of the targeted MPI implementations.
You can switch the automatic selector through the
-\c smpi/coll_selector configuration item. Possible values:
+\c smpi/coll-selector configuration item. Possible values:
- <b>ompi</b>: default selection logic of OpenMPI (version 1.7)
- <b>mpich</b>: default selection logic of MPICH (version 3.0.4)
- mvapich2: use mvapich2 selector for the alltoall operations
- impi: use intel mpi selector for the alltoall operations
- automatic (experimental): use an automatic self-benchmarking algorithm
- - bruck: Described by Bruck et.al. in <a href="http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=642949">
+ - bruck: Described by Bruck et.al. in <a href="http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=642949">this paper</a>
- 2dmesh: organizes the nodes as a two dimensional mesh, and perform allgather
along the dimensions
- 3dmesh: adds a third dimension to the previous algorithm
Example: adding a "pair" version of the Alltoall collective.
- - Implement it in a file called alltoall-pair.c in the src/smpi/colls folder. This file should include colls_private.h.
+ - Implement it in a file called alltoall-pair.c in the src/smpi/colls folder. This file should include colls_private.hpp.
- The name of the new algorithm function should be smpi_coll_tuned_alltoall_pair, with the same signature as MPI_Alltoall.