Selecting Unrepeatable Random Starting Nodes in Large-Scale Networks for Parallel Computing and Searching

Abstract

With the prosperity and development of information technology, the scale of extractable networks becomes larger and larger. The current sizes of the networks often turn up to millions, tens of millions, or even hundreds of millions of orders of magnitude. The analysis and mining of large-scale network is bound to cost a lot of time. Therefore, the research of parallel computing and searching on large-scale networks is an urgent task. Parallel computing on large-scale networks often requires the identification of unrepeatable random nodes, but this research is very rare. Aiming at this problem, three algorithms for selecting unrepeatable random starting nodes are proposed based on the existing algorithm. In order to carry on comparison and analysis, firstly designs the corresponding experiment algorithms, and then carries on the experiments, finally carries on the analysis to the experiment results. The results show that two algorithms are more suitable for the selection of unrepeatable random starting nodes in large-scale networks. This paper will be helpful for parallel computing and searching on large-scale networks.