Network management and operation has traditionally, and quite obviously, been a technologyfocused
endeavor. Getting enterprise networks up and running and maintaining performance
over time has required a distinct focus on connection facilitation. The emphasis has been to
make sure various clients, such as personal computers (PCs), laptops and personal digital
assistants (PDAs), can link to the network from various locations, such as a local area network
(LAN) or dial-up connection.
The network’s principal responsibilities, therefore, have been discovering devices, ensuring they
are properly configured and establishing the linkage between those devices and the services
residing on the network. Largely disregarded, however, have been users’ varying access,
application, bandwidth and quality of service (QoS) needs.
With this old model, all network intelligence and decision-making abilities are placed in the core
devices, handling device identification and enforcement of access and security policies. Basic,
simplistic configuration is employed for basic, simplistic connectivity across multiple domains to
ensure the core switches can handle all identification and connection decisions. Conversely,
edge devices are essentially brainless and unable to assist in the authentication and connection
process. They are able only to pass packets to the core routing switch, with no recognition or
decision-making capacity.
As a result, the infrastructure behaves uniformly no matter what user is connecting to the
network, whether it is a guest or a CIO. In fact, the network is unable to distinguish among
different users, and is capable only of recognizing the devices through which these users are
trying to connect.
This traditional model of network management and access facilitation not only hinders
workforce productivity, but also creates several problems and limitations. Chief among them are
challenges associated with network security, management, performance and operation.