An extended least-hop distributed routing algorithm
Don J. Nelson, Khalid Sayood, Hyunseok Chang
Abstract
Don J. Nelson, Khalid Sayood, Hyunseok Chang
Abstract
A routing strategy called NELHNET has been developed for networks with multiprecedence traffic and operating under dynamic traffic and topological conditions. An adaptive distributed algorithm that uses least-hop and least-hop-plus-1 routes in a table of routing vectors, as opposed to the usual table of routing scalars, is described. Current delays are passed backward and forward with the packets to allow development of expected delays to each node via all acceptable routes. The route then selected is the acceptable route with the least expected delay. For speedier recovery, a node returning to service receives the current network status from an adjoining node as soon as the link connecting them is operational. The resultant algorithms show far greater than the marginal improvements originally expected over Arpanet simulations. NELHENET strategies also permit the network to function stably under more heavily loaded conditions than do the Arpanet strategies.>
OpenAlex reports 24 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
A routing strategy called NELHNET has been developed for networks with multiprecedence traffic and operating under dynamic traffic and topological conditions. An adaptive distributed algorithm that uses least-hop and least-hop-plus-1 routes in a table of routing vectors, as opposed to the usual table of routing scalars, is described. Current delays are passed backward and forward with the packets to allow development of expected delays to each node via all acceptable routes. The route then selected is the acceptable route with the least expected delay. For speedier recovery, a node returning to service receives the current network status from an adjoining node as soon as the link connecting them is operational. The resultant algorithms show far greater than the marginal improvements originally expected over Arpanet simulations. NELHENET strategies also permit the network to function stably under more heavily loaded conditions than do the Arpanet strategies.>
Key concepts: Routing table, Hop (telecommunications), Routing algorithm, Computer network, Computer science, Network packet, Routing (electronic design automation), Node (physics)