I'm running a simulation in Riverbed modeler, and I've got two broadcast domains, each having one Ethernet advanced hub and 16 Ethernet advanced stations connected to it through 10Base-T links. Subsequently, another 10Base-T link has been adopted to connect together the hubs and switch, whereby the switch (ethernet advanced switch) would be in the middle of both hubs.
After executing the simulation the results are like this;
- Traffic forwarded by switch : 4 Mbps
- Traffic received by switch : 8 Mbps
- Point to point through switch to hub1 : 4 Mbps
- Point to point through switch to hub2 : 2 Mbps
I've studied all the possible factors, such as collision count in collision domains, queuing delay for point to points, and the questions are:
- Why does the switch behave in this manner?
- Why does the traffic being forwarded to hub1 vary from the traffic being forwarded to hub2?
- Why is the traffic on the point-to-point throughput to hub2 2 Mbps, whereas the traffic on the point-to-point throughput to hub1 is 4 Mbps?
I've read all about the switch technologies, but I haven't found any relative topic that describes how a switch behaves in such cases. Any source or suggestion would be highly appreciated.
This scenario is in an office (100m x 100m).