Data-Intensive Computing Initiative (DICI)
Demonstration: Decision Support & Control
The Decision Support and Control (DSC) demonstration addresses increasingly large and complex data volumes to enable near-real-time informed human decisions or automated response actions. Current I/O capacity limitations hinder the timely acquisition, processing, and presentation of actionable information to decision makers for rapid response. To minimize these I/O limitations, PNNL researchers will leverage a coherent collection of sensor technologies for data transformation, augmentation, and fusion routines that will provide decision makers with actionable information in near-real-time.
Goals:
- Rapid data monitoring, processing, analysis, and actions for streaming internet traffic.
- Identification of features relevant to analytic mission objectives (sensor-to-analyst with feedback to the sensor).
- Dynamic interactions among analysts, sensors, and Analysis Fusion Environment to facilitate real-time decision making and information sharing.
Approach:
- Cell processor-hosted algorithms attach context (CON-Tags) to raw streaming traffic augmented with data from additional sensors (e.g., FLO, SNO).
- Automated Analysis Fusion Environment infers context and associates it with data of interest (e.g., behavior, attribution, temporal patterns, content).
- MeDICI architecture enables plug-and-play workflow modules for constructing and editing workflows.
The DSC demonstration is planned in three phases, shown here in green (I), gray (II), and blue (III).
Projects that support this demonstration in Phase I:
- MeDICI Architecture (Adaptive Software Architecture), (A Data Virtualization Architecture), (Adaptive Workflow in Data-Intensive Environments)
- Adaptive Network Traffic Analysis on the Cell Processor
- Multithreaded Architectures for DI Computing Applications
- Near-Real-Time Situational Awareness from Massive Sensor Data
Projects that support this demonstration in Phase II:
- MeDICI Architecture
- Data-Intensive Machine Learning for Real-Time Decision Analysis
- Adaptive Composite Analysis for Complex Systems
Projects that support this demonstration in Phase III:
- Adaptive Workflow in Data-Intensive Environments