?– Since it is the ends (the bridges in which the I-components are embedded) that
decide which paths to use, the bridges in between are not involved. The worst case
for failure detection and recovery is both fast and predictable.
?– The price paid for the speed and predictability of recovery is that multiple failures
can bring down both the primary and secondary links. The Spanning Tree
Protocols can take longer, but can provide connectivity between nodes as long as a
physical path remains. However, this is a situation with which most carriers are
both accustomed and comfortable.
418 Chapter 13
IETF TRILL
A project is underway in the Transparent Interconnection of Lots of Links (TRILL)
working group of IETF. It is not explicitly tackling the problem of provider networks,
but is mentioned here for completeness. Its basic purpose is to employ routing techniques
to forward bridged traffic. The TRILL website should be consulted for further
information at: www.ietf.org/html.charters/trill-charter.html.
Economic Assessment
The cost of creating a Carrier Ethernet network using bridging technology is dependent
on a number of questions that will have very different answers for different
carriers:
?– Does the access network??”those parts of the network that multiplex and demultiplex
Ethernet frames, but never switch them (bridge or route them)??”exist?
?– If the access network exists, does it deliver Ethernet frames to the bridges in a
manner that they can digest easily, or does it require complex edge interfaces?
?– Is the bridged Carrier Ethernet network a provider bridge (802.
Pages:
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943