A method to distribute loss and loss distribution server

Abstract

The invention relates to a consumption distribution method and consumption distribution server. Wherein, said method via the consult between network sections of flow local channel, calculates out the total minimum consumption of flow channel; based on total consumption target, total minimum consumption and the minimum consumption of each network section, each network section can calculate out available range of consumption, and select reasonable value. The inventive server comprises a consumption target receiving module, a flow channel fixing module, a total minimum consumption fixing module, a maximum consumption fixing module and a consumption selecting module. The invention can be used to distribute consumptions for operator network section and internal sub network sections.

Description

technical field [0001] The invention belongs to the technical field of quality of service assurance (QoS-Quality of Service) in network communication, and in particular relates to a loss distribution method and a loss distribution server among network segments on an end-to-end flow path. Background technique [0002] In the next generation network (NGN-Next Generation Network), end-to-end service quality assurance is a very important link. Applications such as voice, data, fax, and video require high-quality end-to-end performance guarantees. If the end-to-end paths cross different operator network segments, how to ensure the end-to-end performance is a very challenging problem. That is to say, for the total loss target of an end-to-end service, how to allocate the total loss target to each operator network segment on its path, so that each operator network segment can achieve its sub-loss target, so as to achieve the total loss target Loss target to ensure end-to-end serv...

AI Technical Summary

Problems solved by technology

[0008] There are two disadvantages of the static allocation method: 1) The performance of the allocation may exceed the actual required performance

In the above static divisor method, if the estimation of the number of operators in the convection path is inaccurate, the maximum number of operators used in the calculation formula is 4 instead of the actual 3, that is, when the parameter of the static divisor method is larger than the actual value, the transmission is calculated. Take delay as an example, the calculated delay budget will be smaller than the actual required transmission delay, so the transmission performance of the stream will exceed the actual required performance

2) Unable to adapt to changing network

This signaling negotiation allocation method only considers the situation that the operator's network cannot meet the loss target, but does not consider that if a resource-constrained operator is assigned a resource-intensive loss budget for the network segment, the operator happens to It can be satisfied, but the end-to-end service may not need such a strict loss budget, which will cause a waste of network resources

Moreover, the initial distribution of loss is based on a static distribution algorithm, and the consideration of operator differences is also extensive.

Images