|
Footnotes
|
|
There are still some vendors that try to put as many buffers as possible on their routers. A recent example is the buffer bloat problem that plagues some low-end Internet routers [GN2011]_.
|
|
Some examples of the performance of various types of commercial networks nodes (routers and switches) may be found in http://www.cisco.com/web/partners/downloads/765/tools/quickreference/routerperformance.pdf and http://www.cisco.com/web/partners/downloads/765/tools/quickreference/switchperformance.pdf
|
|
Some networking technologies allow to adjust dynamically the bandwidth of links. For example, some devices can reduce their bandwidth to preserve energy. We ignore these technologies in this basic course and assume that all links used inside the network have a fixed bandwidth.
|
|
This name should not be confused with the duration of a transmission slot in slotted ALOHA. In CSMA/CD networks, the slot time is the time during which a collision can occur at the beginning of the transmission of a frame. In slotted ALOHA, the duration of a slot is the transmission time of an entire fixed-size frame.
|
|
In this section, we focus on congestion control mechanisms that regulate the transmission rate of the hosts. Other types of mechanisms have been proposed in the literature. For example, `credit-based` flow-control has been proposed to avoid congestion in ATM networks [KR1995]_. With a credit-based mechanism, hosts can only send packets once they have received credits from the routers and the credits depend on the occupancy of the router's buffers.
|
|
For example, the measurements performed in the Sprint network in 2004 reported more than 10k active TCP connections on a link, see https://research.sprintlabs.com/packstat/packetoverview.php. More recent information about backbone links may be obtained from caida_ 's real-time measurements, see e.g. http://www.caida.org/data/realtime/passive/
|
