TCP/IP protocol suites 2

Learning Outcomes:

Know the basic concepts of IP addressing including the scheme and classes of IP addresses.
Compare TCP/IP protocol suites such as Internet Protocol version 4 (IPv4) and Internet Protocol version 6 (IPv6) from an evolutionary perspective.

Different versions of TCP/IP protocol suites are compared in terms of address size and their support for audio and video.

Internet Protocol (IP):

Features: Similar to UDP, it is unreliable and connectionless

Functions: IP addressing, encapsulation, routing and ultimately delivery of the IP packets to correct destination

IPv4 vs IPv6:

IPv4 address is 32-bit long. The number cannot cope with the huge demand from the expanding network. Thus, IPv6 is developed with its address which is 128-bit long. Additional features are also available.

Features of IPv6:

128-bit long addresses

Efficient packet handling: e.g. simplified header, traffic class, flow label

Improved performance: multicast – sending datagrams to multiple receivers in a single transmission

Improved security: e.g. IPsec fully integrated

Relevant past paper:

DSE ICT Elect B(SP-2017): 2012 2d.

CE CIT(2005-2011): 2009 2c.

IPv4

Address classes:

Binary notation 11000000.10101000.00000000.00000001

Decimal notation 192.168.0.1

The IP address is separated by dots into 4 parts. Each part is called an octet. The leading bits determine the class.

Network address is for identifying the network while the host address is for identifying the devices in that network.

Class	leading bits	1st octet	Network address portion	Host address portion
A	0-------	0-127	1^st octet	2^nd-4^th octet
B	10------	128-191	1^st-2^nd octet	3^rd-4^th octet
C	110-----	192-223	1^st-3^rd octet	4^th octet
D	1110----	224-239
E	1111----	240-255

Class D and E are not important for the exam purpose.

Relevant past paper:

DSE ICT Elect B(SP-2017): SP 4e. PP 1di. 2012 4ai.

CE CIT(2005-2011): 2006 4e. 2007 2b. 2008 1f. 2011 3f.

Local(private) vs global(public) IP address:

Some IP addresses are reserved for private use (use in a LAN). i.e. it cannot be used to identify a network/host through the internet.

A router is in a special location where it is between 2 different networks, usually a LAN and the internet. The ISP assigns a public IP address for the router to use so that others can identify and reach the router through the internet. The router also has a private/local IP address so that devices in the LAN can reach the router. 192.168.0.1 is used by many routers as the default (local) IP address.

Relevant past paper:

DSE ICT Elect B(SP-2017): 2017 2ci.

CE CIT(2005-2011): 2009 2f. 2010 4bvii.

AS CA(2000-2013): 2009 9ei.

Network address translation (NAT):

The translation of local IP address into global IP address and vice versa. This allows the hosts in a LAN to communicate with the outside world.

A special form of NAT called NAT overload allows multiple hosts in a LAN to communicate with the outside world using 1 public address only. This helps to reduce the demand of the limited number of IP addresses. Port number is used to identify different hosts in the LAN.

NAT table:

Local IP address and source port	Port number used in NAT
192.168.0.3:10100	20000

When data is sent out from an application of a host, the router received the information 192.168.0.3:10100. This information is saved in the NAT table in the router and mapped to a new port number. Suppose the public IP address of the network is 11.100.100.100. The outgoing data will now have a source IP address and port number as 11.100.100.100:20000.

When returning data targeting 11.100.100.100:20000 reaches the router, it maps the port number 20000 to the local IP address and port 192.168.0.3:10100. The data will therefore be directed to the relevant host and application.

Relevant past paper:

DSE ICT Elect B(SP-2017): PP 1div.

CE CIT(2005-2011): 2009 4c. 2010 4biv.

AS CA(2000-2013): 2007 2c.

Subnetting:

Reason for subnetting:

Usual IP address can be divided into network and host number based on its class as mentioned before. This represents 1 network with certain number of hosts. To facilitate network management, for example a large company, we want to divide the single network into a number of networks.

Use binary notation for understanding

Without subnet:

IP address 192.168.123.132 11000000.10101000.01111011. | 10000100

(Default) subnet mask 255.255.255.0 11111111.11111111.11111111. | 00000000

Network address Host address

The subnet mask determines which point to divide the IP address into network address and host address.

The default subnet masks for class A, B and C addresses are 255.0.0.0, 255.255.0.0 and 255.255.2552.0, respectively.

Subnetting a private network:

To divide a network into 4 networks, we use subnet mask 255.255.255.192 for a class C address.

IP address 192.168.123.0 11000000.10101000.01111011.00 | 000000

Subnet mask 255.255.255.192 11111111.11111111.11111111.11 | 000000

Network address Host address

So, the four networks are:

11000000.10101000.01111011.00(000000) 192.168.123.0 host addresses: 1-62

11000000.10101000.01111011.01(000000) 192.168.123.64 host addresses: 65-126

11000000.10101000.01111011.10(000000) 192.168.123.128 host addresses: 129-190

11000000.10101000.01111011.11(000000) 192.168.123.192 host addresses: 193-254

For a class C network:

No of subnets	Subnet mask 255.255.255.x x =	In binary notation x =	Number of subnets	Number of host addresses in each subnet	Number of hosts in each subnet *
2	128	10000000	2¹	2⁷=128	128-2
4	192	11000000	2²	2⁶=64	64-2
8	224	11100000	2³	2⁵=32	32-2
16	240	11110000	2⁴	2⁴=16	16-2
32	248	11111000	2⁵	2³=8	8-2
64	252	11111100	2⁶	2²=4	4-2