Hello, I use HDF on a MPI based cluster without any MPI access calls, so each core should read / write to the HDF file. Is there any need to compile HDF with MPI support? Each core need to have only write access to different datasets and read access to all (after write). Should I need a mutex or HDF with MPI support enable?
Thanks Phil _______________________________________________ Hdf-forum is for HDF software users discussion. [email protected] http://mail.hdfgroup.org/mailman/listinfo/hdf-forum_hdfgroup.org
