Working Group Charters
Working Group 1 – High Spectrum Computing and Analysis Needs and Priorities
Identify the characteristics for advanced high spectrum computing and data infrastructure that enables integrated discovery involving experiments, observations, analysis, theory, and simulation. Evaluate the match between resources and demand for the high spectrum of systems, for both compute- and data-intensive applications, and the impacts on the research community if NSF can no longer provide state-of-the-art computing and data analysis for its research community. Provide prioritization guidelines for NSF to meet the needs of high spectrum computing and analysis.
Working Group 2 – Access to Sufficient High Spectrum Resources
There are challenges facing researchers in obtaining access to advanced computational and data analysis resources. These range from allocation processes, over subscription of resources and the need to frequently migrate from one resource to another. Some have said the easiest part is actually getting the at-scale application to work. In other cases, the communities may be self limiting requests to known resources rather than requesting what is actually needed. In reviewing the white papers and other sources, estimate the true computational requirements that would enable all areas of science and engineering to make timely progress in both best of breed problems and common practice problems. Where possible, provide quantitative data on computing needs.
Working Group 3 – Risk, Opportunities and NSF’s Role Fostering High Spectrum Science and Engineering
There are multiple technical challenges to building future, more capable advanced computing and data systems for the next decade. Technology limitations will make some approaches more difficult for applications to use in a productive manner. How should NSF best respond to the challenges for sustained application performance and researcher productivity in the future? What is the optimum ways and frequency to collect requirements and status in the rapidly changing environments? What is the best way to motivate and enable science teams to be ready to use high spectrum resources in a highly efficient manner to do best of breed problems in the most timely manner? What are the risks and opportunities to scientific leadership in current US plans for extreme scale computing and what can NSF do to address and reduce the risks and maximize opportunities
Working Group 4 – Computation and Data Analysis
In reviewing the submitted white papers and other information, analyze the requirements for computation and data analysis for open science and engineering that will need to be met to enable high-spectrum science and engineering in the next decade. Provide a break-down of the system and architectural requirements that are anticipated to enable high spectrum science and engineering.
Working Group 5 – Storage and Data Movement
In reviewing the submitted white papers and other information, analyze the requirements for storage and data movement for open science and engineering that will need to be met to enable high-spectrum science and engineering in the next decade. Data movement is expensive both in terms of energy, investment and human effort. Identify strategies to minimize data movement to make science and engineering teams as productive as possible. Discuss the current number of storage hierarchies (memory, rotating disk, tape) and project the types and levels of hierarchies that will be likely and effective over the next decades.
Working Group 6 – Workflow and Methods
In reviewing the submitted white papers and other information, analyze the requirements for advanced workflows and methods that will need to be met to enable high-spectrum science and engineering in the next decade. Time to insight in many science areas is not only related to the largest scale work steps, but may be dominated by other workflow steps. Today, many workflows have been developed on a project or team basis, with many different assumptions. Project, where possible, the commonalities of workflows and where there could be opportunities for synergies and optimizations.