Osi Model Is Split Into 7 Layers Computer Science Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

The OSI model is split into 7 separate layers. Each layer is responsible for carrying out functions used to transmit data from one network to another. The OSI model is separated into two sets, the Application Set and the Transport Set. The Application Set uses all three upper layers (Layers 5, 6 and 7); The Transport Set uses the lower layers.

The OSI model works by starting at the application layer which carries out its function then passes control to the next layer till it reaches the bottom. It then passes its way over the network connection and starts the process all over again starting from the bottom layer (Physical Layer) and working its way up the hierarchy till it reaches the required destination.

The Upper Layers of the OSI are responsible for all application functions such as managing the connection, formatting an encryption.

Layer 7 Application: This layer allows the user to communicate with the OS (Operating System) and provides application services such as SMTP, browsing the internet, network data sharing, allowing file and data transactions etc.

Layer 6 Presentation: This layer translates data into a format. It encrypts/decrypts data, compresses/decompresses data and handles graphics.

Layer 5 Session: Controls the timing of each session of user application processes.

The Lower Layers of the OSI are mainly used for network based functions such as addressing, routing and flow control.

Layer 4 Transport: This layer controls error recognition and recovery of lost or corrupted data.

Layer 3 Network: This layer decides which routing path is best for data transmission between nodes. All hardware devices such as routers and switches are used at this stage.

Layer 2 Data: This layer decides which physical protocol is to be used depending on what type of network and it also splits the data into frames for transmitting.

Layer 1 Physical: This layer manages all physical hardware such as physical connections, voltages etc.


The model below shows the comparison and relationship between both the TCP/IP model and the OSI model. Both models are very similar in which they both use a layered structure. The TCP/IP model splits up into four layers as you will see below, each layer has specific protocols. TCP/IP model is used to allow computers to communicate over various networks. http://technet.microsoft.com/en-us/library/cc786128(v=ws.10).aspx



Host to Host









Network Access

TCP/IP Model

OSI Model

Protocols of Each OSI Layer

Layer 7

DNS (Domain Name System): Changes the domain name e.g. glasgow.com into an IP address which makes it simpler for a computer to understand. DNS server works by matching all domain names to IP addresses which are held within a massive database. This protocol uses standard RFC 0974.

HTTP (Hypertext Transfer Protocol): This protocol is a request - respond protocol. It is divided into 4 different sections; connection, request, response and close.

Connection - To determine whether a connection has been established. A message will be displayed on the connection as to whether it has been successful or not.

Request - The user makes a HTTP request which is passed to the server.

Respond - If the connection to the Web server has been successful it will allow Web pages to be downloaded. If it has been unable to connect an error message will be displayed.

Close - Once all requested information has been successfully delivered, the connection can then closed.

HTTP is used for all communications over the WWW (World Wide Web) to exchange information. It is known as a stateless protocol as it does not need previous knowledge of which stage was run before it. This protocol uses standard RFC 2616.


DHCP (Dynamic Host Configuration Protocol): DHCP assigns IP addresses to devices on a network by determining whether the IP is static or dynamic.

Static Addresses: This is an address that is manually applied and does not change.

Dynamic Addresses: This is the opposite of a static address which means that addresses are automatically applied and will automatically change in accordance to the network. http://business.bt.com/broadband-and-internet/business-broadband/static-ip/

This protocol uses standard RFC 3131.

POP3 (Post Office Protocol 3): A method used in order to retrieve emails from a remote server. This protocol uses standard RFC 1939.

SMTP (Simple Mail Transfer Protocol): Used to send emails to remote server. This protocol uses standard RFC 1869.

Layer 6

This layer provides several different functions these include:

Data conversion which helps decipher image, video, sound and audio formats.

Translation from different character codes such as RTF (rich text Format) and (ASCII) American standard code for information interchange, among others.

Layer 5

PAP (Password Authentication Protocol): is known to use a two way process for authentication. A username and password is requested and sent. A response is then sent determining whether it has been a successful attempt or a failed attempt. http://books.google.co.uk/books?id=hJ-OF2w5iOMC&pg=PA140&dq=ppp+components&hl=en&sa=X&ei=eDt4Ub3jK8LkPJftgYgK&ved=0CFAQ6AEwBQ#v=onepage&q=ppp%20components&f=false

Layer 4

TCP (Transmission Control Protocol): TCP is known as a 'suite' of protocols as it is used within IP, UDP, ARP etc. TCP transmits data in the 0correct order between a point to point connection. Although if the connection fails then the server will get send the lost part of the file or message. TCP allows communication between an IP and an application. This protocol uses standard RFC 0793. http://www.cyberciti.biz/faq/key-differences-between-tcp-and-udp-protocols/

UDP (User Datagram Protocol): Unlike TCP UDP is connectionless which means there is a greater chance data packets may get lost or corrupted while being transferred to its destination. It does not provide a correct order delivery and does not track connections. Although there is a greater chance of packets being lost UDP allows for each individual packet to arrive whole. This protocol uses standard RFC 0768. http://www.cyberciti.biz/faq/key-differences-between-tcp-and-udp-protocols/

Layer 3

IPSec (Internet Protocol Security): IPSec offers improved security which means it is offers a secure network connection. IPSec transmits data safely and securely making it completely confidential due to the authentication process. IPSec is commonly used within L2TP tunnelling as is supports data encryption.

