科学工作流溯源表示和查询技术综述.pdfVIP

科研信息化技术与应用 2 0 1 5 , 6 ( 6 ) : 1 8 – 3 2 TECHNOLOGY ·技术 / · 科学工作流溯源表示和查询技术综述 1, 2 1 1 林晨 ，罗万明，阎保平 1. 中国科学院计算机网络信息中心，北京 100190 2. 中国科学院大学，北京 100049 摘 要：溯源管理是科学工作流系统的核心功能之一。科学工作流语境下的溯源，可分为工作流定义溯源 和工作流执行溯源，分别描述工作流定义和执行阶段的元数据、过程依赖及数据演化。本文重点 关注工作流定义溯源和执行溯源的表示及查询技术，并阐释针对科学工作流领域内独有问题，如 “黑盒”问题、依赖区分问题以及细粒度溯源等问题的解决方案。文中还将介绍现存的一些面向 科学工作流的溯源系统，并提出对溯源技术未来的展望。 关键词： 溯源技术；科学工作流； “黑盒”问题；细粒度溯源 doi: 10.11871/j.issn.1674-9480.2015.06.003 A Survey of Modeling and Querying Techniques for Scientific Workflow Provenance 1, 2 1 1 Lin Chen , Luo Wanming , Yan Baoping 1. Computer Network Information Center, Chinese Academy of Sciences, Beijing 100190, China 2. University of Chinese Academy of Sciences, Beijing 100049, China Abstract: Provenance management is one of the core functionalities of scientific workflow systems. Provenance in scientific workflow can be categorized as workflow definition provenance and workflow execution provenance, which describe the meta-data, module dependencies and data evolution during workflow definition and execution phases respectively. This paper focuses on the modeling and querying techniques of scientific workflow provenance, and explains the existing methods on solving those problems unique to scientific workflow such as: “black-box” problem, dependency differentiation problem and fine-granularity provenance. Finally, this paper lists some of the existing scientific workflow-oriented provenance systems, and shows our prospect of the future of scientific workflow provenance. Keywords: provenance; scientific workflow; “Black-Box” problem; fine-granularity pr