Keynote Speaker III: Analysing Large-Scale Data

Le résumé fourni par la source

Scientific communities have to cope with steadily increasing data rates. Even smaller communities recognize that their traditional "best-practice solutions" are no longer sufficient anymore. The project "Large-Scale Data Management and Analysis (LSDMA)" of the German Helmholtz Association supports their research centres since 2012 in analyzing their data. The activities include developments of community-specific solutions and provide, in addition, generic technologies for data management, data access, data security, data preservation. and data analysis. The talk gives an overview on selected activities and results of LSDMA. New Big Science projects are entering the scene putting the term "large-scale data" to a new level. The European XFEL will generate extremely intense flashes of X-rays for exploring the nano-world at unprecedented precision. The Square Kilometre Array (SKA) will consist of thousands of telescopes to be built in deserts of South Africa and Australia. The images taken from the Universe will be the sharpest of all astronomy. The talk will present the challenges of these projects seen from the IT's perspective. For example, even in the first phase with only 10% of the telescopes installed, SKA will produce data rates of the order of a few Zettabytes per year. Topics considered in the talk are new IT-concepts like "taking computing to the data", integrating cloud computing into a scientific work flow, secure data federation, developing long-term e-infrastructures for supporting multi-disciplinary science.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.