Earlier in the IPv6 series we covered an overview of the protocol and a look at the advanced addressing capabilities of the next generation IP. In this article we look at the routing and security provided by the protocol.
Routing
IPv6has the same routing as with IPv4 with the only difference being the 128-bit address versus the 32-bit address found in IPv4. Using extensions IPv4`s routing algorithms may be used to route IPv6.
IPv6 has extended capabilities to support enhanced functionality such as the provider selection, which is based on cost, performance and policy. Plus mobility that will allow for the re-route to a new location and auto-readdressing, which simply routes to a new specified address.
The added functionality is achieved by means of creating a string of IPv6 addresses using the routing option. This option is used by the source to indicate the immediate nodes or groups that will be used to pass the packet on to its final destination. The function is similar to that of the loose source used in the IPv4.
An interesting enhancement to the IPv6 routing is that the IPv6 hosts need to install reverse routes in a packet to make the address sequencing a general function. This makes it possible to return the packet to the source. The reason for this enhancement is that it will enable the implementation of new features to operate with IPv6. Three examples of address sequences may be used with the first and most common being the source address with intermediate addresses and lastly the destination address. In the intermediate address the provider can be stipulated. Which will ensure that the return route will follow the route of the provider.
Quality capabilities
IPv6 header may also take advantage of flow label and priority headers that will single out those packets for special handling by the routers to deliver `real time` service. Ideally 6thse packets would be put to use in such demanding applications as multi-media or video conferencing. The issue at play here is that the packets need some sort of consistent delivery to ensure the quality of service.
A flow is regarded as a sequence of packets originating from a particular source with a particular destination. The source requires the packets to gain some special handling by the routers. The special handling will be specified by the use of a control protocol, which will include a resource reservation instruction, or a requested flow of the packets.
A flow is assigned by the source; the ensuing flow labels will have to follow a random choice with a uniform range from 1 to FFFFFF hex. The reason for this is that the random yet uniform assignment is to allow for the bits within the label to be able to use a hash key by the routers.
The packets that belong to the same flow must have the same source address with the same flow label and destination address. If any of the packets include either a routing header or a hop-by-hop header then they must all include the same header and originating header. If the conditions are violated and the conditions are not met a message will be sent to the source identifying the error in the flow label.
Router in turn is able to take advantage of the flow states and set-up their own if the source has not provided for. The router will process the header and extensions as 0 and include the next hop interface. The processing could include such aspects as the queue or priority the packets would take. The router would then cache the data and with subsequent packets handled in much the same way.
Flow labels are 24-bit IPv6 headers utilised by the source for special handling by the routers. The flow labels are still in development and will likely be so for some time as the protocol evolves. Host and routers hat do not support the flow label will set the field to 0, in the case of forwarding the packet will be passed on unchanged and on receipt ignore the field.
The priority of the packet in IPv6 is set by means of a 4-bit field in the header, which identifies the delivery priority in respect of other packets emanating from the source. The priorities can be set to accommodate traffic control to assist in alleviating congestion on the network and the other is to set according to the `real time` data needed as in the case of multimedia. Values set between 0 and 7 will allow the packets to queue in respect of traffic congestion and values of between 8 and 15will specific the priority traffic for the multi-media applications.
In respect of the non congestion controlled traffic the lowest setting of 8 should be assigned to those packets which the source is able or willing to discard, the highest level of 15 should be those packets that the source would least like to see discarded.
Security
Perhaps the most pressing issue of the Internet as we enter the age of e-commerce is that of security. The IPv4 protocol has already indicated that there are a number of flaws with little in the way of privacy and authentication under the application layer.
The two options of IPv6 security mechanisms of authentication and encapsulating security enable a solution that can be tailored to meet various users` needs. The authentication header is an extension header that delivers authentication to IPv6 traffic. The authentication is algorithm independent and will operate using different techniques. The keyed MD 5 is designed to be interoperable on the Internet and reduce the possibility of network attacks, such as host masquerading.
The encapsulating security header is the mechanism that provides both confidentiality and integrity to the packet. The header is algorithm independent yet flexible enough to ensure interoperability with the rest of the Internet.
The Internet will continue to grow beyond these capabilities but for the present and most immediate future the work in progress will go along way to making the Internet more pleasurable and pervasive medium across the world.
Share
