[mpas-developers] derived data type redesign
Michael Duda
duda at ucar.edu
Tue Dec 20 12:07:04 MST 2011
Hi, All.
Attached is a first draft of a requirements and design document for the
first item from the meeting between Doug, Conrad, and I last week; I've
attached a PDF as well as the Latex source for editing. My assessment is
that the changes to the DDTs themselves should be simple -- it's the
remaining items on the list, below, that will be tricky.
Cheers,
Michael
On Wed, Dec 14, 2011 at 03:47:35PM -0700, Michael Duda wrote:
> Hi, Doug and Conrad.
>
> Here are the key development issues that we identified this morning, in
> roughly the order we'll probably need to tackle them:
>
> 1) Update/extend the fundamental derived types in mpas_grid_types.F.
> In order for other parts of the infrastructure to handle multiple
> blocks per task in a clean way, we'll need to be able to pass a head
> pointer to a field into a routine, and have that routine loop through
> all blocks for that field, with information about which cells/edges/vertices
> in that field need to be communicated.
>
> 2a) Decide on a new MPAS I/O abstraction layer, which will provide a
> high-level interface to the PIO layer for the rest of MPAS. This layer
> should work with blocks of fields, and make it possible to define an
> arbitrary set of I/O streams at run-time.
>
> 2b) Add a new module to parse a run-time I/O configuration file that
> will describe which fields are read or written to each of the I/O
> streams that a user requests via the file. This module will make calls
> to the new MPAS I/O layer to register the requested fields for I/O in
> the requested streams.
>
> 3) Update the mpas_dmpar module to support communication operations on
> multiple blocks per task. This will likely involve revising the
> internal data structures used to define communication of cells
> between tasks, and also require revisions to the public interface
> routines themselves.
>
> 4) Modify the block_decomp module to enable a task to get a list of
> cells in more than one block that it is to be the owner of.
> Implemented in the simplest way, there could simply be a namelist
> option to specify how many blocks each task should own, and the
> block_decomp module could look for a graph.info.part.n file, with
> n=num_blocks_per_task*num_tasks, and assign blocks k, 2k, 3k, ...,
> num_blocks_per_task*k to task k.
>
> Some of the changes, e.g., to the dmpar module, are significant enough
> that we should consider whether a general re-write of the module would
> be appropriate; in such cases, we should consider whether the code can
> be made more concise and more maintainable if it were written in C.
>
> Michael
More information about the mpas-developers
mailing list