Agent-based Distance Vector Routing: A Resource Efficient and Scalable approach to Routing in Large Communication Networks Page: 16
The following text was automatically extracted from the image on this page using optical character recognition software:
60ADVR ith 20 Agents
35 ADVR with 30 Agents
20 DVR 5 thrO gzt L
(a) Path-Cost Convergentsce (b) Route Discovery
Fig. 9. Comparison of Convergence and Route Discovery
resource overhead, it is imperative to consider the structure of the agents. If the
agent code segment is excessive, the agent will consume significant amounts
of resources in terms of bandwidth, memory, and computing. Conversely, if
the code segment is severely restricted, it may be impossible to supply some
of the agents with sufficient intelligence to optimize their task performance.
In order to reduce the size of the code segment, it is possible to supply the
agents code as pre-loadable software modules at each node. The behavior
of these modules is controlled by a set of parameters that are provided by
the agent upon arrival at that node. These parameters will replace the code
segment that is otherwise carried by the agents, resulting in smaller light-
weight agents that may consume less bandwidth. Nevertheless, this approach
does not eliminate the problem, it does only shift the resource overhead from
the link (i.e., bandwidth) to the node (i.e., computation).
4.2.2 Analysis of Path-Cost Convergence and Route Discovery
It is the characteristics of DVR that every change in the routing table of an
individual node is broadcasted to its immediate neighbors. Additionally, the
entire routing table of every node is broadcasted periodically to each of its
neighbors. These events occur asynchronously making use of message concur-
rency, which in turn causes DVR to be highly reactive to small changes. Hence,
any change in a single routing table has a cascading effect initiating a sequence
of broadcasts throughout the network. Such an aggressive parallelism in DVR
results in bursts of update messages within the network. Conversely, ADVR
implements controlled parallelism characterized by the number of agents in
the network. Although ADVR can replicate the behavior of DVR, routing in-
formation, encapsulated in the agent payload, is generally propagated to only
one neighbor. Such an approach restricts the outburst of routing packets due
to small changes. Nevertheless, controlled parallelism reduces the sensitivity
of the algorithm, thereby exhibiting a relatively slow convergence. Figure 9(a)
Here’s what’s next.
This article can be searched. Note: Results may vary based on the legibility of text within the document.
Tools / Downloads
Get a copy of this page or view the extracted text.
Citing and Sharing
Basic information for referencing this web page. We also provide extended guidance on usage rights, references, copying or embedding.
Reference the current page of this Article.
Amin, Kaizar A. & Mikler, Armin R. Agent-based Distance Vector Routing: A Resource Efficient and Scalable approach to Routing in Large Communication Networks, article, March 25, 2002; [New York, New York]. (digital.library.unt.edu/ark:/67531/metadc111275/m1/16/: accessed October 23, 2018), University of North Texas Libraries, Digital Library, digital.library.unt.edu; crediting UNT College of Engineering.