Parallel i/o: Difference between revisions
Jump to navigation
Jump to search
(Created page with "Currently statefiles are written from rank 0. Each process, however, reads (only) the data section it needs. NetCDF-4 has HDF5 support - it calls the appropriate HDF5 rout...") |
mNo edit summary |
||
| (2 intermediate revisions by the same user not shown) | |||
| Line 1: | Line 1: | ||
'''NOTE:''' | |||
NetCDF-4 has HDF5 support - it calls the appropriate HDF5 routines that the user would call, and so overhead is nominal. Adding parallel i/o though netCDF would actually be a simplification to the io_save_state routine(!), but as the serial write is unlikely to be a bottleneck when using fewer than 1000 cores, and HDF5 library installation can be tricky, this option has not yet been included. | |||
NetCDF-4 has HDF5 support - it calls the appropriate HDF5 routines that the user would call, and so overhead is nominal. Adding parallel i/o though netCDF would actually be simplification to the | |||
Please contact me if you are interested in seeing this feature added. | Please contact me if you are interested in seeing this feature added. | ||
Latest revision as of 07:37, 27 June 2016
NOTE: NetCDF-4 has HDF5 support - it calls the appropriate HDF5 routines that the user would call, and so overhead is nominal. Adding parallel i/o though netCDF would actually be a simplification to the io_save_state routine(!), but as the serial write is unlikely to be a bottleneck when using fewer than 1000 cores, and HDF5 library installation can be tricky, this option has not yet been included.
Please contact me if you are interested in seeing this feature added.