Collective I/O on Chunked Datasets

Collective I/O on chunked datasets can be performed either as a fully collective I/O operation on all the chunks at once (“linked-chunk I/O”), or partially collective, with collective I/O on each chunk individually (“multi-chunk I/O”).  The general flowchart for this decision making looks like this:


Pasted Graphic 2.pdf


The user input for this algorithm is provided with the following API routines which set properties on a dataset transfer property list (DXPL) that can be used in a call to H5Dread or H5Dwrite.





Describing this in pseudo-code: (source code is in H5D_chunk_collective_io() in src/H5Dmpio.c)

Page Index