37. SCIENTIFIC VISUALIZATION AND DATA UNDERSTANDING

Scientific visualization and data management are critical enabling technologies for computational science research, which provide scientists with the capability to extract the scientific insights from data sets generated by simulations and experiments.  The visualization systems that are sought must be attuned to the needs of domain scientists and be integrated with important data management and domain-specific science.  In addition, to be part of a useful investigatory scientific research environment, visualization systems and data analytics must further be integrated with supporting computational science technologies such as petascale computing, data management and storage/retrieval; I/O capabilities, and networking capabilities for remote visualization.

a. Scientific Visualization and Management—Scientific discoveries enabled by petascale computational sciences require advanced visualization systems to extract the scientific insights from data generated by simulation and experiments.  With petascale computing and other experiments expected to generate several petabyte of unstructured multi-dimensional data sets per year, next-generation scientific visualization systems will outstrip the performance of today’s systems.  Next-generation data analytics will be needed to:  1) compare data between different simulation runs, 2) perform statistical analysis on data, 3) perform comparison between simulation data and experimental data, and 4) accommodate uncertainties in data.  Grant applications in this subtopic should focus on, but are not limited to:  1) parallel visualization tools and services, 2) tools to enable knowledge discovery at petascale, 3) GPU programming model to multi-GPU systems, 4) high-level I/O libraries optimized for large-scale scientific visualization, 5) full-feature visualization tools for unstructured data, and 6) uncertainty management in scientific visualization.

Questions – contact Thomas Ndousse-Fetter (tndousse@science.doe.gov

b. Petabyte-Scale Data Transformation, Discovery, and Distribution—Science is increasingly becoming more and more data-intensive.  In many large-scale scientific experiments and simulations, the data challenge already exceeds the compute-challenge in its needed resources.  The scientific importance of storing, discovering, and distributing scientific data on an unprecedented scale to scientists in different geographical locations is clear—it is the limiting or enabling factor of scientific discoveries in many large-scale data-intensive science involving distributed resources and research teams.  Grant applications are sought to focus on the development of scalable tools to facilitate the transformation, discovery, and distribution of scientific data (unstructured data).  These include but are not limited to:  1) high-speed data transfer protocols and services optimized for optical links, 2) salable tools for storage systems coordination and scheduling, 3) metadata and data replication services to support data distribution, and 4) tools to interface data management systems to network and storage systems. Grant applications focusing on commercial database management extension are beyond the scope of this subtopic and will not be peer-reviewed.

Questions – contact Thomas Ndousse-Fetter (tndousse@science.doe.gov

References:

1.      Bunn, J. and Newman, H., “Data-Intensive Grids for High-Energy Physics,” Grid Computing, Making the Global Infrastructure a Reality, Berman, Fox and Hey, eds., UK:  Wiley, 2003.  (ISBN:  0-470-85319-0)

2.      “Planning ASCR/Office of Science Data-Management Strategy,” Data Management Challenge Workshop Report, 2004.  (Full text available at:  http://www-conf.slac.stanford.edu/dmw2004/docs/DM-strategy-final.doc)

3.   Childs, H. and Miller, M., “Beyond Meat Grinders:  An Analysis Framework Addressing the Scale and Complexity of Large Data Sets,” Proceedings of SpringSim High Performance Computing Symposium (HPC 2006), Huntsville, AL, April 2-6, 2006, pp. 181-186, 2006.  (Full text available at:  http://graphics.idav.ucdavis.edu/publications/print_pub?pub_id=891)

 4.      “Visualization Group:  Current Projects,” Research Activities at Lawrence Berkeley National Laboratory Website, at http://vis.lbl.gov/Research/

 5.      “Scientific Visualization,” Research Activities at Lawrence Livermore National Laboratory Website, at http://www.llnl.gov/graphics/

6.      Visualization Research Activities at Pacific Northwest National Laboratory, at http://www.pnl.gov/news/experts/visualization.stm

 

Return to the Complete List of Topics

 

Program Information, Instructions and Requirements  |  Technical Topic Descriptions  |  Download Program Information  | Download Technical Topics |