Nortel Networks 3500 User Manual
2-72 Operation, administration, and maintenance (OAM) features
OPTera Metro 3500 Multiservice Platform NTRN10AN Rel 12.1 Standard Iss 1 Apr 2004
interfaces for data management, such as the Bay Command Console
command line interface (BCC CLI) and Simple Network Management
Protocol (SNMP).
command line interface (BCC CLI) and Simple Network Management
Protocol (SNMP).
Distributed multilink trunking
OPTera Metro 3500 supports network-to-network interface (NNI) redundancy
through trunk groups. A trunk group is a logical group of two NNI ports in the
same ring that are on separate 2xGigE circuit packs. Ports in a trunk group
share traffic according to a hashing algorithm. In case of a failure on one port
in a trunk group, all traffic is switched to the other port until the failure clears.
Trunk groups enable redundancy between Resilient Packet Rings (RPR) by
providing two NNI links between RPRs instead of one. The extra NNI link
protects traffic between rings.
through trunk groups. A trunk group is a logical group of two NNI ports in the
same ring that are on separate 2xGigE circuit packs. Ports in a trunk group
share traffic according to a hashing algorithm. In case of a failure on one port
in a trunk group, all traffic is switched to the other port until the failure clears.
Trunk groups enable redundancy between Resilient Packet Rings (RPR) by
providing two NNI links between RPRs instead of one. The extra NNI link
protects traffic between rings.
For more information, see:
•
OPTera Packet Edge System Planning Guide, NTRN10YK
•
OPTera Packet Edge System User Guide, NTN465YG
Bandwidth Reservation Protocol (BRP)
OPTera Packet Edge rings support a bandwidth reservation protocol (BRP)
algorithm that allows you to provision a guaranteed bandwidth rate, called the
reserved rate, for each node in the ring. The BRP algorithm attempts to provide
access to the ring at or above the reserved rate of the node. The BRP algorithm
works by sending credit packets from a source node to a downstream node
when the source node is exceeding its reserved rate such that the downstream
node cannot access the ring at its reserved rate. The credit packets are dropped
by the downstream node, creating "holes" (gaps in the data stream) that allow
the downsteam node to increase the rate at which it adds traffic to the ring. The
source node sends enough credit packets to allow the downstream node to
access the ring at or above its reserved rate.
algorithm that allows you to provision a guaranteed bandwidth rate, called the
reserved rate, for each node in the ring. The BRP algorithm attempts to provide
access to the ring at or above the reserved rate of the node. The BRP algorithm
works by sending credit packets from a source node to a downstream node
when the source node is exceeding its reserved rate such that the downstream
node cannot access the ring at its reserved rate. The credit packets are dropped
by the downstream node, creating "holes" (gaps in the data stream) that allow
the downsteam node to increase the rate at which it adds traffic to the ring. The
source node sends enough credit packets to allow the downstream node to
access the ring at or above its reserved rate.
For more information about OPTera Packet Edge services, see:
•
•
OPTera Packet Edge System Planning Guide, NTRN10YK
•
OPTera Packet Edge System OPTera Metro 3000 User Guide, NTN465YG
1024 TDIs on a mapped UNI
You can have up to 1024 TDI values for each 2xGigE circuit pack and up to
256 TDI values for each port of the 4x100BT or 4x100FX circuit pack. You
can configure up to 1024 TDI to VLAN ID mappings for each circuit pack.
You must limit the number of mappings on each port of the 4x100BT or
4x100FX circuit pack to 256. On the 2xGigE circuit pack, there is no limit to
the number of mappings on a port, as long as the total number of mappings on
both ports does not exceed 1024.
256 TDI values for each port of the 4x100BT or 4x100FX circuit pack. You
can configure up to 1024 TDI to VLAN ID mappings for each circuit pack.
You must limit the number of mappings on each port of the 4x100BT or
4x100FX circuit pack to 256. On the 2xGigE circuit pack, there is no limit to
the number of mappings on a port, as long as the total number of mappings on
both ports does not exceed 1024.