| Tools and Capabilities |
|
The most important aspect of the SDAV institute is to deliver tools and technologies to DOE scientists that can not only scale to large core counts but also can be sustainable. Scientific software often emerges from research as a prototype, and with additional work to package, document, and test becomes a software product, a production tool that is ready to be deployed and used by others. Our Institute favors these software products - libraries and frameworks - over custom solutions. Products that benefit from an economy of scale in development and deployment, often receive leverage from other funding agencies, are higher quality software, and enjoy more sustainable futures. Of course, our overall goal is to assist applications teams with their data management, analysis, and visualization needs to achieve breakthrough science, so in some cases we may deploy custom solutions to meet these goals at minimum cost. Simply devoting the majority of our resources to libraries and frameworks is still a daunting task: there is a large number of each that we could potentially support. We approach this challenge by judiciously supporting technologies required by application scientists, and, when possible, converging on a minimal set of technologies that can easily be supported by our team. Further, many of these technologies provide complementary functionality and, where appropriate, we plan to increase their interoperability, providing new capabilities to application scientists at reduced development costs.
Data Management Chairs: This e-mail address is being protected from spambots. You need JavaScript enabled to view it , This e-mail address is being protected from spambots. You need JavaScript enabled to view it Experts in our field expect that, as concurrency grows, there will be a widening gap between computational and I/O capacity, and this will be further stressed by energy demands. Our approach is to perform as much work as possible while the data is still resident in application memory, a use model often referred to as “in-situ.”
Data Analysis Chairs: This e-mail address is being protected from spambots. You need JavaScript enabled to view it , This e-mail address is being protected from spambots. You need JavaScript enabled to view it Early analysis of the data generated by large scale simulations is crucial to effectively monitor the correct progress of simulations and the understanding of simulation results. Such early analysis capability requires an in-situ monitoring framework that allows analysis tasks to take place “on-the-fly” while data is generated.
Visualization Chairs: This e-mail address is being protected from spambots. You need JavaScript enabled to view it , This e-mail address is being protected from spambots. You need JavaScript enabled to view it While the idea of a single library upon which applications are built is well grounded in software engineering principles, the current ubiquitous library was designed during the era of the single core processor. Our plan is to engage a wide segment of leading visualization and system researchers and developers to produce a new library that is architecture-aware and can serve as the basis for future visualization applications.
Scientific Software Tools Chairs: This e-mail address is being protected from spambots. You need JavaScript enabled to view it , This e-mail address is being protected from spambots. You need JavaScript enabled to view it , This e-mail address is being protected from spambots. You need JavaScript enabled to view it A sustainable software infrastructure requires quality assurance, regression testing, distribution, and tracking feedback from the users. Our intent is to deliver a software infrastructure to the scientific community that couples the best practices from both research and development. |