Bipartite graph sampling method and device for test bed topology structure

Abstract

The invention is applied to a bipartite graph topology structure widely existing in a computer network, and provides a bipartite graph sampling method and device for a test bed topology structure. Byadopting the bipartite graph sampling method, the simulation resource node number needed by test bed construction is decreased in a large scale on the premise that the topology properties are kept stable and invariable, the test bed construction difficult problems under the limited simulation resource condition are solved, and therefore the economic benefits that the test bed construction cost isreduced significantly, and the test task operation time complexity is improved are achieved; and the scale of the test bed bipartite graph topology structure in the computer network can be reduced ina large scale on the premise that the important topology properties such as the node degree distribution, the clustering coefficient and the path length are kept stable, that is, approximately invariable, and the scale reduction rate can reach 96% or above.

Description

technical field [0001] The invention relates to the field of network topology, in particular to a bipartite graph sampling method and device for a test bed topology. Background technique [0002] The development, testing, and evaluation of computer network technologies such as routing protocols and resource positioning require the construction of corresponding simulation test beds. Among them, the scale reduction of the test bed topology is an important guarantee to reduce the cost of test bed construction and improve the efficiency of test running time. In the diverse topological connections of computer networks, the bipartite graph topology is a widely existing connection mode. When it is necessary to test and evaluate the rationality of the topology connection relationship between subnets and subnets in a multi-subnet structure, a bipartite graph topology between subnets needs to be configured on the simulated computer network test bed. Graph sampling is a common method...

AI Technical Summary

Problems solved by technology

However, existing graph sampling techniques (for example, forest fire, random walk, breadth first search, and snowball) have a wide range of applications and can be applied to diverse graph structures such as social networks and transportation networks. The perception of the unique attributes of the graph topology cannot achieve a large scale reduction of more than 70% of the bipartite graph topology under the condition of maintaining the stability of the attributes

Images