IPv4: IPv4 is an Ethernet based protocol that sends packets of data across the network and routes traffic over the internet or in other words it is a packet switched link layer protocol. IPv4 is known as a connectionless protocol, this means that data sent has only one intention and that is to reach its destination and not taking into account duplication, security and integrity of data. This is handled by TCP; which is an upper layer protocol. IPv4 uses a 32bit format in 4 octets. Each octet provides a number ranging from 0 to 255; for example IPv4 are needed to communicate on a network or the internet, each sending and receiving point need an IPv4 in order to do this. http://pc.net/glossary/definition/ipv4 (For IPv4 In-depth refer to - Section 2)

IPv6: IPv6 is more or less and upgrade to its earlier predecessor IPv4. IPv6 has the same feature as IPv4, just not as faulty and with fewer limitations. IPv6 has some additional features such as new header formats, larger address space (it has a 128 bit address space), configuration settings for stateless and stateful addressing, IPSec header support and is improved for prioritised delivery. http://books.google.co.uk/books?id=jag8LFbABSkC&printsec=frontcover&dq=ipv6&hl=en&sa=X&ei=c0Z4UZnKBMWsO8_MgfAF&ved=0CGAQ6AEwBg

RIPv2 (Routing Information Protocol): RIPv2 is a distance vector protocol and is an improved version of RIP as it provides features such as authentication for routing updates and also includes the subnet mask of the network address. RIPv2 has become a popular choice of protocol for routing. RIPv2 is not used for large networks as is it limited to only 15 hops maximum. This protocol uses standard RFC 1058. http://www.orbit-computer-solutions.com/RIPV2.php

ICMP (Internet Control Message Protocol): ICMP is used for error information and error messages. For example when a node is overloaded ICMP reports the problem by sending a message. This protocol uses standard RFC 0792. http://www.erg.abdn.ac.uk/~gorry/eg3567/inet-pages/icmp.html

Layer 2

ARP (Address Resolution Protocol): ARP is used to convert IP addresses into physical addresses such as a MAC address. A host which needs a physical address will send an ARP request to the TCP/IP network. The host which has the IP address requested then replies with the physical hardware address. This protocol uses standard RFC 0826. http://www.webopedia.com/TERM/A/ARP.html

IEEE802.11: Has two modulation techniques DSSS (Direct Sequence Spread Spectrum) and FHSS (Frequency Hopping Spread Spectrum). DSSS uses different Phase shift keys dependant of the speed, 1Mbps uses Differential Binary Phase Shift Keying (DBSPK) and for 2Mbps it uses Differential Quadrature Phase Shift Keying (DQPSK). FHSS is an entirely popular modulation technique when compared to its counterpart DSSS, mainly due to speculation on whether or not FHSS adds to the quality of security or not but no proof has been provided as to say it does. As the years progressed IEEE 802.11 standard has seen improvements upon the system such as 802.11a, 802.11b and 802.11g; among others. http://www.cisco.com/web/about/ac123/ac147/archived_issues/ipj_5-1/ieee.html

WPA2 (Wi-Fi Protected Access 2): This standard is a method of authentication and security for a WLAN, one the most important aspect of WPA2 lies in its encryption algorithm known as AES (Advance Encryption Standard) which outdone the DES (Data Encryption Standard). In terms of security, WPA2 has two modes in which it can implement. The first being Personal mode or otherwise know home user mode in which a pre shared secret is used; essentially a password using 64 ASCII characters or a 256 bit number but is just as bad when compare to WEP. The second method and a much better one being the Enterprise mode or otherwise know corporate mode. This mode uses 802.11X, EAP or more specifically EAP-TLS (Extensible Authentication Protocol - Transport Layer Security) as well as RADIUS to keep it secure. http://books.google.co.uk/books?id=L__jFgCNJuEC&pg=PA139&dq=WPA2&hl=en&sa=X&ei=7j5wUZ3QJYHBONOrgJgF&ved=0CFAQ6AEwBA#v=onepage&q=WPA2&f=false

PPP (Point to Point): Used for communicating between two points. PPP is commonly used for a fast connection needed for broadband communications. PPP handles three main components; link control protocol, protocol encapsulation and network control protocols.

