[mvapich-discuss] Announcing the release of MVAPICH2 2.2a, MVAPICH2-X 2.2a and OMB 5.0

[Panda, Dhabaleswar]

The MVAPICH team is pleased to announce the release of MVAPICH2 2.2a,
MVAPICH2-X 2.2a (Advanced MPI Features and Hybrid MPI+PGAS (OpenSHMEM,
UPC and CAF) with Unified Communication Runtime), and OSU
Micro-Benchmarks (OMB) 5.0.

* Features and Enhancements (since 2.1 GA):

  - Based on MPICH 3.1.4
  - Support for backing on-demand UD CM information with shared memory
    for minimizing memory footprint
  - Reorganized HCA-aware process mapping
  - Dynamic identification of maximum read/atomic operations
    supported by HCA
  - Enabling support for intra-node communications in RoCE mode without
    shared memory
  - Updated to hwloc 1.11.0
  - Updated to sm_20 kernel optimizations for MPI Datatypes
  - Automatic detection and tuning for 24-core Haswell architecture

* Bug Fixes (since MVAPICH 2.1 GA):

  - Fix for error with multi-vbuf design for GPU based communication
  - Fix bugs with hybrid UD/RC/XRC communications
  - Fix for MPICH putfence/getfence for large messages
  - Fix for error in collective tuning framework
  - Fix validation failure with Alltoall with IN_PLACE option
         - Thanks for Mahidhar Tatineni @SDSC for the report
  - Fix bug with MPI_Reduce with IN_PLACE option
     - Thanks to Markus Geimer for the report
  - Fix for compilation failures with multicast disabled
     - Thanks to Devesh Sharma @Emulex for the report
  - Fix bug with MPI_Bcast
  - Fix IPC selection for shared GPU mode systems
  - Fix for build time warnings and memory leaks
  - Fix issues with Dynamic Process Management
         - Thanks to Neil Spruit for the report
   - Fix bug in architecture detection code
     - Thanks to Adam Moody @LLNL for the report

MVAPICH2-X 2.2a provides support for advanced MPI features (Dynamic
Connected Transport and Non-blocking Collectives with Core-Direct)
and hybrid MPI+PGAS (UPC, OpenSHMEM and CAF) programming models with
unified communication runtime for emerging exascale systems. This
library also provides flexibility for users to write applications
using the following programming models with a unified communication
runtime: MPI, MPI+OpenMP, pure UPC, pure OpenSHMEM, and pure CAF
programs as well as hybrid MPI(+OpenMP) + PGAS (UPC, OpenSHMEM and
CAF) programs.

Features and enhancements for MVAPICH2-X 2.2a are as follows:

* Features and Enhancements (since 2.1 GA):

    - MPI Features
        - Based on MVAPICH2 2.2a (OFA-IB-CH3 interface)

    - MPI (Advanced) Features
        - Support for Dynamically Connected (DC) transport protocol
            - Support for pt-to-pt, RMA and collectives
        - Support for Hybrid mode with RC/DC/UD/XRC
        - Support for Core-Direct based Non-blocking collectives
            - Support available for Ibcast, Ibarrier, Iscatter, Igather,
              Ialltoall and Iallgather

    - OpenSHMEM Features
        - Support for RoCE
        - Support for Dynamically Connected (DC) transport protocol

    - UPC Features
        - Based on Berkeley UPC 2.20.2 (contains changes/additions in
          preparation for upcoming UPC 1.3 specification)
        - Support for RoCE
        - Support for Dynamically Connected (DC) transport protocol

    - CAF Features
        - Support for RoCE
        - Support for Dynamically Connected (DC) transport protocol

    - Unified Runtime Features
        - Based on MVAPICH2 2.2a (OFA-IB-CH3 interface). All the runtime
          features enabled by default in OFA-IB-CH3 interface of
          MVAPICH2 2.2a are available in MVAPICH2-X 2.2a
        - The advanced MPI features listed above are available with the
          unified runtime

New features, enhancements and bug Fixes for OSU Micro-Benchmarks
(OMB) 5.0 are listed here.

* New Features & Enhancements (since OMB 4.4.1)

    - Support for a set of non-blocking collectives. The benchmarks
      can display both the amount of time spent in the collectives
      and the amount of overlap achievable
        * osu_iallgather
        * osu_ialltoall
        * osu_ibarrier
        * osu_ibcast
        * osu_igather
        * osu_iscatter
    - Add startup benchmarks to facilitate the ability to measure
      the amount of time it takes for an MPI library to complete MPI_Init
        * osu_init
        * osu_hello
    - Allocate and align data dynamically
        - Thanks to Devendar Bureddy from Mellanox for the suggestion
    - Add options for number of warmup iterations [-x] and number of
      iterations used per message size [-i] to MPI benchmarks
        - Thanks to Devendar Bureddy from Mellanox for the suggestion

* Bug Fixes (since OMB 4.4.1)
    - Do not truncate user specified max memory limits
        - Thanks to Devendar Bureddy from Mellanox for the report and patch

For downloading MVAPICH2 2.2a, MVAPICH2-X 2.2a, OMB 5.0, associated
user guides, quick start guide, and accessing the SVN, please visit
the following URL:

http://mvapich.cse.ohio-state.edu

All questions, feedback, bug reports, hints for performance tuning,
patches and enhancements are welcome. Please post it to the
mvapich-discuss mailing list (mvapich-discuss at cse.ohio-state.edu).

Thanks,

The MVAPICH Team
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.cse.ohio-state.edu/pipermail/mvapich-discuss/attachments/20150819/43a3a9e0/attachment-0001.html>