Analytics and Informatics Management Systems

Mission Statement: The objective of the Analytics, Informatics, and Management Systems (AIMS) project is to establish Lawrence Livermore National Laboratory (LLNL) as a leader and visionary architect for all aspects of data discovery and knowledge integration. LLNL has demonstrated a viable path forward and attracted key national and international collaborators and potential sponsors. The technical motivation for this program is based on the success of the Earth System Grid Federation (ESGF) and the Ultra-scale Visualization Climate Data Analysis Tools (UV-CDAT). Over the past two decades, the AIMS project has demonstrated LLNL’s ability to enable the scientific climate community to self-organize and build an information infrastructure that has revolutionized how climate modeling and intercomparison are performed. The AIMS program leads collaborations to develop an uncertainty quantification diagnostic test bed and data infrastructure for the DOE Accelerated Climate Modeling for Energy (ACME) project. Moreover, by working with many others in a global setting, the team is resolving many of the technical challenges regarding scaling and federation (e.g., authentication, sharing, location of data and processing resources, interface standards, etc.) issues that face any attempted large-scale information system. This has been accomplished by using an open systems approach that leverages the efforts of a global community.

The AIMS program is essential for the future of the climate science community. To date, global funding sponsors have started to achieve a critical mass to tackle fundamental data-intensive and data-driven science problems. In the climate application arena, projects are consolidating on the ESGF Peer-to-Peer (P2P) infrastructure. This realization places AIMS at the forefront. Our program plays a stewardship role in the transformation in climate science that will lead to integration of data driven infrastructure and analysis with science research and discoveries. It enables the world to better organize and integrate all climate knowledge via a cooperative federation. With that said, the ESGF P2P software has distinguished itself from other collaborative knowledge systems in the climate community, as evidenced by its widespread adoption, federation capabilities, and broad developer base. It is the leading source for current climate data holdings, including the most important and largest data sets in the global-climate community. Through the system, users access, analyze, and visualize data using a globally federated collection of networks, computers, and software. The enterprise system is used to support national and international climate model and observation intercomparison activities as well as high profile U.S. DOE, NOAA, NASA, and NSF projects.

Just as important as building a large distributed system for the dissemination and management of petascale climate resources is the UV-CDAT, which enables analysis, diagnosis, and visualization of data for atmosphere, ocean, and land model components, and other impact studies. As part of the larger ESGF problem-solving environment, UV-CDAT is an open-source, easy-to-use application that links together disparate software subsystems and packages to form an integrated environment for analysis. Because of UV-CDAT’s design and openness, climate related visual analytics software is easily shared among the community. Along this line of thought, other DOE projects involving visualization and analysis software intended for use in the climate community are first—and in many cases, solely—deployed within the UV-CDAT environment. Leveraging work from other scientific communities, UV-CDAT manipulates data via NumPy, a numerical package used by thousands for doing diagnostics, statistics, and other various types of data manipulation. UV-CDAT is the only climate analysis package that is fully compliant with the conventions for Climate and Forecast (CF) metadata, also designed to promote NetCDF file processing and sharing. The UV-CDAT concept is simple and flexible enough to interchange its parts and expand into the future.