System and method for implementing self-governing QoS based on service network differentiation and IPv6 spreading head

Abstract

The invention relates to a system based on a differentiated service network and an IPv6 extension head for realizing autonomous QoS (quality of service) and a method thereof. The creative feature of the method lies in: a service terminal is provided with a service sensing component for sensing the service type and flow feature of a service data message to be transmitted; and the service sensing component takes the IPv6 QoS extension head to bear a sensed service flow parameter, adds the IPv6 QoS extension head in the service data message and transmits the service data message to a network. An ingress edge node is provided with an IPv6 QoS extension head detector for fast checking the service data message accessing the network and judging whether the service data message bears the IPv6 QoS extension head, and a service QoS self-configuration component for extracting service information born in the IPv6 header of service data message and the IPV6 QoS extension head and updating a service data table adaptively according to extracted information. A classifier and a marker of the edge node respectively conduct classification, marking and parameter configuration of a flow controlling mechanism to the service data message according to the service information table, thereby providing support to QoS guarantee and realizing autonomous QoS.

Description

technical field [0001] The present invention relates to a system and method for implementing autonomy in quality of service (QoS, Quality of Service) for an IP network. Specifically, it relates to a differentiated service network based on a service terminal that perceives service types and service flow characteristics , and then use a newly designed IPv6QoS extension header to carry the perceived service flow parameter information, and then the edge nodes of the DiffServ network analyze and process the service data packets carrying the IPv6QoS extension header, and according to the obtained service flow The parameter information is used to configure the service flow control mechanism on the node, so as to provide support for QoS guarantee and realize the network system and method of autonomous QoS. It belongs to the technical field of the Internet. Background technique [0002] In the existing IP network, the use of network resources is very unbalanced due to the increase i...

AI Technical Summary

Problems solved by technology

However, the IntServ model also has many shortcomings: First, the scalability is not good, it requires all routers on the service path must support the RSVP protocol, otherwise end-to-end resource reservation cannot be completed; second, each router needs to maintain a large number of state information, consumes a lot of memory, and the complexity of flow-based operations will increase as the number of flows increases; moreover, the IntServ model is not suitable for short-lived packet flows, because the reserved resources of short-lived packet flows The overhead may be greater than the overhead of processing all packets in the stream

[0013] The disadvantages of the DiffServ model are: it cannot solve the end-to-end QoS problem, only promises relative service quality, and cannot provide users with absolute service quality guarantee; it needs the cooperation of a large number of network elements to provide end-to-end service quality; the aggregated DiffServ The service flow cannot guarantee the full realization of service fairness between service micro-flows in the converged flow; in addition, due to the lack of detailed standards, there are still problems in the intercommunication between DiffServ networks of different operators

In addition to being executed on edge nodes, business perception can also be used in conjunction with business management systems, which can better implement overall policy deployment. However, this method still has unavoidable shortcomings in centralized management.

Images