CBSD:一种基于Chord的模糊服务发现方法

CBSD:A Fuzzy Service Discovery Algorithm Based on Chord

针对基于DHT的结构化服务发现方法不支持模糊查找的问题,采用服务聚类技术与结构化服务发现技术相结合的方式,提出了一种基于Bloom filter聚类优化的结构化Web服务发现方法。该方法利用Bloom filter实现服务语义映射并通过服务训练队列实现服务描述聚类特征向量的提取,利用相关性计算实现服务描述的预分类,利用Chord算法实现服务的发布/发现,无需冗余发布,既可保证服务语义相近的服务发布到相同的节点上,又可有效地支持服务的模糊查找,并在此基础之上提出了一种基于Bloom filter的分布式服务组合算法。最后,通过仿真验证了所提方法的可行性。

The main drawback of the structured service discovery method based on DHT doesn＇t support fuzzy search in the distributed computing environment. A Bloom filter based structured service discovery method was raised that combines the service clustering and structured service discovery technology. This method uses Bloom filter to represent the service semantics. The clustering feature vectors are got by service training queue. Before the services are published into the chord ring, they are clustered by the relevance among the feature vectors. Without redundancy advertisement, this method can guarantee that the services with similar semantics can be published to the same node and can support fuzzy service discovery. Based on this method, a distributed service composed algorithm was raised. At last, the feasibility of the proposed method was demonstrated by simulation.