As a scientific center, CAVCaM has three inter-related research agendas.

1) High Performance Computing (HPC) is the discipline that utilizes high-end computing systems and parallelism for numerically intensive, scale intensive and time critical analysis. HPC programs emphasize computational intensity so as to increase the size of the problems that can be solved and to decrease the time to solution.

2) Advanced Visualization applies human-in-the-loop computing for discovery, training, collaboration and decision support using appropriate visual and interactive technologies. It seeks to effectively couple the human mind and its capacity for analog investigation with the digital capabilities of computers for data exploration and analysis.

3) Data Intensive Computing (DIC) involves processing and managing massive amounts of data – as might be derived from remote or ground based sensing – that enables informational content to be effectively and efficiently extracted. The over-arching goal in this effort is to use various approaches to process, condition, archive and access data for on-demand knowledge discovery and data mining.

CAVCaM endeavors to bring together government leaders, industry partners and researcher expertise to develop data and a cyber infrastructure that leads to practical applications for environmental problems.

Some of Center’s focus for advancing the research and educational mission of the DRI include:

• Computation: Support the utilization of efficient numerical procedures to solve problems using deterministic or statistical methodologies.
• Visualization: Encourage and support the use of visualization for engaging the domain knowledge and analytical capabilities of human subject matter experts through techniques that include: remote graphics, stereographic immersion, interaction and analytics.
• Software: Foster the expert use, adaptation, and development of computer applications and tools through training and consulting;
• Simulation: Advocate appropriate hardware and software approaches for the use with computational models that results in numerical and visual output; as might be used in forecasting, optimization, validation, design, analysis, quantification and what-if scenarios.
• Data: Support effective strategies for storage and analysis of datasets – particularly massive datasets – that might derive from field acquisition, laboratory instruments or numerical computation.
• Collaboration: Drive collaboration, partnership and interdisciplinary research in academic science and industrial engineering to solve problems.
• Education: Provide instructional support and outreach to both specialists and non-specialists, especially policy planners and students at all levels.