Enables communication across the internet between different devices.

ApplicationTransportNetworkLink

Interact with the user, using application software (e.g. web browser) and select the correct protocol to use.

Establishes a virtual pathThen splits the transmission into equal sized packets. Assigns packages a sequence number and port number.

Provides the correct source and destination IP addresses, The routers within the network layer, use these to send each packet to its destination.

Controls the physical connections between hardware in the network. Add MAC addresses to packets, which change from each "hop" through the network.

An end-to-end connection between the sender and the recipient.

Used to identify which protocol is in use.

1. Link layer removed MAC addresses.
2. Network Layer removes IP addresses.
3. Transport layer ensures that packages are in correct sequence.
4. Transport layer uses the port number to route the packet to the appropriate application.
5. Application Layer receives packets and displays information to the user.

20 & 21

22

80

443

110 & 995

25, 587 & 465

Sending files between devices on a network, anonymously or with a login & passwordThere is an FTP sever and many clients

makes a TCP connectionremote management of computers, provides command line accessrequires username & password for securityand provides encryption

1. Web servers deliver content to clients with HTTP or HTTPS.
2. Clients render HTML files.
3. Only HTTPS encrypts the information and prevents unauthorised or unintended modification during transmission.

Post-Office-Protocol 3 retrieves emails from an email server. It works by checking for new messages and downloads them on the client.

Used for sending emails, to and between email servers.

The first part is network identifierThe second part host identifier

Identify which part of the IP address is the network identifier. Identify if the IP address is within the same network or outside of our network.

1. Convert the subnet mask to binary.

E.g. 255.255.255.244.0 → 11111111.11111111.11100000.00000000

2. There are 19 ""1""s → so the network identifier part is 19 bits.
3. Then subnet mask and IP addresses is ANDed bitwise.

107.56.94.111 → 01101011.00111000.01011110.01101111

• 01101011.00111000.01011110.01101111
• 11111111.11111111.11100000.00000000
• 01101011.00111000.01000000.000000000
Checking: 1101011.00111000.01000000.00000000 → 107.56.94.111/19

Dividing a large organisation into a small subnetworks can improve speed, stability, reduce network traffic and collision frequency.

Host identifier made longer, split into subnet identifier and host identifierNetwork identifier made a few bits shorter.

IPv4 = dotted quad numbers = 192.168.43.7IPv6 = eight blocks of 4 hexadecimal characters, separated by colons = 2071:0eb8:85a3:8a2f:0000:0000:0370:7264

4 parts (quad), each 8 bits = 4*8 = 32IPv4 = 32 bits = 2^32 unique addresses8 parts (blocks), each 16 bits = 8*`6 = 128IPv6 = 128 bits = 2^128 unique addresses.
IoT and increasing number of connected devices reduce supply of unique IPv4 addresses

Unique IP addresses are in short supply, so each site is given a few or one routable (public) IP addresses While each device within the network is given its own non-routable private IP address.

Only public IP addresses are globally unique. Public IP addresses are assigned to sites by global authorities, ensuring no address is issued twice

Ensures that the limited number of private IP addresses is never reached. So that devices which join once and never again do not take up unique addresses.

Dynamic Host Configuration Protocol assigns each device its own private IP address from a pool of available IP addresses for the duration of its sessionWhen a device disconnects its IP address returns to the pool.

As the packet leaves the private network the router makes a record of the packet's private sender IP and replaces it with the sides routable IP.

Router receives response at the public routable IP, uses the record previously made to forward the response packets to the device at the correct private IP.

Network Address Translation enables one device with only a private IP address to communicate with another similar device across the internet.

Enables communication between a client and a sever on a private network with a non-routable IP.

The clients sends messages to the public IP belonging the servers private network. The port number on the packets is used by router of the private network to forward it to the correct server computer via NAT.

1. A client computer sends request messages to server computers.
2. Server computers reply with response messages.
3. Each server is specialised in one task (file server, database server, web server, email server).

An Application Programming Interface is a set of protocols relating to how different applications may communicate with each other. Standard definitions allows applications to make use of other applications.

Provides a constant stream of information between two devices (client web browser, and server).

The API, operates on the application layer of the TCP /IP where it establishes a full-duplex connection, for bidirectional transmission.

1. Data transmission is fast, because there are smaller packet headers.
2. Data packets are frequently lost and must be retransmitted, as only TCP has error checking functions.

1. Video streaming.
2. Online real-time-multiplier games.
3. Instant messaging.

1. Create → INSERT
2. Retrieve → SELECT
3. Update → UPDATE
4. Delete → Delete

POST, GET, PUT and DELETE.

1. POST → INSERT
2. GET → SELECT
3. PUT → UPDATE
4. DELETE → DELETE

1. Client makes HTTP request to web server.
2. Web server responds with HTML web page.
3. Browser displays web page.
4. JavaScript calls API, which uses the REST methodology to make a HTTP request to the database server using a HTTP verb.
5. Database server responds with XML or JSON formatted data.
6. JavaScript processes XML or JSON to display information on web page.

1. More compact.
2. Easier to read - less markup.
3. Easier to create - fewer characters.
4. Dictionary implementation, makes it faster to process.

1. More flexible.
2. More extensible.

Majority of networks processing power is at server computer, while clients are far less powerful.

Each client pc is powerful enough to provide for its owns processing and storage needs.

1. Cheap to add new clients.
2. Easier to add new clients.
3. Greater security possible through centralisation of management.

1. Network hardware is expensive to set up.
2. Cost and expertise of maintenance is drastically lower.
3. Lower volumes of traffic, lower likelihood of collisions.

1. Server pcs may be much more expensive than the combined total of a few thick clients.
2. There is cost to hiring experts to manage, maintain and setup a thin-client network.
3. Higher volumes of traffic, so higher likelihood of collisions.

Harder to maintains as there is no centralisation of updates, security and backup.