to another node. Hot potato routing algorithms have been observed to work well in practice and have been used in parallel machines such as the HEP multiprocessor [17] the Connection machine [10] and the Caltech Mosaic C [16] as well as high speed communication networks [14]. Hot potato routing algorithms are well suited for optical networks [1; 9; 14; 19; 20] because it is dicult to bu er optical messages. The network we consider here is the two dimensional n n mesh, one of the simplest networks for multiprocessor architectures. The nodes in the network are ....

N. F. Maxemchuk. Comparison of de ection and store and forward techniuques in the Manhattan street and shue exchange networks. In Proc. IEEE INFOCOM, pages 800-809, 1989.

No context found.

N. F. Maxemchuk. Comparison of de ection and store and forward techniuques in the Manhattan street and shue exchange networks. In Proc. IEEE INFOCOM, pages 800-809, 1989.

No context found.

Maxemchuk, N. F. (1989) Comparison of de ection and store-and-forward techniques in the Manhattan street and shue-exchange networks. In Proc. IEEE INFOCOM '89, Ottawa, Ontario, Canada, April 23-27, pp. 800-809. IEEE Comput. Soc. Press, Los Alamitos, CA, USA.

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