The tag has no usage guidance.

learn more… | top users | synonyms

0
votes
0answers
7 views

How to avoid looping of packets while implementing k-shortest paths algorithm in Network Simulator-3?

I am trying to implement k-shortest paths algorithm in NS-3 for IPv4GlobalRoutingProtocol. I am concerned about how to avoid looping of packets. My implementation calculates k-shortest paths from ...
0
votes
1answer
12 views

In the Data Link Layer, is it better to use large or small frames?

I understand how the layer works to an extent, but cannot seem to figure out if the data link handles big frames more efficiently due to error detection or small frames better. I am assuming from my ...
2
votes
0answers
14 views

Benes (Fat Tree) Network for number of inputs/outputs that is not 2^n

I was reading about Benes Network construction in this book. Their construction is easy for a number of inputs and outputs that is an exponent of two. However it seems to me that for a number of i/o ...
0
votes
0answers
24 views

Measure of network branchiness on a weighted graph

I'm working with road networks, in which each edge is a physical street segment with a length attribute. Nodes represent junctions. However, my question should be generalizable to any weighted graph. ...
2
votes
0answers
29 views

How to know for sure that a router is congested?

Suppose a router X is dropping packets. Another router A will investigate whether X is dropping packets due to congestion. What are the procedures that A should follow? A already knows X's buffer ...
0
votes
3answers
139 views

P2P and C/S, is it possible to determine which one is more bandwidth consuming?

In peer-to-peer (P2P) networks nodes are interconnected with each other. In client/server (C/S) networks clients talks to the server. If in a C/S network clients also want to talk with each other ...
2
votes
1answer
113 views

Seeking “gold standard” to evaluate accuracy of network clustering algorithm

I'm currently looking at network clustering algorithms (we're currently looking at both directed and undirected, unweighted networks). The algorithms we've tried produce visually nice clusters. ...
24
votes
2answers
438 views

Why do neural networks seem to perform better with restrictions placed on their topology?

Fully connected (at least layer to layer with more than 2 hidden layers) backprop networks are universal learners. Unfortunately, they are often slow to learn and tend to over-fit or have awkward ...
10
votes
1answer
203 views

Analysis of and references for Koch-snowflake-like (and other exotic) network topologies

In computer networking and high-performance cluster computer design, network topology refers to the design of the way in which nodes are connected by links to form a communication network. Common ...