Networking Basics

• IP Address: is a numeric assigned to each device connected to a computer network to identify the device.

  • Two flavors of IpAddressing
    • IPv4: 32 bit number
    • IPv6: 128 bit number

• IPv4 addresses:

  • has the size of 32 bits which theoritically allows to connect to 2^32 addresses. We are not allowed to use all the addresses as some of them reserved
  • An IPv4 address is represented as x.x.x.x . Each x is 8 bit number so x can be any where between 0 to 2^8-1 (255). Range of ip address (0.0.0.0 to 255.255.255.255)
  • Launch cmd and execute ipconfig
  • IP address uniquely identifies your device. In the above case ip address was 192.168.0.106. But why do i require subnet mask and Default gateway?

• Basic Networking Principles

  • A device in a network can communicate with other devices in the same network

• When we look at ip address it is representation of

  • network identifier:
    • id of your network
    • All the devices with same id can communicate with each other
  • host identifier:
    • device id on your network

• But how would i know what is network id and what is host id by looking at ipaddress?

  • It is not possible to know network id from a ip address without knowing what subnet mask.

  ip addresss: 192.168.0.106
  subnet mask: 255.255.255.0
  network id: 192.168.0.0
  host id: 106
  
  ip address: 10.10.12.14
  subnet mask: 255.255.0.0
  network id: 10.10.0.0
  host id: 12.14
  

• The host id size will reveal size of network i.e. how many devices you can connect on the network. Size of network => 2^n – 2 (n = size of host id in bits)

  ip addresss: 192.168.0.106
  subnet mask: 255.255.255.0
  network id: 192.168.0.0
  host id: 106
  size (hostid): 8 bits : 2^8 -1-1 : 254
  
  ip address: 10.10.12.14
  subnet mask: 255.255.0.0
  network id: 10.10.0.0
  host id: 12.14
  size (hostid): 16 bits : 2^16-2 : 65534
  

• Older approach: Use classful addressing Refer Here

  • Possible subnet masks which i can have in this approach is
    • Class A => 255.0.0.0 => number of address => 2^24-2 ~= 16,777,216
    • Class B => 255.255.0.0 => number of address => 2^16-2 ~= 65000
    • Class C => 255.255.255.0 => number of address => 2^8-2 ~= 250

• Scenarios:

  • If i need a network which has to connect to 200 devices. Based on classful networking i need to use class C network
  • If i need a network which has to connect to 400 devices. Use class B. But class B is of size 65000 which is way beyond 400 (so many address will be wasted)

• A new ip addressing scheme called as CIDR (Classless inter domain routing) convention came into play

  • Now convert subnet mask into binary

  network: 192.168.0.0 
  subnet mask: 11111111.11111111.11111111.00000000
              255.255.255.0
  cidr notation => 192.168.0.0/24
  
  how many zeros => 8 (size of network)
  till what position do you consecutive 1 => 24
  size of network => 2^8-2 = 254
  
  network: 192.168.0.0 
  subnet mask: 11111111.11111111.11111110.00000000
              255.255.254.0
  cidr notation => 192.168.0.0/23
  how many 1s => 23
  how many 0 => 9
  size of network => 2^9-2 => 510
  
  
  

  • Create a classless network for 1000 devices. Refer Here from this table find smallest number that is approximately thousand

  value=> 2^10 =1024
  so our value = 10
  10 for host id 
  subnet mask: 11111111.11111111.11111100.00000000
          : 255.255.252.0