I developed an F90-interface to MPI providing
* explicit interfaces to all procedures defined by the MPI-1.1 standard
* direct support for scalar and vector arguments
* enhanced safety due to strong type checking and reduced risk of
message buffers copying
* support for optional argument
e.g., you can use
call MPI_SEND(buf, count, datatype, dest, tag, comm)
or
call MPI_SEND(buf, dest = dest)
where "count" and "datatype" are set automatically (for scalars "buf"
and vectors "buf(:)")
and "tag" and "comm" are set to defaults "0" and "MPI_COMM_WORLD",
respectively.
Sources and documentation are available at
http://rcswww.urz.tu-dresden.de/~jstiller/software/software.html
Joerg
Choi-Hong Lai wrote:
>Dear Colleagues
>
>I wonder whether there is any MPI implementation to be used
>with F90/F95?
>
>Choi-Hong
>--
>#============================================
># LAI, Choi-Hong (BSc, PhD, FIMA, CMath, MBCS, CEng)
># Reader in Scientific Computing
># T: +44 (20) 8331 8712 F: +44 (20) 8331 8665
># [log in to unmask] http://cms1.gre.ac.uk/
># http://www.gre.ac.uk/~C.H.Lai
># Office location: M318
># University of Greenwich - Maritime Greenwich Campus
># School of Computing and Mathematical Sciences
># Old Royal Naval College, Greenwich, London SE10 9LS, U K
>#============================================
>
>This message and any attachment are confidential and
>may be privileged or otherwise protected from disclosure.
>If you are not the intended recipient, please telephone
>or email the sender and delete this message and any
>attachment from your system. If you are not the intended
>recipient you must not copy this message or attachment
>or disclose the contents to any other person.
>
>.
>
--
Joerg Stiller
TU Dresden, Institute for Aerospace Engineering
Tel: 0351-463-38328
Fax: 0351-463-38087
mailto:[log in to unmask]
http://www.tu-dresden.de/mw/ilr/tfd/stiller
|