中文摘要：

针对服务起源数据存储在关系型数据库和文档型数据库中无法提供高效的服务追踪以及采用图形数据库存储无法进行快速聚合运算等问题,提出了一种基于图的服务起源追踪机制。该机制以图形数据库存储服务起源数据为基础,重新定义服务起源在图形数据库中的存储结构,并提供基于此存储结构的聚合操作。然后提出了基于静态权值、基于混合操作与基于实时任务三种不同的服务起源追踪模型。实验结果表明该服务起源追踪机制能够满足聚合、追踪等不同类型的服务起源数据的查询需求,减少了服务追踪耗时,提高了服务起源数据的追踪效率。

英文摘要：

Service provenance data stored in relational database and document database cannot provide effective service tracking operations and graphic database storage cannot execute rapid aggregation operations. In order to solve the problems,a new service provenance tracking mechanism based on graph was proposed. On the basis of graphic database storage service provenance tracking mechanism,the service provenance storage structure in the graphic database was defined,and the aggregation operation for this storage structure was provided. Then the three different service provenance tracking models which were separately based on static weight,mixed operation and real-time task. The experimental results show that the proposed service provenance tracking mechanism can meet different query requirements of different types of service provenance data such as aggregation and tracking operation,reduces the service tracking time-consuming and improves the tracking efficiency of service provenance.