Network Handout

ISO/OSI Seven Layer Model

The goal of this protocol stack is a regulated, well-defined exchange of data between the application processes. The layering is based on the principle that every layer can take advantage of the services of the next lower layer without knowing how their services are provided. A layer offers its own service to the next respective higher layer. This makes it possible to achieve a division of labor within the layers.

Consequently, every layer:

- Has limited, defined tasks
- Has a precisely defined interface to the neighboring higher and lower layers.
- Attaches its own layer-specific header to the data package being passed on. The corresponding layer on the other side interprets and removes the header.

The individual layers of the OSI model are listed in Table1-1.
 
 
Table1-1  ISO/OSI Network Model Layers
ISO/OSI Layer
Description
Application
Provides a service for managing the application.
Presentation
Manages the presentation of the data to be 
independent of the architecture.
Session
Administers communication relationships.
Transport
Makes sure that messages reach their destination 
system via an optimal transmission path.
Network
Manages data addressing and delivery between networks, 
as well as fragmenting data for the Data Link Layer. A 
router functions at this layer (routing).
Data Link
Manages the delivery of data across the physical 
network. This layer provides error detection and
 packet framing. A bridge/switch functions at this layer.
Physical
Describes the network hardware, including electrical 
signal characteristics such as voltage and current. A 
repeater or hub function at this layer.


TCP/IP Five-Layer Model

Layered Model

The TCP/IP protocol suite is structured as a hierarchy of five layers, sometimes referred to collectively as a protocol stack. This architectural scheme provides the following benefits:

- Each layer is designed for a specific purpose and exists on both the sending and receiving hosts.
- Each layer is designed so that a specific layer on one machine sends or receives exactly the same object sent or received by its peer process on another machine.
- Each layer on a host acts independently of other layers on the same machine, and in concert with the same layer on other hosts.

Table1-2 lists each layer in the TCP/IP network model.
 
 

Table1-2 TCP/IP network model
TCP/IP Layer
Description
Application
Consists of user-accessed application programs and 
network services. This layer is also responsible for defining 
the way in which cooperating networks represent data. A 
gateway functions at this layer.
Transport
Manages the transfer of data using acknowledged and 
unacknowledged transport protocols. This layer also 
manages the connections between cooperating applications.
Internet
Manages data addressing and delivery between networks, as 
well as fragmenting data for the network interface layer. A 
router functions at this layer.
Network Interface
Manages the delivery of data across the physical network. 
This layer provides error detection and packet framing. A 
bridge/switch functions at this layer.
Hardware
Describes the network hardware, including electrical signal 
characteristics such as voltage and current. A repeater or 
hub functions at this layer.



 

IPv4 Netmask


Computing Network Numbers

Given this definition of the logical AND operator, consider the IPv4 address of 171.63.14.3, a Class B address whose netmask is, therefore, 255.255.0.0. Figure 1-1 illustrates how a network number is computed.
 
 
 

IPv4 address in decimal 171.63.14.3
IPv4 address in binary 10101011  00111111  00001110  00000011
Class B netmask in decimal 255.255.0.0
Class B netmask in binary 11111111  11111111  00000000  00000000
Apply the logical AND operator
IPv4 Address (decimal) 171.63.14.3
IPv4 Address (binary) 10101011  00111111  00001110  00000011
AND netmask 11111111  11111111  00000000  00000000
Network # (binary) 10101011  00111111  00000000  00000000
Network # (decimal) 171.63.0.0
Figure1-1  Network Number Computation

Thus, the resulting network number is 171.63.0.0 in decimal. Notice that the host portion of the address is zero (masked out).


Table1-3 shows the possible Class C subnet masks.
 

Table1-3 Class C Subnet Masks
Mask in Decimal
Mask in Binary
Number of Subnets
Number of Hosts per subnet
255.255.255.0
11111111  11111111  11111111  00000000
1
254
255.255.255.128
11111111  11111111  11111111  10000000
2
126
255.255.255.192
11111111  11111111  11111111  11000000
4
62
255.255.255.224
11111111  11111111  11111111  11100000
8
30
255.255.255.240
11111111  11111111  11111111  11110000
16
14
255.255.255.248
11111111  11111111  11111111  11111000
32
6
255.255.255.252
11111111  11111111  11111111  11111100
64
2

 
 
 

Table1-4 shows the possible Class B subnet masks.
 

Table1-4 Class B subnet masks
Mask in Decimal
Mask in Binary
Number of Subnets
Number of Hosts per Subnet
255.255.0.0
11111111  11111111  00000000  00000000
1
65534
255.255.128.0
11111111  11111111  10000000  00000000
2
32766
255.255.192.0
11111111  11111111  11000000  00000000
4
16382
255.255.224.0
11111111  11111111  11100000  00000000
8
8190
255.255.240.0
11111111  11111111  11110000  00000000
16
4094
255.255.248.0
11111111  11111111  11111000  00000000
32
2046
255.255.252.0
11111111  11111111  11111100  00000000
64
1022
255.255.254.0
11111111  11111111  11111110  00000000
128
510
255.255.255.0
11111111  11111111  11111111  00000000
256
254
255.255.255.128
11111111  11111111  11111111  10000000
512
126
255.255.255.192
11111111  11111111  11111111  11000000
1024
62
255.255.255.224
11111111  11111111  11111111  11100000
2048
30
255.255.255.240
11111111  11111111  11111111  11110000
4096
14
255.255.255.248
11111111  11111111  11111111  11111000
8192
6
255.255.255.252
11111111  11111111  11111111  11111100
16384
2


Easy way to determine subnets, network numbers, broadcast numbers for a subnetted Class C network.

Given : 192.168.11.140/27

Convert IP address to binary : 11000000  10101000  00001011  10001100

A Class C network masks the first 24 bits (from the left). In this example, we are masking 27 bits. That means the 3 bits, on the left, of 10001100 will be used for the network and the remaining 5 bits will be used for hosts.

# of subnets available = 23  = 8 (the power is the number of additional bits beyond the 24 that will be used for the network)

# of hosts available = 25  -  2  = 30 (the power is the number of bits used for hosts) Note: we subtract the first and last because those are reserved for the network number and the broadcast number.

So, here are the network and broadcast numbers for each of the subnets:
(Our IP address is on the 100 subnet)

Subnets are all binary permutations, of the available bits.
 

Subnet
Network Number
Broadcast Number
000
0
31
001
32
63
010
64
95
011
96
127
100
128
159
101
160
191
110
192
223
111
224
255

 

Modified by Marty Froomin on  4/7/04