复杂网络松耦合分布式计算框架的设计与实现

发布时间：2018-07-26 19:28

【摘要】：为更快地计算大尺度复杂网络结构的相关参数,设计并实现一种松耦合分布式计算框架。将分散于网络中的松耦合计算节点汇集起来,通过任务队列使各计算节点共同参与复杂网络的相关分布式计算,并能随时加入或者退出计算,利用分散于网络中松耦合的计算节点提高复杂网络相关分析的计算速度。基于该框架,实现对大尺度复杂网络的平均最短路径长度、网络直径和网络效率的分布式计算。实验结果表明,在保证计算结果正确的前提下,该框架可充分利用网络中闲散的计算资源,提高运算效率。
[Abstract]:A loosely coupled distributed computing framework is designed and implemented in order to calculate the parameters of large-scale complex network structure more quickly. The loosely coupled computing nodes scattered in the network are gathered together, and each computing node can participate in the related distributed computing of the complex network through the task queue, and can join or withdraw from the computation at any time. The computing speed of correlation analysis of complex networks is improved by using loosely coupled computing nodes dispersed in the network. Based on this framework, the distributed computation of the average shortest path length, network diameter and network efficiency for large scale complex networks is realized. The experimental results show that the framework can make full use of the idle computing resources in the network and improve the computational efficiency on the premise that the calculation results are correct.
【作者单位】：北京化工大学信息科学与技术学院;
【基金】：北京高等学校青年英才计划基金资助项目(YETP0506)
【分类号】：TP338.8;O157.5

【参考文献】