Title :
Enhancements to the eXtensible Data Model and Format (XDMF)
Author :
Clarke, Jerry A. ; Mark, Eric R.
Author_Institution :
US Army Res. Lab., Aberdeen
Abstract :
The extensible data model and format (XDMF) is an active common data hub used to pass meta data and value between application modules in a standard fashion. XDMF views data as consisting of two basic types: light data and heavy data. Light data is both meta data and small amounts of values. Heavy data typically consists of large arrays of values. Support for XDMF is included in the widely used visualization package ParaView, computational codes like ALE3D from Lawrence Livermore National Laboratory (LLNL) and the SIERRA computational framework from Sandia National Laboratory (SNL). A major effort is now under way to enhance XDMF to more directly support concepts like adaptive mesh refinement (AMR), mixed topologies, higher order elements and parallel datasets. Discussed here are the motivations for the enhancements and the benefits of the new version.
Keywords :
electronic data interchange; meta data; Lawrence Livermore National Laboratory; adaptive mesh refinement; computational codes; data hub; extensible data model and format; heavy data; higher order elements; light data; meta data; parallel datasets; Adaptive mesh refinement; Data models; Data visualization; HTML; High performance computing; Java; Laboratories; Packaging; Topology; XML;
Conference_Titel :
DoD High Performance Computing Modernization Program Users Group Conference, 2007
Conference_Location :
Pittsburgh, PA
Print_ISBN :
978-0-7695-3088-5
DOI :
10.1109/HPCMP-UGC.2007.30
