EarthCube Building Blocks: Collaborative Proposal: Cloud-Hosted Real-time Data Services for the Geosciences (CHORDS)

EarthCube 构建模块：协作提案：地球科学云托管实时数据服务 (CHORDS)

• 批准号: 1440116
• 负责人: Branko Kerkez
• 金额: $ 4万
• 依托单位: Regents of the University of Michigan - Ann Arbor
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2014
• 资助国家: 美国
• 起止时间: 2014-09-01 至 2016-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1440116&HistoricalAwards=false
• 关键词: EarthCube; Building; Blocks; Collaborative; Proposal

The importance of real-time scientific data is ever increasing, particularly in mission critical scenarios, where informed decisions must be made rapidly. Furthermore, advances in the distribution of real-time data are leading many new transient phenomena in space-time to be observed. Presently however, realtime decision-making is infeasible in many cases that require streaming scientific data to be coupled with complex models. While EarthCube will provide an unprecedented framework for disseminating data sources, the use of real-time data raises an additional set of complex challenges that must be considered. This project is a pilot to demonstrate the importance of coodinating the geosciences around their real-time data gathering and use. The work will demonstrate a Cloud-Hosted Real-time Data Services for the Geosciences (CHORDS)The vision behind CHORDS is to provide a real-time data management infrastructure that will: a)Provide a system to archive, navigate and distribute real-time data streams via the Internet; b)Be easily deployed and configured; c)Run on cloud infrastructure; d)Use transactions built on RESTful protocols (i.e. via URLs); e)Employ data and metadata formats that adhere to standards, which simplify the user experience; f)Be free and open source.Science derived from observing platform data is the result of years of planning before a deployment, for example, and millions of dollars are spent on the deployment itself in hopes of obtaining the desired dataset. Many geo-scientific experiments are often resource constrained. In such cases it is vital to guarantee the optimal allocation of resources to ensure that important events do not go unobserved. In geo-scientific domains it is not uncommon to analyze data after a field campaigns, only to detect sensor faults, calibration offsets or anomalous behaviors when it is already too late. Real-time data will enable the optimal allocation of constrained experimental resources by automating the detection of faults and anomalies. CHORDS will provide a framework to enable adaptive sampling, discovery and fusion of various real-time geoscientific data sources, thus facilitating a new means by which geoscience experiments are carried out. The use cases will illustrate this by showing traditional experiments would have missed events of interest due to lack of access to real-time data. Focus on the initial work of defining requirements, design and specifications. Begin to ingest a small subset of geosciences data streams into a prototype CHORDS structure built in the cloud. Participate in activities that strengthen the integration of real-time data being ingested via CHORDS into other EarthCube Building Block systems that are under development. They will focus on some initial test cases, in hydrology sensor data, radar data streams, the NCAR Lower Atmosphere Observing Facilities, and outreach to earth and oceans communities.

实时科学数据的重要性日益增加，特别是在使命关键情景中，必须迅速做出知情决策。此外，实时数据分布方面的进展正在导致许多新的时空瞬态现象被观测到。然而，目前，实时决策在许多情况下是不可行的，需要流科学数据与复杂的模型相结合。虽然EarthCube将为传播数据源提供一个前所未有的框架，但使用实时数据提出了一系列必须考虑的额外复杂挑战。该项目是一个试点项目，展示了围绕实时数据收集和使用协调地球科学的重要性。这项工作将展示一个用于地球科学的云托管实时数据服务（CHORDS）CHORDS背后的愿景是提供一个实时数据管理基础设施，该基础设施将：a）提供一个通过互联网存档、导航和分发实时数据流的系统; B）易于部署和配置; c）在云基础设施上运行; d）使用基于REST风格协议的事务（即通过URL）; e）采用符合标准的数据和元数据格式，这简化了用户体验; f）是免费和开源的。从观察平台数据中获得的科学是部署前多年规划的结果，例如，数百万美元花费在部署本身上，希望获得所需的数据集。许多地球科学实验往往受到资源的限制。在这种情况下，必须保证资源的最佳分配，以确保重要事件不会被忽视。在地球科学领域，在野外活动后分析数据并不罕见，只是为了检测传感器故障，校准偏移或异常行为，当它已经太晚了。实时数据将通过自动检测故障和异常来优化有限实验资源的分配。CHORDS将提供一个框架，使各种实时地球科学数据源能够进行自适应采样、发现和融合，从而为开展地球科学实验提供一种新的手段。这些用例将通过展示传统实验由于无法访问实时数据而错过感兴趣的事件来说明这一点。专注于定义需求、设计和规范的初始工作。开始将地球科学数据流的一个小子集摄取到云中构建的原型CHORDS结构中。参与活动，加强通过CHORDS摄取的实时数据到其他正在开发的EarthCube积木系统的集成。他们将专注于水文传感器数据，雷达数据流，NCAR低层大气观测设施以及地球和海洋社区的外展方面的一些初步测试案例。