Elements: FastTract: Web-Based Exploratory Visualization of Gigapixel Astronomical Images

点击翻译按钮获取中文摘要

Like scientists in virtually every field, astronomers are struggling under a"data deluge". While the ever-larger images obtained by the newestobservatories lead to cutting-edge science, they also overwhelm traditionaltools designed to work with files hundreds or thousands of times smallerthan the state-of-the-art. In particular, astronomers are rapidly losing theability to simply look at the images of the sky that are coming out of theirtelescopes. The FastTract Project will solve this problem by marrying existingWeb-based visualization technologies in the AAS WorldWide Telescope softwaresystem with new features and tools needed to efficiently work with the largeastronomical images of the 2020's, enabling US astronomers to fully exploittheir world-class facilities, in particular the ones that open new Windows onthe Universe. The project team will synthesize the insight gained in thisundertaking to establish the Little Big Data University (LBDU), a learningresource for scientists across disciplines who are struggling to stay afloatin the "data deluge". LBDU will follow the Khan Academy model and extensivelyuse interactive environments to help researchers and others learn strategiesfor Harnessing the Data Revolution. These efforts will especially benefitpeople who do not have access to top-tier computational resources, such asthose at small institutions and interested non-specialists.The FastTract Project will create a sustainable cyberinfrastructure (CI)system and associated community of practice that enable exploratory scientificvisualization of large (gigapixel+) astronomical images over the Web. The CIwill build on the NSF-funded AAS WorldWide Telescope (WWT) software system.The research team will extend WWT's image tiling architecture to work withFITS data files, develop the tooling needed for easy creation of such tiles,and produce workflows and tutorials allowing the broad astronomical communityto take ownership of the infrastructure. The team will work with threeNSF-funded partners on specific science applications. Annual workshops willnucleate a core group of project user-contributors and ensure that developmenteffort meets the needs of the broader community. The project will adopt bestpractices in the open-source, open-development paradigm, laying the groundworkfor FastTract infrastructure to be leveraged by non-astronomical communitiesfacing similar data challenges. In parallel, the team will distill itsexpertise to create the Little Big Data University (LBDU), an onlineprofessional development resource aimed at domain scientists who do not intendto become CI specialists. These scientists will learn core strategies forcoping with ever-larger data sets through the empirically successful KhanAcademy model. Telemetry analysis and focus groups will guide curriculumdesign and assess efficacy.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.