[mpiwg-persistence] Summary of Semantics for Partitioned Communication Operations

Bangalore, Purushotham puri at uab.edu
Wed Jul 15 12:24:18 CDT 2020


Hi All,

I was planning to bring this topic at today's meeting. Since we did not meet, I would like all of you to review the attached spreadsheet and complete the corresponding columns for stages, completion, blocking/nonblocking, type of operation, and resources blocked/comment.

I have copied the persistent point-to-point section of the table and used it as a starting point. Please review and let me know what you all think.

For the operation column, I have entered ?-op. Do we need to define "pp-op" to denote partitioned communication operation? I have added a footnote that indicates the difference between persistent initialization functions and partitioned initialization functions.

Thanks,
Puri
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mpi-forum.org/pipermail/mpiwg-persistence/attachments/20200715/f987198a/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: MPI-semantics-appendix.xlsx
Type: application/vnd.openxmlformats-officedocument.spreadsheetml.sheet
Size: 12723 bytes
Desc: MPI-semantics-appendix.xlsx
URL: <http://lists.mpi-forum.org/pipermail/mpiwg-persistence/attachments/20200715/f987198a/attachment-0001.xlsx>


More information about the mpiwg-persistence mailing list