IPv6 Header structure and comparison with IPv4

Although IPv6 headers have two times the length of 40 bytes and 20 bytes), IPv6 has a simplified header structure. The IPv6 Header has eight fields, while IPv4 has 14 fields. IPv6 discards and uses several IPv4 header fields to make packet processing more efficient.
The following two headers are compared. The Version field does not change in the two Protocols. IPv6 discards IPv4 Internet Header Length (Internet Header Length), Type of Service Type), Identification recognition), Flags mark), Fragment Offset) and Header Checksum) field. Total Lenght Total length), Time to Live survival Time) and Protocol) The field has a new name in IPv6, and the function is slightly redefined. The Option in IPv4 has been removed from the header and changed to the Extension function. Finally, IPv6 adds two new fields: Traffic Class Traffic type) and Flow Label Flow tag ). This section describes each header field of an IPv6 package.
Version: the length of the Version field is still 4 bits, indicating the Protocol Version number.
Traffic Class: This 8-bit field can assign different classes or priorities to the package. It is similar to the Type of Service field of IPv4, leaving room for differentiated services.
Flow Label: The Flow Label field is a new field of IPv6. The source node uses this 20-bit field, which is better than the best-effort forwarding for packet requests in a specific sequence ). For real-time data transmission such as voice and video, you can use the Flow Label field to ensure QoS.
Payload Length: This 16-bit field indicates the Payload Length. Unlike the Total Length field in an IPv4 packet, the value of this field is not counted as the IPv6 40-bit header. Only the extensions behind the header and the length of the Data Section are calculated. Because this field is 16 bits long, it can represent a data payload of up to 64 KB. If the payload is larger, it is represented by the jumbogram Extension Section of the ultra-large package.
Next Header: This 8-bit field is similar to the Protocol field in IPv4, but there are some differences. In an IPv4 packet, the transport layer header, such as TCP or UDP, always follows the IP header. In IPv6, the extension part can be inserted in the IP header and transport layer header. Such extensions include verification, encryption, and sharding. The Next Header field indicates whether the transport layer Header or the extended part is behind the IPv6 Header.
Hop Limit: This 8-bit replaces the TTL field in IPv4. After a specified number of Route segments, the packet is discarded to prevent packet forwarding. When a package passes through a vro, the value of the Hop Limit field is reduced by one. IPv4 uses the time value). Each time a route segment passes through, the TTL field is subtracted from one second. IPv6 uses the segment value hop value) to change the value.
Source Address: this field specifies the starting Address of the origin host. The length is 128 bits.
Destination Address: this field specifies the Destination Address of the transmission signal. The length is 128 bits.
Network personnel may be surprised to find that the checksum and fragment fields have disappeared from the IPv6 Header. The header checksum of discarded packets is used to improve the routing efficiency. Although there may still be errors in the packet header, the designers of the new protocol think this risk is acceptable, especially considering the upper and lower layers of the IP layer: the data link layer and the transport layer will check for errors.
As for sharding, IPv6 does allow packet segmentation, but this process is performed in the header extension part rather than the header itself. In addition, the IPv6 package can only be split by the source node and re-assembled by the target node: the router is not allowed to intervene to split or re-assemble the package. The purpose of this sharding feature is to reduce the processing overhead in transmission. Assuming that the size of the network frame is large enough today, most packets do not need to be sharded.
If you have to split the IPv6 package, the source node determines the maximum transmission unit MTU of each link ). One way is to send a test package to the target address. If the test package is too large for a link, the link returns an Internet Control Message Protocol (ICMP) message to the source node, and the source node reduces the packet size accordingly.
The extension mechanism for implementing the sharding and other options is an important feature of IPv6 re-design. It replaces the Options field of IPv4, which enhances the security function of IPv4 packets and enriches the source route selection.
Instead of adding this enhancement feature to the IPv6 Header, the designer designs an extension that can be inserted between the IP header and the higher-level protocol header. This makes the unscalable package process faster and provides a series of extensible options, such as encryption, verification, sharding, source routing, segment, and target options. As mentioned above, these extensions are calculated in the total payload length of the package.