A large-scale derived sub-graph extraction algorithm for adaptive cloud

A large-scale derived sub-graph extraction algorithm for adaptive cloud

Guo Xin Dongjian Zhouqingping

In view of the problems of the existing cloud computing platform resource allocation and the traditional derivation of the sub graph mining efficiency is lower, further enhance the resource integration utilization efficiency of cloud computing platform and the mining efficiency of large-scale derivation of the child graph, and put forward an adaptive cloud extraction algorithm of large scale derived sub graph, In order to solve the problem of resource optimal utilization and massive graph mining. Firstly, the concept of cloud computing and the related concepts and problem description are introduced. An adaptive task Dynamic assignment algorithm is designed based on MapReduce parallel processing model sac_ta (self re-use Cloud Dynamic allocation), which allocates system resources to achieve the optimal cost consumption according to the computing task, and designs the algorithm of large-scale derived sub-graph mining based on adaptive Cloud SFGFF (Sac_ta, Find_ve, G_F1, FINDPARTFG, FINDALLFG), which is divided into 4 stages of mining, the application of all algorithms to the adaptive cloud can constitute the entire export map mining system, and finally in the artificial simulation data and real environment data are tested, the results show that the adaptive cloud operation is good, the algorithm is effective and feasible, It has higher acceleration ratio and running efficiency, and can effectively meet the demand of large scale and frequent export of sub graph mining.

A large-scale derived sub-graph extraction algorithm for adaptive cloud