directory

  • An overview of the
  • The physical layer
  • Data link layer
  • Network layer (1) Basic knowledge
  • Network layer (2) IP address design and exhaustion
  • Network layer (3) Network layer packet transmission process
  • Network layer (4) Network layer multicast and other applications
  • Transport layer (1) Reliable transport and TCP/UDP
  • Transport layer (2) congestion control and handshake waving
  • Application layer (1) HTTP and the World Wide Web

The IP address design of the classification is incorrect

  1. There is A problem of low utilization rate. The number of class A address hosts is more than 10 million, and the number of class B hosts is 60,000. Some units apply for class C address, but the number of hosts itself is not very many, but in order to consider the future development, it leads to waste.
  2. Each physical network has a network number, which causes the router to have too many routing table items, increasing the storage cost and forwarding efficiency
  3. Not flexible enough if an organization now needs a new network, but applies for a new IP address first.
  4. The entire IPV4 address space will soon be allocated.

Historically, the following improvements have been made to address the above problems

Divide the subnet

Add a subnet number, that is, borrow some bits from the host number as the subnet number, and the number of hosts will be reduced accordingly. Subnets are internal affairs of the unit that owns the network number, and from the outside, the unified network. When the router reaches this network number, it will forward the router to the corresponding subnet according to the subnet number. So how do we mark this subnet? We have to use the subnet mask, so we can’t tell the subnet number from the IP address. Its job is to find its network number and subnet number, and filter out the rest of the host number. Set the bits of the network number and subnet number to 1 and the host number to 0 using the & operation. Specify that all networks use subnet masks. If there is no subnet, use the default subnet mask (the network number is 1 and the host number is 0). Subnets increase flexibility, but reduce the number of hosts on the network.

Unclassified address

As you can see from this name, this address does not use the classification, no ABCD address constraints. As you can imagine, this adds a lot of network numbers, because the original address of the network number number is fixed. This makes assigning IP addresses more efficient. In this way, the slash notation is used, that is, the/network number digit is added after the IP address. It’s also more flexible, so you don’t need to assign 8/16/24.

Build super net

Indicates an aggregation of addresses. For example, merge multiple Class C addresses. Then these class C addresses are one item in the routing table, reducing the number of items in the routing table. In addition, the number of network can also be allocated according to the actual situation, which requires several CLASS C addresses to reduce the network waste.

IPV6 address

Since the IPV4 address space is running out (and is running out), the solution to this problem, of course, is to use more multi-digit IP addresses, known as IPV6. The difference

  1. The “larger address space” was increased four times to 128 bits, increasing the address space by 2 ^ 96.
  2. The “more flexible head” improves the router’s processing efficiency because it can be processed less efficiently. The length of the front is fixed.
  3. “Allow protocol expansion” IPV4’s functionality is fixed
  4. “Improved options” IPV4 options are fixed and placed in the header variable section, v6 in the payload.
  5. “Plug and play, no NEED to use DHCP“

The V6 datagram consists of two parts, the base header and the payload behind it. The payload consists of multiple extension headers and subsequent data sections. The length field of the header is removed because the header is fixed. The number of fields in the header has changed to 8: version, traffic class, stream label, payload length, next header, hop limit, wish/destination address.

If V4 uses the header option, each router along the way will have to check these options one by one, which reduces router transmission efficiency. V6 puts the original header option in the payload extension header and sets the header length to a fixed value. Give the options at the head to the origin and the destination. Greatly improve the efficiency of the router.

V6 uses hexadecimal notation, using: split. Zero compression is allowed, and the same 0 can be omitted.

How are IPV4 and V6 compatible

Since it is impossible for all addresses to use V6 addresses at once, there are two ways to solve this problem

Double protocol stack

Equipped with both stacks, the machine can communicate with BOTH V4 and V6. But it’s inevitable that some data will be lost in the process.

The tunnel technology

V6 datagrams in V4 datagrams in V4 datagrams in V6 datagrams in V4 datagrams in V4 datagrams