XDMF (eXtensible Data Model and Format) is a library providing a standard way to access data produced by HPC codes. Data format refers to the raw data to be manipulated, the description of the data is separate from the values themselves. It distinguishes the metadata (Light data) and the values themselves (Heavy data). Light data is stored using XML, Heavy data is typically stored using HDF5, so some information is stored redundantly in both XML and HDF5. A Python interface exists for manipulating both Light and Heavy data. ParaView, VisIt and EnSight visualization programs are able to read XDMF.