Method, device and system for obtaining transmission path in network

Abstract

The invention discloses a method, device and system for obtaining a transmission path in a network. The method comprises: collecting state data of n terminals contained in the transmission network, wherein the state data at least comprise quality information of every terminal and data transmission weight between any two terminals; dividing the n terminals into multiple terminal sets according to a preset rule, wherein the n is a natural number; calculating the shortest paths from any one terminal node contained in every terminal set to other terminal nodes in parallel; determining the shortest path between any two terminals in the transmission network according to the shortest paths corresponding to every terminal set. According to the invention, the technical problem in the prior art that the cycle of obtaining the shortest path is long under a condition of network fluctuation or path failure when centrally calculating the paths in a single machine through a dynamic accelerating technique is solved.

Description

technical field [0001] The present invention relates to the field of Internet data transmission, in particular to a method, device and system for acquiring a transmission path in a network. Background technique [0002] In the field of Internet data transmission, a relatively common business scenario is the data transmission acceleration service, which can effectively reduce the response delay for users to access remote services. At the same time, in the domestic Internet environment, it can solve the problem of cross-operator access. Effectively improve the end-to-end connectivity. Among the existing acceleration technologies, there are mainly Tencent Interactive Entertainment TGP acceleration system and Tencent CDN acceleration system. [0003] Tencent Interactive Entertainment TGP acceleration system adopts static acceleration technology. It is an acceleration system based on pre-established nodes, statically configured fixed acceleration paths, and based on node support ...

AI Technical Summary

Problems solved by technology

[0005] 1. It cannot adapt to the network environment, and its fault tolerance is poor in scenarios such as network failure or jitter;

[0006] 2. Unable to provide users with the optimal path based on quality

[0009] 1. The single-machine centralized calculation path leads to poor cluster scalability. After the calculation amount reaches the single-machine bottleneck, the cluster nodes will not be able to expand capacity, and in scenarios such as network fluctuations or path failures, the path update response cycle is longer;

[0010] 2. The path selection dimension is single, and the optimal path is selected only based on the quality dimension. Without considering the path capacity, it is easy to overload the capacity of a single path, resulting in unbalanced global capacity and affecting path stability.

[0011] Aiming at the technical problem of using dynamic acceleration technology to centrally calculate the path in a single machine in the existing technology, resulting in a long period of obtaining the shortest path in the case of network fluctuations or path failures, no effective solution has been proposed yet

Images