IP Addressing - University of Maine System

Download Report

Transcript IP Addressing - University of Maine System

IP Addressing
Lecture 11
October 30, 2000
Background



The goal of Universal Service is such that all
computers on all physically different networks
can communicate.
Physical addresses allow communication
between computers on one network.
A new level of abstraction must be be
introduced for internet communication.
IP Addresses



An IP address is the next layer of abstraction.
The IP address provides virtual addressing.
The address is software controlled, whereas
the address for the network card is hardware
based.
The IP addressing scheme is quite complex,
and there have been many revisions to the IP
scheme.
IP Addresses (cont.)


IP addressing allows for seamless integration
amongst heterogeneous networks.
To send a packet, the destination address is
the IP address of the computer, not the
hardware address! This allows for
communication across networks.
IP Addresses (cont.)





32 bits in length (IPv4)
64 bits in length (IPv6)
Addresses are divided into a prefix and suffix
The suffix is the host address
The prefix is the network number
IP Classes




People commonly throw around terms like
“Class C”, but it should really be termed “Class
C address” or “Class C address space.”
Class A: 16777216 hosts!
Class B: 65536
Class C: 256
IP Class Scheme
IP Class Scheme



From the previous figure, we see that the 32-bit
address is split into 4 octets.
IP addresses are self identifying.
If the first 4 bits of the first octet are
–
–
–
–
–
0xxx: Class A address
10xx: Class B address
110x: Class C address
1110: Class D address (Multicast)
1111: Class E address
Dotted Decimal



IP addresses are generally read in dotted
decimal format.
0.0.0.0 through 255.255.255.255
Much better than reading:
10000001 00110100 00000110 00000000
Dotted Decimal with Classes

Class A:
–
–

Class B:
–
–

1 prefix octet (128 networks)
3 suffix octets (16777216 hosts)
2 prefix octets (16384 networks)
2 suffix octets (65536 hosts)
Class C:
–
–
3 prefix octets (2097152 networks)
1 suffix octet (256 hosts)
Address Space
Address Delegation



A central authority exists for IP address
delegation.
In the US, it’s ARIN – American Registry for
Internet Numbers (http://www.arin.net)
People just can’t arbitrarily use any IP network
if their network is publicly accessible! That
would lead to routing conflicts.
Address Delegation (cont.)

RFC 1597 – Private networks
–
–
–
10.0.0.0 – 10.255.255.255 (Full Class A)
172.16.0.0 – 172.31.255.255 (16 Class B’s)
192.168.0.0 – 192.168.255.255 (Full Class B)
Special Addresses

Network address
–
–
–
Host 0 address for specific class type
16.0.0.0 is the network address for the Class A
prefix of 16.
130.111.0.0 is the network address for the Class B
prefix of 130.111.
Special Addresses (cont.)

Directed Broadcast Address (Network
Broadcast Address)
–
–
–
A network suffix of all 1’s.
16.255.255.255 is the directed broadcast address
for the Class A prefix of 16.
130.111.255.255
Special Addresses (cont.)

Limited Broadcast Address
–
–
–
All 1’s in the entire address.
Limited broadcast address is restricted to the local
subnet.
255.255.255.255
Special Addresses (cont.)

Loopback addresses
–
–
–
Loopbacks are used for testing. An IP looback is
application-level testing.
Any information sent to the loopback address is
never passed to the network segment. It is handled
internally in the TCP/IP stack.
127.x.x.x
Special Addresses (cont.)

This computer’s address
–
–
If a computer doesn’t know what it’s own address is,
but needs to communicate to another machine, it
designates the address of 0.0.0.0 for itself.
Applications include DHCP, BOOTP
IP and Routers
IP and Routers (cont.)


We’ve seen from our project that routers do not
necessarily have a single IP address.
Commonly have 2 (or more addresses)
–
–
IP address for the LAN interface
IP address for the WAN interface