[svn-r5680] Purpose:

    New feature.

Description:
    There is some discussion among the SAF team as to whether it is better
    to use MPI derived types for raw data transfers (thus needing a
    MPI_File_set_view() call), or whether it is better to use a sequence of
    low-level MPI types (i.e. MPI_BYTE) for the raw data transfer.

Solution:
    Added an in internal flag to determine whether derived types are preferred
    (the default), or whether they should be avoided.  An environment variable
    ("HDF5_MPI_PREFER_DERIVED_TYPES") can be set by users to control whether MPI
    types should be used or not.  Set the environment variable to "0" (i.e.:
    'setenv HDF5_MPI_PREFER_DERIVED_TYPES 0' to avoid using MPI derived types.

Platforms tested:
    IRIX64 6.5 (modi4) w/parallel

1 files changed, 7 insertions, 0 deletions

diff --git a/release_docs/RELEASE.txt b/release_docs/RELEASE.txt
index aa608a3..c07a2a3 100644
--- a/release_docs/RELEASE.txt
+++ b/release_docs/RELEASE.txt
@@ -185,6 +185,13 @@ Documentation
 New Features
 ============
 
+    * Added environment variable flag to control whether creating MPI derived
+      typed is preferred or not.  This can affect performance, depending on
+      which way the MPI-I/O library is optimized for.  The default is set to
+      prefer MPI derived types for collective raw data transfers, setting the
+      HDF5_MPI_PREFER_DERIVED_TYPES environment variable to "0" (i.e.:
+      "setenv HDF5_MPI_PREFER_DERIVED_TYPES 0") changes the preference to avoid
+      using then whenever possible.  QAK - 2002/06/19
     * Changed MPI I/O routines to avoid creating MPI derived types (and thus
       needing to set the file view) for contiguous selections within datasets,
       which should result in some performance improvement for those types of