CEO:P--A Data-Intensive Cyberinfrastructure Component for Coastal Forecasting and Change Analysis

摘要OCI 0619041多年来，人们在环境观测和建模方面做了大量工作。许多复杂的系统已经建成或正在建设中。尽管在收集的数据量方面取得了进展(包括来源数量的增加以及时空粒度的增加)，并改进了用于分析这些数据集的技术，但在这一领域仍然存在一些挑战。首先，当前的系统非常紧密地耦合在一起。在不同的系统中几乎没有任何算法实现的重用。测试或整合新的分析算法也是极其困难的。实现与可用资源紧密相关，最后，现有系统不能使分析的粒度适应资源可用性和时间限制。(密切相关的)面向服务的体系结构和网格计算概念的出现趋势可以缓解上述问题。它们可以开发不绑定到特定数据集或终端应用程序的服务，并使用这些服务实现应用程序。然而，这也需要能够支持流应用和数据虚拟化/集成的网格中间件组件的进步。该项目建议开发和评估用于环境应用的网络基础设施组件。这将包括中间件、模型集成、分析和挖掘技术的开发，以及使用服务模型来支持两个密切相关的应用程序。这些应用将用于目前海岸的实时预报和预报，以及长期的海岸侵蚀分析和预报。解决的具体问题如下。在第一个应用中，重点将放在海岸条件的实时预报和预报上。中间件和面向服务的实施将被用来允许插入新的算法(例如，用于海滩关闭和大肠菌群预测)，允许根据资源和时间限制使用更复杂的模型，允许灵活地插入新的数据流，并允许对预测/现在预测模型产生的数据进行分析和解释的新算法。在第二个应用中，将开发用于长期海岸变化和侵蚀模式的先进模型，并允许进行更大规模、分布式和灵活的数据分析。将在五大湖观测系统(GLOS)的范围内实施和评估，并将与国家海洋和大气管理局(NOAA)联合进行。这是一个很好的机会，可以对网络基础设施组件进行现实的设计、部署和评估，也可以影响真实环境观测系统的长期设计和运行。该项目将由俄亥俄州立大学(OSU)和美国国家海洋和大气管理局(NOAA)共同努力。OSU团队包括两名计算机科学研究人员：Gagan Agrawal(网格中间件系统)和Hakan Ferhatosmanoglu(数据库和数据分析)，以及两名环境研究人员：Keith Bedford(环境建模)和Ron Li(地理空间数据分析和遥感)。NOAA的合作者包括NOAA-国家海洋局(NOS)的Frank Aikman博士和NOAA-五大湖环境研究实验室(GLERL)的David Schwab博士。

Abstract OCI 0619041Over the years, much work has been done on observing and modeling the environment. Many complex systems have been, or are being, built. Despite advances in the amount of data being collected, (including larger number of sources as well as increased spatio-temporal granularity) and enhancements in the techniques being used for analyzing these datasets, a number of challenges remain in this area. Firstly, the current systems are very tightly coupled. There is hardly any reuse of algorithm implementations across different systems. It is also extremely hard to test or incorporate new analysis algorithms. The implementations are closely tied to the available resources, and finally, the existing systems cannot adapt the granularity of analysis to the resource availability and time constraints. The emerging trend towards (closely related) concepts of service-oriented architectures and grid computing can alleviate the above problems. They can enable development of services that are not tied to specific datasets or end applications, and implementation of applications using these services. However, this also requires advances in grid middleware components that are able to support streaming applications and data virtualization/integration.This project proposes to develop and evaluate a cyberinfrastructure component for environmental applications. This will include developments in middleware, model integration, analysis, and mining techniques, and the use of a service model for supporting two closely related applications. These applications will be real-time coastal now casting and forecasting, and long term coastal erosion analysis and prediction. The specific problems addressed are as follows. In the first application, focus will be on real-time now casting and forecasting of coastal conditions. Middleware and service-oriented implementation will be used to allow new algorithms to be inserted (for example, for beach closings and coliform forecasts), allow more complex models to be used based on resource and time constraints, allow new data streams to be inserted flexibly, and allow new algorithms for analysis and interpretation to be operated on data being produced from forecasting/now casting models. In the second application, advanced models will be developed for long-term coastal changes and erosion patterns, and allow larger scale, distributed, and flexible data analysis. Implementation and evaluation will be in the context of the Great Lakes Observing System (GLOS) and will be done jointly with the National Oceanic and Atmospheric Administration (NOAA). This is an excellent opportunity to carry out realisticdesign, deployment, and evaluation of the cyberinfrastructure component, and also impact the long-term design and operation of a real environmental observation system.This project will be a joint effort between The Ohio State University (OSU) and the National Oceanic and Atmospheric Administration (NOAA). The OSU team includes two computer science researchers: Gagan Agrawal (grid middleware systems) and Hakan Ferhatosmanoglu (databases and data analysis), and two environmental researchers: Keith Bedford (environmental modeling) and Ron Li (geospatial data analysis and remote sensing). The NOAA collaborators include Dr. Frank Aikman,NOAA-National Ocean Service (NOS), and Dr. David Schwab, NOAA -Great Lakes Environmental Research Lab (GLERL).