[svn-r5429] Purpose:

    Bug fix/Code improvement.

Description:
    Currently, the chunk data allocation routine invoked to allocate space for
    the entire dataset is inefficient.  It writes out each chunk in the dataset,
    whether it is already allocated or not.  Additionally, this happens not
    only when it is created, but also anytime it is opened for writing, or the
    dataset is extended.  Worse, there's too much parallel I/O syncronization,
    which slows things down even more.

Solution:
    Only attempt to write out chunks that don't already exist.  Additionally,
    share the I/O writing between all the nodes, instead of writing everything
    with process 0.  Then, only block with MPI_Barrier if chunks were actually
    created.

Platforms tested:
    IRIX64 6.5 (modi4)

1 files changed, 2 insertions, 0 deletions

diff --git a/src/H5Fpkg.h b/src/H5Fpkg.h
index 72cd321..30439b4 100644
--- a/src/H5Fpkg.h
+++ b/src/H5Fpkg.h
@@ -194,10 +194,12 @@ __DLL__ herr_t H5F_istore_write(H5F_t *f, hid_t dxpl_id,
                                 const hsize_t size_m[], const hssize_t offset_m[],
                                 const hssize_t offset[], const hsize_t size[],
                                 const void *buf);
+#ifdef H5_HAVE_PARALLEL
 __DLL__ herr_t H5F_istore_allocate (H5F_t *f, hid_t dxpl_id,
                                 const struct H5O_layout_t *layout,
                                 const hsize_t *space_dim,
                                 struct H5P_genplist_t *dc_plist);
+#endif /* H5_HAVE_PARALLEL */
 
 /* Functions that operate on contiguous storage wrt boot block */
 __DLL__ herr_t H5F_contig_read(H5F_t *f, hsize_t max_data, H5FD_mem_t type, haddr_t addr,