eScience Workshop 2005 - Computational Data Grid for Scientific and Biomedical Applications

Опубликовано 7 сентября 2016, 16:11

The goal of this project is to develop a Microsoft Windows-based Computer Grid infrastructure that will support high performance scientific computing and integration of multi source biometric applications. The University of Houston Microsoft Windows-based Computer Grid (WING) includes not only the Computer Science and the Technology Department networks, but also includes nodes in China, Germany, and several other countries. The total amount of available storage exceeds 4 Terabytes. Four specific biomedical applications developed at University of Houston are the basis of this project: Computational tracking of Human Learning using Functional Brain Imaging Monitoring Human Physiology at a Distance by using Infrared Technology Multimodal Face Recognition and Facial Expression Analysis Relating Video, Thermal Imaging, and EEG Analysis ΓÇö integrate and analyze simultaneously recorded brain activity, infrared images, and 3D video This Biomedical Data Grid project meets the following technical requirements: Rapid application development (use of the Microsoft Visual Studio .NET technology) Visual modeling interfaces (forms driven Graphical User Interfaces) Database Connectivity (interface with Microsoft SQL Server 2005) Query support (clients can store, update, delete, retrieve database metadata) Context-sensitive, role-based access (Microsoft Windows Server 2003, ASP.NET) Robust security (HIPPA compliance through MicrosoftΓÇÖs Authentication and Authorization from IIS and ASP.NET) Connectivity to other biomedical resources (PACS, DICOM, XML) The Biomedical Data Grid application is developed using Microsoft Windows Server 2003, Microsoft Virtual Server 2005, Microsoft Visual Studio .NET Beta 2, and the Microsoft SQL Server 2005. A web client will be able to securely upload biomedical files to a web server while metadata related to these files will be stored in the SQL Server 2005 database for the purpose of querying, data mining, etc. Post-processing and simulation steps on biomedical data will be using a Master node Web Service that automatically distributes a large set of parameter or sensitivity analysis tasks to Slave nodes on the Computing Grid. We will give an overview of our project and provide a few examples of our biomedical applications.