Skip to main content


HDF5 (the Hierarchical Data Format, version 5) is a library, data model, and file format designed for managing data, especially in HPC where efficient I/O is a high priority. While there are many ways in which HDF5 can be used for modeling data and creating file formats, in this module we focus on how to use HDF5 within a program to save the relevant parts of state needed in order to resume the application with minimal loss of CPU time (this is called application-level C/R). We also cover other topics that are important for any form application-level C/R, even if you are using an API other than HDF5, or no API at all.

Brandon Barker
Cornell Center for Advanced Computing
September 2016