February 2004
Intermediate to advanced
312 pages
8h 49m
English
The primary idea behind DiffServ is to (1) classify traffic at the boundary of the network and (2) regulate (shape) this traffic at these ingress boundaries. The classification algorithm entails the assignment of the traffic to behavioral aggregates. Behavioral aggregates are a collection of packets with common requirements (namely QoS). In this way, we have a small number of classifications for packets, which helps in determining flow types and, finally, according resources. The identified traffic is assigned a value: a DiffServ code point. In IP, this may be the 6-bit TOS (type of service) field (as in IPv4) or the traffic-class octet (as in IPv6).
Following the classification of the packets at the boundary ...
Read now
Unlock full access