一种大规模网络数据缓存方法的改进

An improve data caching method for large scale network

针对传统的网络海量数据缓存方法中容易出现数据丢失和召回错误,数据访问和调度性能差等问题,提出一种大规模网络海量数据缓存方法的改进方法.构建网络缓存空间的数据分布结构模型,进行大规模网络海量数据的信息流模型构建和时间序列分析.采用模糊C均值聚类算法对提取的关联维特征进行聚类处理,实现缓存模型优化.仿真结果表明,采用该方法进行大规模网络海量数据缓存优化设计,有效降低缓存开销,扩展缓存空间,数据的吞吐性和召回性等指标参量优于传统方法.

In view of the problems of traditional method of the storage area network mass data cache method, such as prone to loss of data in the data cache and recall errors, data access and poor schedu- ling performance, an improved method of large-scale network mass data caching method is put for- ward. Web cache space structure model of the distribution of data is constructed as well as a large scale network information flow model of huge amounts of data and time series analysis. Fuzzy C-means clustering algorithm is used to cluster the extracted, correlation dimension characteristics,fufilling cac- hing optimization model. The simulation results show that this method effectively reduces the cache o- verhead, extends the cache space,and indices such as throughput and recall of the data parameter are superior to those of traditional methods.