What Is Effective Bandwidth?
Networks are expensive to install and maintain, and companies want to get the most for their money. These two facts make it important to understand how various network topologies use the bandwidth available to them -- and how to maximize the efficiency of the deployed network topology. Networking topologies have improved in recent years and many of the difficulties in early networks have been resolved by newer network switching technology.
-
Effective Bandwidth
-
The effective bandwidth on a network is not necessarily the same as the available bandwidth. The reasons for this reduction in efficiency are different for different network topologies, but the effect is the same: reduced real capacity for your network.
Ethernet
-
On a Gigabit (1 billion bits per second) Ethernet network, only about 300 megabits per second actually get used. This assumes there is no switching involved and that the network is using simple hubs. The reduction in efficiency is caused by network traffic collisions and retransmissions, which are inherent in Ethernet networks. The maximum utilization for Ethernet is in the 35 percent to 37 percent range; the network will begin to bog down after that point.
-
Token Ring
-
Token Ring networks are much more efficient than Ethernet networks, achieving 75 percent to 80 percent utilization before they begin to slow down. This is due to the more efficient network access method of using a “token” to grant access to a network node that wants to transmit. The token moves in a circle around the network, hence the “ring” part of the name. This avoids collisions and reduces retransmissions.
Switched Networks
-
Switched networks avoid collisions by creating virtual connections between network nodes and granting them the full bandwidth available on the network. The absence of collisions in the case of Ethernet, and the reduction of the ring size in the case of Token Ring both drive the effective bandwidth in a switched network to nearly 100 percent capacity.
-
References
Resources
- Photo Credit Hemera Technologies/AbleStock.com/Getty Images