IPv6 and ICMPv6

1 IPv6

The upgrade of IPV4 was first defined in two R F C. RFC 1883 describes the protocol itself, while R F C 1 8 8 4 introduces the IPV6 address structure. RFC 1884 has now been superseded by RFC 2373, and the summer of 1998 I E T F approved a draft to replace RFC 1883. Changes from 3 2-bit addresses to 1 2 8-bit addresses represent a significant shift.

Overview of 5.1.1 Changes

The changes in IPV6 are reflected in the following five important areas:

Extended addresses.

simplifies header formatting.

Enhanced support for extensions and options.

Stream remember.

Authentication and confidentiality.

These changes to IP have improved the vast majority of the IPv6 development direction developed by I a B in 1991. IPV6 's extended address means that IP can continue to grow without considering the scarcity of resources, which is helpful to improve routing efficiency; the simplification of the header reduces the processing required on the router, thus improving the efficiency of the route selection; Improved support for the right scaling and options means that you can adapt to more specific requirements with little impact on common packets and special packet paths; stream tagging provides a mechanism for handling packet flow more efficiently, which is especially useful for real-time applications , authentication and confidentiality improvements make IPV6 more applicable to commercial applications that require special treatment of sensitive information and resources.

1. Extended Address

In addition to extending the 3 2-bit address space to 1 2 8 bits, the IPV6 address structure also makes some adjustments to the different types of addresses that IP hosts may have. As will be described in detail in chapter 6th, the broadcast address is canceled in IPV6 and replaced by an anycast address. A unicast address in IPV4 that specifies a network interface and a multicast address that specifies that one or more hosts listen for a basic invariant.

2. Simplified Baotou

The IPV6 includes 8 fields with a total length of 4 0 bytes (of which two are source addresses and destination sites). It differs from the IPV4 header in that the IPv4 contains at least 1 2 different fields with a length of 2 0 bytes when there is no option, but up to 6 0 bytes when the option is included. IPV6 uses a fixed-format header and reduces the number of fields that need to be checked and processed, which makes it more efficient to route.

The simplification of the header makes some changes in the way IP works. On the one hand, all headers are of uniform length, so the header length field is no longer required. In addition, you can remove some fields from the header by modifying the rules for packet fragmentation. A fragment in IPV6 can only be performed by the source node: the intermediate routers that the packet passes through cannot be further segmented. Finally, removing the IP header checksum does not affect reliability, mainly because the header checksum will be responsible for higher level protocols (U D p and t C P).

3. Improvements to extension and option support

In IPv4, you can add options to the end of the IP header, and in contrast, IPv6 the options in a separate extension header. In this way, the option headers need to be checked and processed only when necessary. The following and 7th chapters will have more to discuss.

For illustration purposes, consider the following two different types of extensions: Segmented headers and route headers. The segments in IPV6 only occur on the source node, so the node that needs to consider the segmented extension header has only the source node and the destination node. The source node is responsible for staging and creating the extension header, which will be placed between the IPV6 header and the next high-level protocol header. The destination node receives the package and uses the extension headers to reload it. All intermediate nodes can safely ignore the segmented extension header, which increases the efficiency of the package route.

In another option, hop-by (H o p-b y-h o p) option extension headers require each node on the path of the package to process the header field. In this case, each router must also handle the hop-by option while processing the IPV6 header. The first hop-by option is defined for an extra long IP packet (mega net). Packages that contain huge net loads require special treatment because not all links are capable of handling such long transmission units, and routers want to avoid sending them to a network that cannot be processed. Therefore, this requires that the options be checked on each node that the package passes through.