Data-Intensive Computing Initiative (DICI)
Middleware for Data-Intensive Computing (MeDICI)
Technical Contacts: Ian Gorton; Justin Almquist
Executive Summary
The MeDICI technology is providing an integrated information architecture to support advanced analytical applications. The architecture provides a common platform through which to apply the results of multiple information processing and analysis capabilities and to demonstrate their applicability in the areas of decision support and control, scientific discovery and insight, and situational awareness and response.
Accomplishments / Highlights
- The first prototype of the MeDICI platform formed the core of PNNL's network monitoring at Supercomputing 2006 (SC06) in Tampa, Florida.
- Several analytical application demonstrations in homeland security, cybersecurity, and bioinformatics have been built using MeDICI.
- A paper on the MeDICI platform was published at the 7th IEEE International Conference on COTS-Based Software Systems.
Collaboration
- The MeDICI project is strongly coupled with the Data Virtualization and Adaptive Workflow projects.
- Collaborating with National Information and Communications Technology Australia on adaptive architectures for advanced MeDICI deployments.
Demonstration
The MeDICI platform will tightly integrate with the Data-Intensive Computing Initiative's workflow and data virtualization projects and will form the underlying platforms for all the Initiative's demonstrations. The figure on the right shows the design of the SC06 cybersecurity application that was built upon the SIFT architecture, an early version of the MeDICI platform.
Impacts
The MeDICI platform will make it cost-effective and efficient to create advanced, adaptive data-intensive computing applications. The platform will underpin PNNL's solutions to future data-intensive problems in diverse application domains including bioinformatics, cybersecurity, intelligence analysis, and the power grid. These applications will integrate high-throughput scientific instruments, network sensors, and distributed databases with new analytical and visualization capabilities developed at PNNL.