Neighborhood Discovery Function
The neighborhood discovery function (NDF) can be used to provide a means for locating all the other nodes in
the network and to simplify the network configuration process. This tool helps to eliminate the need to
manually configure some of the attributes of interface modules inside the LS2020 switch.
CDF enables the switch to find nodes in the network; the found nodes are placed in a database to make
internal routing decisions. This neighborhood discovery process runs on the network processor (NP) inside the
LS2020’s attached network. The neighborhood discovery function provides two functions:
Discovering the local network topology
•
Determining the network processor processes
•
Discovering the Local Network Topology
This function keeps track of the interface modules added to or removed from service manually or by a
hardware fault. When another module is installed into the switch, the NDF automatically locates and
distributes functions to each network processor. If an interface module is removed from the switch, NDF must
be used to move processes running on that network processor to another or to terminate the processes if they
are no longer necessary.
Determining Network Processor Processes
When a resource is added to or removed from an LS2020, the NDF process notifies the global information
distribution (GID) system. The GID then floods information about the change from the local network
processor module to all the network processors in the network. A function called the connection admission
control (CAC) is used by the network processors to generate new routes through the network. Using this
process, the internal routing module provides a directed set of links from one source node to a destination
node, providing link state and bandwidth information that can be advertised throughout the network.
Virtual Path Connections
Virtual path (VP) is a generic term used to define more than one virtual channel directed to the same ATM
endpoint. A virtual path essentially makes for a much larger pipe by combining the redundant paths and
creating a logical grouping of virtual connections between multiple ATM sites. The advantage of using VPs in
an ATM network is that they enable cell streams from multiple users to be bundled together, resulting in
transit over multiple links that is much faster than transit over one link. This process is referred to as
load−sharing. In an ATM cell (discussed in Chapter 8), the Virtual Path Identifier (VPI) field in the cell
header identifies the VP. If a virtual circuit (VC) cell is traversing the network, the network pays attention to
the Virtual Channel Identifier (VCI) field.
There is another reason to use a VP between LightStream switches. This path can be used if an end user needs
to define the route of certain ATM cells in a manually defined manner that must remain entirely transparent to
the network service provider.
Before you can enable a PVC and an SVC between any two endpoints in the network, an internal routing
database must first be established. The internal database is established during the network configuration
process by downloading the necessary configuration information to each of the LS2020 switches installed in
your network. An internal routing module keeps the database information up to date with the required
information that confirms the state of every link in the network.
This routing database is replicated from the NP used in each of the LS2020 switches in the network. The
database is synchronized with the other ATM switches in the network; the ATM switches then use a routing
186
Summary of Contents for Catalyst 1900 Series
Page 1: ......
