What is Internet Protocol?
I want Internet Part 2
IP is the connection between the applications we use and the network that connects us together (Wireless networks, ATM/Frame Relay, Ethernet LAN, 3G Wireless, WIFI).
IP networks are
- Adaptive and Self Learning
- IP protocol is designed for “Best Effort” delivery
IP networks are deployed everywhere on earth. Components of IP networks such as routers, DNS servers, and others learn network information and keep up with changes on the network. Since IP networks connect both public and private networks, IP is designed to be a “Best Effort” protocol and does not guarantee quality of service.
The IP Address
With each datagram, IP attaches source and destination information. Each address is 32 bits long in IPv4. An example would be 126.96.36.199
A company own the IP address 128.16.x.x and the company can divide its IP address by location which creates subnetworks 128.16.1, 128.16.2, and 128.16.3.
Types of IP Addresses
Public: Each device on the internet has a globally unique IP address. Used for communicating with external networks.
Private: Can be used by any private network, but must be unique within the network.
How a device gets an IP Address
A static address is given to a device and is permanent to that device.
A dynamic IP addresses is given when a device connects to a network and is taken away when the device goes offline. Uses Dynamic Host Configuration Protocol (DHCP) by sending and receiving a broadcast message.
IP Packet Has its Address and Needs to be Sent
The computer/host is assigned an IP address and gets the destination’s IP address. The computer can use the Domain Name System (DNS) to look up the name of a web page and get its IP address. The DNS a distributed network with the top level domains being .com, .edu, and .org, and sub-level domains being google.com, yoyo.edu, and government.org. Since IP networks are adaptive and self-learning, a person can register a new domain on a DNS server and have that domain be registered on all DNS servers in the world.
Now IP packets are forwarded through the internet to the “Last Router” in the destination network. IP packets are sent from a router to a router to a router… until it hits the last router. This is known as hop-by-hop forwarding and only the destination address is used. Each router has a routing table with destination network addresses and a next hop address.
These routing tables need to be updated all the time since new IP addresses are made and new connections are made all the time. Each router learns and updates other routers. When a network is connected to the internet, the router connected to it sends an update message to all of its adjacent routers and those adjacent routers send an update message to all of its adjacent routers. Remember each router only has information about its adjacent connections not about long term connections or new networks. The protocols used here are Routing Information Protocol (RIP), Open Shortest Path First (OSPF) and Border Gateway Protocol (BGP).
When the IP packet goes to the last router, it then delivers it over Ethernet LAN to the web server. Ethernet requires MAC addresses included in the datalink layer header and help identify the device that needs the IP Packet. To find the MAC address, Address Resolution Protocol (ARP) is used which sends a broadcast message over the LAN and receives the MAC address.
What Feeds IP?
I said earlier that applications feed IP. Applications need to break down their data into TCP (Transmission Control Protocol) or UDP (User Datagram Protocol). As data gets broken down into packets, TCP keeps track of if data is lost while being sent to IP and can control the flow of data to IP. UDP is like TCP but usually only used for Video over IP and Voice Over IP.
Where does IP go?
How does TCP know what type of application (FTP, WWW) it needs to go to? Each application is assigned a port number. Email is port 25 while WWW is port 80. The port number is included in the IP packet sent from the source.