Transcript 2.2 Three Primary Classes Of IP Addresses

2.1 Universal Identifier
In general, addressing in a network is closely linked with the issue of
routing. There are really at least 3 aspects of addressing to be considered:
•
•
•
the name of an object, which identifies what it is.
the address of an object, which tells us where it is.
the routing information, which tells us how to get to it.
In networking, an address can either be:
CSE5803 Advanced Internet Protocols and Applications (2)
1
•
•
flat, e.g. a 48-bit Ethernet address. The route must be determined
entirely by some form of look-up on the whole address.
structured, e.g. a telephone number (+61 3 9905 3402), which can
be parsed to determine the route to the object.
In the Internet, a two-stage address is used:
1.
2.
an alphabetic domain name (examined later)
a 32-bit binary network address.
The network address effectively identifies an NSAP, i.e. the Network
layer entity associated with a connection to a particular physical network.
Each host and router has one or more connections, and hence one or more
IP addresses.
CSE5803 Advanced Internet Protocols and Applications (2)
2
The binary IP addresses comprise a two-level hierarchy:
•
•
a netid, which uniquely identifies a subnetwork
a hostid which identifies a link (host/gateway) on that subnetwork.
The address does not specify an individual computer, but a connection to
the network.
Every IP address is associated with a physical address according to the
technology of the underlying physical network.
CSE5803 Advanced Internet Protocols and Applications (2)
3
2.2 Three Primary Classes Of IP Addresses
CSE5803 Advanced Internet Protocols and Applications (2)
4
• The convention is that binary IP addresses are written in “dotnotation”, in which the four octets are written as decimal numbers (0255), with ‘.’ between them, e.g. 211.126.1.49.
• Classes A, B and C in the following table include private, broadcast
and other special addresses that may not normally be used in the
Internet.
• Class D is for multicast and Class E is reserved for experiment and
future use.
Class A IP: 1.0.0.0 - 126.255.255.255
Class B IP: 128.0.0.0 – 191.255.255.255
Class C IP: 192.0.0.0 – 223.255.255.255
Class D IP: 224.0.0.0 – 239.255.255.255
Class E IP: 240.0.0.0 – 254.255.255.255
(Monash's IP addresses are in the form 130.194.x.x. Which Class is this?)
CSE5803 Advanced Internet Protocols and Applications (2)
5
•
Special addressing conventions (RFC3330 for more details):
•
The whole range of 255.x.x.x are normally reserved for
broadcasting.
Destination IP of all 1s means broadcast to host’s own local
network. While net plus all 1s means targeted broadcast to a specific
net.
•
CSE5803 Advanced Internet Protocols and Applications (2)
6
• Class A allows up to 126 networks and up to 16,777,214 host links
p.n. (why? 0 net cannot be used, 127 net is reserved for loopback,
all 0s and 1s hosts cannot be used)
• Class B allows up to 16,384 networks and 65,534 host links p.n.
• Class C allows up to 2,097,152 networks and 254 host links p.n.
• IANA (Internet Assigned Number Authority) controls the numbers
• Address distributed by Internet registries.
2.3 Private and Global Connectivity
• Global connectivity means that hosts inside an organization have
access to both internal and Internet hosts.
• Private connectivity means that hosts have only internal access (Bank
ATM & ?)
• Private hosts do not have to be assigned global unique IP, in order to
save IP address.
CSE5803 Advanced Internet Protocols and Applications (2)
7
• IANA has designated the following IP addresses for private
connectivity only (check www.iana.org for latest definition):
Class A
10.x.x.x (a complete class A)
Class B
172.16.0.0 – 172.31.255.255 (16 class Bs)
Class C
192.168.0.0 – 192.168.255.255 (256 class Cs)
• These address can be selected for internal connectivity without IANA
or an Internet Registry. Users must keep them within the internal nets
(as per RFC1918).
• RFC1700, 2544, 3068, 3171 are also related with IP addressing.
2.4 Network Natural Masks
• A natural network mask is implied by address class and is used to
determine the network and host portion of an IP address. In a mask,
any bit set to a one (1) defines the network address, zero (0) defines the
host address.
• Class A natural Mask 255.0.0.0
• Class B natural Mask 255.255.0.0
• Class C natural Mask 255.255.255.0
CSE5803 Advanced Internet Protocols and Applications (2)
8
2.5 Subnet Addressing
• This allows a site to subdivide a single Internet (network) address for
use across multiple physical networks.
• Limited IP network address from registries.
• IP datagrams are routed based upon the network portion of the address.
Routing is not possible with multiple physical networks without
subnetting.
CSE5803 Advanced Internet Protocols and Applications (2)
9
• Subnet addressing reallocates certain bits of the host portion of the
Internet address to identify the subnet.
CSE5803 Advanced Internet Protocols and Applications (2)
10
• Subnet address restrictions: more bits for subnetworks ->
fewer bits for host numbers, vice versa. (remember all 0s
and all 1s are not used as normal host numbers. What’s the
smallest subnet? two bits and two hosts)
• Subnet expansion – a related planning issue.
• Subnet mask: 1 for every bit of IP network address and
subnet address, 0 for every bit of host number
• A basic subnetting rule for RIP-1: the same mask must be
applied throughout the physical networks that share the
same network address.
CSE5803 Advanced Internet Protocols and Applications (2)
11