SDCI NMI Improvement: Development of Kepler CORE -- A Comprehensive, Open, Robust, and Extensible Scientific Workflow Infrastructure

国家科学基金会NSF网络基础设施软件开发（SDCI）计划网络基础设施办公室提案编号：0722079主要研究员：Bertram Ludaescher机构：加州大学戴维斯分校提案标题：SDCI NMI改进：开普勒核心的开发 Kepler CORE的目标是将流行的Kepler科学工作流系统从研究原型改进和升级为全面，开放，可靠和可扩展的科学工作流基础设施。与其他科学工作流程系统不同，开普勒不是由一个团体或机构驱动或拥有，而是来自不同NSF（以及较小程度上的DOE）赞助项目的基层社区努力，分布在许多站点和机构，利用预先存在的强大建模系统（来自加州大学伯克利分校的Ptolemy II）。基层性质导致了非常多样化的开普勒社区：当前的应用领域包括生物多样性和生态信息学，生物信息学（例如ChIP芯片和系统发育分析工作流程），化学信息学，地理信息学，真实的时间陆地和海洋观测工作流程，大规模模拟的数据管理支持（例如粒子物理学），以及其他，例如，数字图书馆和档案应用。作为迈向社区可持续发展和继续广泛采用的一步，开普勒核心将重新设计和增强当前的开普勒系统，主要目标是独立的可扩展性，即，由不直接与开发重构后的开普勒软件核心的团队合作的团队组成。Kepler CORE将收集利益相关者（对支持Kepler感兴趣的团体）的要求，以确定重新设计和开发活动并确定其优先级。修改后的架构的主要目标是Kepler的接口，组件和打包机制，这将使系统在现有或新领域的采用和扩展变得更容易，同时进一步提高当前系统的模块化和功能。

National Science FoundationNSF Software Development for Cyberinfrastructure (SDCI) ProgramOffice of CyberinfrastructureProposal Number: 0722079Principal Investigator: Bertram LudaescherInstitution: University of California-DavisProposal Title: SDCI NMI Improvement: Development of Kepler CORE A Comprehensive, Open, Robust, and Extensible Scientific Workflow Infrastructure Abstract The goal of Kepler CORE is to improve and upgrade" the popular Kepler scientific workflow system from a research prototype to a Comprehensive, Open, Reliable, and Extensible scientific workflow infrastructure. Unlike other scientific workflow systems, Kepler is not driven or owned by one group or institution, but grew out of grass roots community efforts across different NSF (and to a lesser extent DOE) sponsored projects, spread across many sites and institutions, leveraging a pre existing powerful modeling system (Ptolemy II from UC Berkeley). The grass roots nature has led to a very diverse Kepler community: current application areas include biodiversity and ecoinformatics, bioinformatics (e.g. ChIP chip and phylogenetic analysis workflows), cheminformatics, geoinformatics, real time terrestrial and ocean observatory workflows, data management support for large scale simulations (e.g. particle physics), and others, e.g., digital library and archival applications.As a step towards communitybased sustainability and to continue broad adoption, Kepler CORE will reengineer and enhance the current Kepler system, with the primary goal of independent extensibility, i.e., by groups which are not directly collaborating with the team that develops the refactored Kepler software core. Kepler CORE will gather requirements from stakeholders (groups interested in supporting Kepler) in order to identify and prioritize the reengineering and development activities. A primary goal of the revised architecture are interfaces, components, and packaging mechanisms for Kepler that will make adoption and extension of the system in existing or new areas easier, while further improving the modularity and functionality of the current system.