Tech Note #22: Asynchronous Transfer Mode (ATM) Addresses
1998 Bionic Buffalo Corporation; All Rights Reserved.
14 September 1998
9 of 12
VPCs and VCCs, along with their identifier VPCIs and VCCIs, are used also in discussing
configurations which do not include cross-connects. In general, they are used when discussing
complex paths or connections that span multiple links and nodes.
Switched Services and Control Plane Signalling
In addition to facilitating dynamic assignment of VPI and VCI, the control plane signalling
protocols give the user a regional or global view of the network.
If ATM networks were limited to fixed VPI and VCI numbers, then the total address space
would be no greater than 28 bits (or 32 bits for NNI nodes). By dynamically assigning VPI and
VCI numbers, a much greater addressing range is possible.
The dynamic assignment of addresses at each node is based on a very different philosophy from
that of common IP networks (using TCP/IP or UDP/IP). In IP networks, each node knows the
global address (IP number) of the nodes to which it is connected. The total number of
addresses (or nodes) in an IP network is limited to the largest IP number (32 bits), while there is
no practical limit to the total number of nodes in an ATM network. The limitation at an ATM
node is on the number of simultaneous channels and paths, and is not practically reached.
(Note that IPv6 increases the address space to 128 bits, and that there are various address
translation schemes to ameliorate the problems of having a restricted address space in IP
networks. We also emphatically state that this situation, out of any specific context, does not
make ATM’s addressing scheme “better” than IP’s in any general sense. There are
disadvantages to ATM’s approach, as well as advantages.)
A user on an ATM network, who wishes to connect to, or call, another user, sends a setup
message to the network. In this case, the “network” is the exchange or switch, which is reached
using a pre-assigned VPI and VCI in the control plane. In response, the network sends a similar
setup message to the destination node. The response to a setup message is a connect message.
After the destination node connects, then the network connects to the caller.
The setup message contains the global or regional addresses of the calling and called nodes, as
well as other parameters regarding the type of service and protocols to be used for the
connection. The addresses used are not VPI/VPI numbers, and are described in the next
In addition to these messages, there are additional messages to indicate progress and to
acknowledge successful connection. The overall procedure is: