Enterasys Networks CSX7000 User Guide
USER’S GUIDE
260 CyberSWITCH
F
RAME
R
ELAY
A
CCESS
B
ACKGROUND
I
NFORMATION
Frame Relay is a frame mode service in which data is switched on a per frame basis, as opposed to
a circuit mode service that delivers packets on a call-by-call basis. This feature will allow the system
to efficiently handle high-speed, bursty data over wide area networks. It offers lower costs and
higher performance than a X.25 packet switched network for those applications that transmit data
at a high speed in bursts.
a circuit mode service that delivers packets on a call-by-call basis. This feature will allow the system
to efficiently handle high-speed, bursty data over wide area networks. It offers lower costs and
higher performance than a X.25 packet switched network for those applications that transmit data
at a high speed in bursts.
In private line network implementations, network bandwidth is dedicated to a particular
destination, whether via private lines or circuit switched connections. In any event, these resources
are only available to traffic bound for that location and are reserved for that traffic whether that
traffic is present or not. Conversely, in a frame relay network, bandwidths within the network and
in the access lines are only allocated between any two end devices if there is traffic moving between
those devices. At other times, this bandwidth is made available to other network devices.
Therefore, the performance in a frame relay network is then only limited by the bandwidth
available at the access point to the frame relay network and not necessarily by any preallocated
end-to-end bandwidth as would be the case of a private line network. In a manner of speaking, this
provides bandwidth on demand since network bandwidth is allocated to this data path (virtual
circuit) only when traffic is present.
destination, whether via private lines or circuit switched connections. In any event, these resources
are only available to traffic bound for that location and are reserved for that traffic whether that
traffic is present or not. Conversely, in a frame relay network, bandwidths within the network and
in the access lines are only allocated between any two end devices if there is traffic moving between
those devices. At other times, this bandwidth is made available to other network devices.
Therefore, the performance in a frame relay network is then only limited by the bandwidth
available at the access point to the frame relay network and not necessarily by any preallocated
end-to-end bandwidth as would be the case of a private line network. In a manner of speaking, this
provides bandwidth on demand since network bandwidth is allocated to this data path (virtual
circuit) only when traffic is present.
Bandwidth is provided by the network’s Permanent Virtual Circuit (PVC) service: each data frame
to be transmitted by an endpoint contains and is identified by a Data Link Connection Identifier
(DLCI). The DLCI identifies a pre-established path, or permanent virtual circuit, within the access
line to the frame relay network. The frame relay switch at the edge of the frame relay network, the
one to which the access line is directly connected, routes the packet to the intended destination
based upon the DLCI therein. Hence, each packet is routed independently through the network
based on the addressing information provided by this identifier.
to be transmitted by an endpoint contains and is identified by a Data Link Connection Identifier
(DLCI). The DLCI identifies a pre-established path, or permanent virtual circuit, within the access
line to the frame relay network. The frame relay switch at the edge of the frame relay network, the
one to which the access line is directly connected, routes the packet to the intended destination
based upon the DLCI therein. Hence, each packet is routed independently through the network
based on the addressing information provided by this identifier.
The two line protocols used for data encapsulation on a permanent virtual circuit are Point to Point
Protocol or FR_IETF. PPP specifies the operation of the PPP protocol over Frame Relay links.
Although the CyberSWITCH supports this method of encapsulation, Inband Protocol
Demultiplexing is not performed when a PVC is initiated because the system statically configures
the line protocol used for a PVC. FR_IETF is a multiprotocol encapsulation for Frame Relay,
currently specified by RFC 1490. FR_IETF protocols include IP, MAC Layer Bridge, IPX, and
AppleTalk.
Protocol or FR_IETF. PPP specifies the operation of the PPP protocol over Frame Relay links.
Although the CyberSWITCH supports this method of encapsulation, Inband Protocol
Demultiplexing is not performed when a PVC is initiated because the system statically configures
the line protocol used for a PVC. FR_IETF is a multiprotocol encapsulation for Frame Relay,
currently specified by RFC 1490. FR_IETF protocols include IP, MAC Layer Bridge, IPX, and
AppleTalk.
Although Frame Relay is transparent to each of the protocols specified by FR_IETF, there are a few
special considerations to note. ARP, RARP, and IARP are protocols provided by FR_IETF for IP
over Frame Relay. These protocols are used to determine the IP and DLCI information used on the
virtual circuits. However, since this information is configured in the CyberSWITCH, these
protocols are not supported. The CyberSWITCH’s implementation of FR_IETF supports the Bridge
Point to Point model. The Virtual port model or Extended Spanning Tree is not supported.
special considerations to note. ARP, RARP, and IARP are protocols provided by FR_IETF for IP
over Frame Relay. These protocols are used to determine the IP and DLCI information used on the
virtual circuits. However, since this information is configured in the CyberSWITCH, these
protocols are not supported. The CyberSWITCH’s implementation of FR_IETF supports the Bridge
Point to Point model. The Virtual port model or Extended Spanning Tree is not supported.
The PVC name associates the permanent virtual circuit with a device table entry, whether it is
defined in an on-node or off-node database. If an on-node device database is used, the PVC name
must match the device name if outbound authentication has been disabled for a device associated
with a PPP virtual circuit, or if FR_IETF has been configured as the line protocol. However, if
outbound authentication has been enabled for a PPP device, the PVC name isn’t required to match.
If an off-node device database is used, the PVC name must match the device name for both line
protocols. FR_IETF requires that all PVC names match a configured device database entry, since no
identification mechanism is provided by this line protocol. When upgrading from a previous
release of the UAA software, the CyberSWITCH will process the previous PVC name, which was
defined in an on-node or off-node database. If an on-node device database is used, the PVC name
must match the device name if outbound authentication has been disabled for a device associated
with a PPP virtual circuit, or if FR_IETF has been configured as the line protocol. However, if
outbound authentication has been enabled for a PPP device, the PVC name isn’t required to match.
If an off-node device database is used, the PVC name must match the device name for both line
protocols. FR_IETF requires that all PVC names match a configured device database entry, since no
identification mechanism is provided by this line protocol. When upgrading from a previous
release of the UAA software, the CyberSWITCH will process the previous PVC name, which was