H3C Technologies H3C S10500 Series Switches - Manual (Page 73)

58
Area based OSPF network partition
Network partition
In a large OSPF routing domain, the LSDB becomes very huge and SPF computation consumes many
storage and CPU resources.
In addition, because topology changes can easily occur, OSPF packets generated for route information
synchronization are enormous, occupying excessive bandwidth.
To solve these problems, OSPF splits an AS into multiple areas, each of which is identified by an area ID.
The boundaries between areas are routers rather than links. A network segment (or a link) can only
reside in one area. An OSPF interface must be specified to belong to its attached area, as shown in
Figure 17 .
Figure 17 Area based OSPF network partition
After network partition, ABRs perform route summarization to reduce the number of LSAs advertised to
other areas and minimize the effect of topology changes.
Backbone area and virtual links
Each AS has a backbone area that distributes routing information between none-backbone areas.
Routing information between non-backbone areas must be forwarded by the backbone area. OSPF
requires the following:
• All non-backbone areas must maintain connectivity to the backbone area.
• The backbone area itself must maintain connectivity.
In practice, the requirements may not be satisfied due to lack of physical links. OSPF virtual links can
solve this problem.
A virtual link is established between two ABRs through a non-backbone area and is configured on both
ABRs to take effect. The non-backbone area is called a transit area.
In the following figure, Area 2 has no direct physical link to the backbone area 0. You can configure a
virtual link between the two ABRs to connect Area 2 to the backbone area.