Classless Inter-Domain Routing (CIDR)
For years, the Internet
has been growing at an alarming pace. The proliferation of the 'dot com' and
e-commerce companies caused a huge surge in the use of Internet
Protocol (IP)
addresses, and an increase in the number of destinations on the Internet.
Each destination has a range of IP
addresses associated with it (a 'block of IP's' in NetSpeak).
Routers
use routing
protocols such as BGP
exchange information about how to reach these destinations. As the number of
destinations grew, so did the number of routes (paths) to reach these networks.
Soon, it became clear that the routers
couldn't store the growing number of routes, they also couldn't handle the
optimum path calculations much longer, and the IP
address space was being handed out far too quickly because it was carved into
large classful
blocks.
CIDR: SUBNETTING, SUPERNETTING AND ROUTE AGGREGATION
The solution to the depletion of IP
addresses and the proliferation of the number of routes was two fold. ARIN used a three-pronged solution to this
problem.
To reduce the number of routes in the Internet
backbone, supernetting was used to aggregate destinations together into
larger blocks of IP's.
Larger blocks of IP's
would only be allocated to the ISP's. The ISP's were, in turn, expected to
aggregate their routes so that routing
protocols could make fewer announcements of larger blocks. Of
course, this created other problems for customers who were wise enough to not
count on a single ISP for their Internet access.
ROUTE AGGREGATION
ARIN recommended that all backbone
Internet providers combine or 'aggregate' their routes to reduce the total
number of routes being advertised. ARIN
further recommended that all companies and organizations wishing to connect to
the Internet request IP
addresses from their Internet provider instead of ARIN.
If a company gets a small block of IP
addresses from the larger block of IP
addresses owned by their upstream provider, their provider can more easilly
aggregate those routes because the customer's IP
addresses originally came from the larger block of IP
addresses owned by the Internet provider.
Classless Inter-Domain Routing Notation
BGP uses Classless
Inter-Domain Routing (CIDR) notation for masks. When advertising
routes, BGP will include prefixes in it's advertisements. A prefix is the network
IP address plus the mask
in CIDR notation.
Below are tables showing how IP masks and CIDR masks are related. In every
IP address, certain bits are used to identify the network, and certain bits are
used for host. The mask allows the receiver to tell which bits are network, and
which are host. Ones are used to mark the Network bits, and zeroes are used to
mark the Host bits.
|
|
CLASS 'A' NETWORKS
|
|
BINARY
|
11111111.00000000.00000000.00000000
|
|
DECIMAL
|
255.0.0.0
|
|
CIDR
|
/8
|
|
|
CLASS 'B' NETWORKS
|
|
BINARY
|
11111111.11111111.00000000.00000000
|
|
DECIMAL
|
255.255.0.0
|
|
CIDR
|
/16
|
|
|
CLASS 'C' NETWORKS
|
|
BINARY
|
11111111.11111111.11111111.00000000
|
|
DECIMAL
|
255.255.255.0
|
|
CIDR
|
/24
|
You will note that the number in the CIDR mask notation is equal to the number of 1's in the binary mask. The same holds true for subnets:
|
|
1/2 CLASS 'C'
NETWORK
|
|
BINARY
|
11111111.11111111.11111111.10000000
|
|
DECIMAL
|
255.255.255.128
|
|
CIDR
|
/25
|
And for supernets as well:
|
|
2 CLASS 'C'
NETWORKS
|
|
BINARY
|
11111111.11111111.11111110.00000000
|
|
DECIMAL
|
255.255.254.0
|
|
CIDR
|
/23
|
So for CIDR notation, the number after the slash is equal to the number of network bits (1's) in they mask.
No comments:
Post a Comment