Protocol Encapsulation - Transports IP Datagrams and encapsulates them in order to transmit then over the physical layer.

Link Control Protocol - Configures, maintains and stops the link between devices.

Network Control Protocols - Allows additional configuration for the encapsulation of layer 3 datagrams. http://books.google.co.uk/books?id=Pm4RgYV2w4YC&pg=PA146&dq=ppp+components&hl=en&sa=X&ei=_j54UcfOD4HCPOexgdAJ&ved=0CDMQ6AEwAA#v=onepage&q=ppp%20components&f=false

PPP goes through many stages to complete a PPP connection. Firstly PPP parameters are setup using LCP (Link Control Protocol). After this stage is complete and the configuration negotiated is decided on then the authentication process is then carried out. If the authentication process is successful then the answered PPP stops and allows a connection to the original PPP; this process is known as callback. After PPP is setup, the original PPP is successfully authenticated and it is then up to NCP (Network Control Protocols) to carry out services such as encryption and compression to other data protocols. http://books.google.co.uk/books?id=BS7vAFa07FoC&pg=PT157&dq=ppp+protocol+process&hl=en&sa=X&ei=dUZ4UYfrDceCOM6jgcgH&ved=0CEEQ6AEwAg

CHAP (Challenge-Handshake Authentication Protocol): CHAP is part of PPP. CHAP works by sending a challenge message to the requestor which then returns a response using a value using a one-way hash function. The server then compares its own end value calculation against a calculation of the hash value. If the calculation is a match then the authentication is acknowledged, if not then the connection will be dropped. Both PPP and CHAP use standard RFC 1661, RFC 1662 and RFC 1663. http://searchcio-midmarket.techtarget.com/definition/CHAP

Layer 1

FDDI (Fibre Distributed Data Interface): FDDI transmits data at a speed of 100 mbps which means it can manage high traffic networks. FDDI uses a physical ring in order to transmit data, this means that if one ring goes down then a secondary ring is there to provide backup. Fibre optic connections transmit data by sending pulses of light which creates a wave through the optical fibre. This protocol uses standard ANSI. http://books.google.co.uk/books?id=zXDWNXRPos4C&pg=PA49&dq=802.11+standards+in+relation+to+FDDI&hl=en&sa=X&ei=P01sUYWtHoWMOK_2gfAC&ved=0CDIQ6AEwAA#v=onepage&q=802.11%20standards%20in%20relation%20to%20FDDI&f=false

10BASE-T: Twisted pair telephone cable also known as twisted pair Ethernet. It runs at a speed of 10Mbps which makes it a suitable speed. This protocol uses standard IEEE 802.3. http://www.webopedia.com/TERM/1/10BaseT.html

100BASE-TX: 100 BaseTX is a fast Ethernet standard which uses a twisted pair cable media system to receive and transmit data signals. This protocol uses standard IEEE 802.3. http://books.google.co.uk/books?id=MRChaUQr0Q0C&pg=PA142&dq=100BASE-TX&hl=en&sa=X&ei=Q_N2Uc6NKael4ATp7IHABA&ved=0CDsQ6AEwAQ#v=onepage&q=100BASE-TX&f=false

100BASE-FX: Uses fibre optic cabling which operates at 100 Mbps. It uses two strands of fibre; one for data transmission and the other for reception. This protocol uses standard IEEE 802.3. http://books.google.co.uk/books?id=UD0h_GqgbHgC&pg=PA212&dq=100base+fx&hl=en&sa=X&ei=DSB4UZC4JMHiOpLggZgL&ved=0CEsQ6AEwBA#v=onepage&q=100base%20fx&f=false

V.34 Modem: Allows for data to be sent and received across telephone lines.

The table below shows the process of the individual segments within the network provided to us.

Laptop To Access Point

Layer 1 Protocol

802.11 (Wireless connection)

Layer 2 Protocol


Layer 3 Protocol


Layer 7 Protocol


Access Point to Router 1

Layer 1 Protocol

10Base-T (Twisted Pair Cable)

Layer 2 Protocol


Layer 3 Protocol


Layer 7 Protocol


Router 1 to Router 2

Layer 1 Protocol

V.34 (Telephone Connection)

Layer 2 Protocol

Ethernet and PPP (CHAP is used for Authentication)

Layer 3 Protocol


Layer 4 Protocol


Router 2 to Switch

Layer 1 Protocol

100Base-FX (Fibre Optic Cable)

Layer 2 Protocol


Layer 3 Protocol


Layer 7 Protocol


Switch to DNS and Web Servers

Layer 1 Protocol

100Base-TX (Unshielded Twisted Pair)

Layer 2 Protocol


Layer 3 Protocol


Layer 7 Protocol