[mvapich] Announcing the release of MVAPICH2 0.6.0 (MPI-2 for InfiniBand) + shared library support for MVAPICH (MPI-1) 0.9.4

[Dhabaleswar Panda]

You are receiving this e-mail because of your interest in the MPI over
InfiniBand package, developed in the Network-Based Computing
Laboratory at the Ohio State University.

This is to announce the release of MVAPICH2 (MPI-2 for InfiniBand)
0.6.0 for multiple platforms (EM64T, G5, IA-32, IA-64, and Opteron)
and network interfaces (PCI-X and PCI-Express). 

Starting with this 0.6.0 release, MVAPICH2 is being distributed as a
single integrated package (with the latest MPICH2 0.971). It can be
downloaded with a `single click' and installed. It is also available
under BSD license.

This new release has the following features:

      - MPI-2 functionalities (one-sided, collectives, datatype)
      - all MPI-1 functionalities
      - optimized one-sided operations (Get, Put, and Accumulate)
      - support for active and passive synchronization
      - optimized two-sided operations
      - scalable job start-up
      - optimized and tuned for the above platforms and different 
        network interfaces (PCI-X and PCI-Express)
      - single code base for all of the above platforms
      - memory efficient scaling modes for medium and large clusters

Other features of this release include:

- Excellent performance: For two-sided operations, MVAPICH2 0.6.0
  delivers 5.0 microsec latency (with the switch), up to 965 MB/sec 
  unidirectional bandwidth, and up to 1725 MB/sec bidirectional 
  bandwidth on EM64T system with PCI-Express. For one-sided Put 
  operation, MVAPICH2 0.6.0 delivers 6.98 microsec latency (with 
  the switch), and up to 972 MB/sec unidirectional bandwidth on the 
  above platform. Detailed performance numbers for two-sided and all
  one-sided operations on all platforms are available on the project's
  web page.

- A set of benchmarks to evaluate one-sided operations (Put, Get, and 
  Accumulate)

- A detailed `User Guide' to assist users to install this package on
  different platforms with different options.

- A detailed `Performance Tuning Guide' to vary different parameters
  of the MPI installation to extract maximum performance and achieve
  scalability, especially on large-scale systems.

--------------------------------------------------------------------

In addition to the above MVAPICH2 release, we are also pleased to
announce the `shared library' support for MVAPICH 0.9.4. This support
provides flexibility for using software from ISVs in binary form to be
compiled, linked, and executed with MVAPICH 0.9.4. We also introduce a
new bug fix page which provides patches for recent bug fixes. Current
users of 0.9.4 are welcome to download these patches.

--------------------------------------------------------------------


You are welcome to download the MVAPICH2 0.6.0 package and the shared
library support for MVAPICH 0.9.4 and access relevant information from
the following URL:

http://nowlab.cis.ohio-state.edu/projects/mpi-iba/

All feedbacks, including bug reports and hints for performance tuning,
are welcome. Please send an e-mail to mvapich-help@cis.ohio-state.edu.

Thanks, 

MVAPICH Team at OSU/NBCL 

----------

PS: If you would like to be removed from this mailing list, please end
an e-mail to mvapich_request@cis.ohio-state.edu.