Developer Reference

  • 2020
  • 09/11/2020
  • Public Content

What's New

Intel® oneAPI Releases
Intel oneAPI Beta 04 – 09
  • No documentation changes specific to Intel oneAPI. For a list of all changes in these releases, refer to the see "Intel® Parallel Studio XE Releases" section below.
Intel oneAPI Beta 03
  • Reworked directory layout:
    • Removed
      intel64/
      .
    • Mpivars.[c]sh
      and
      mpi modulefile
      moved to
      env/
      .
    • Mpivars.[c]sh
      renamed to
      vars.[c]sh
      .
  • Removed deprecated symbolic links.
  • Removed static libraries for debug configurations.
Intel® Parallel Studio XE Releases
Intel MPI Library 2019 Update 8
  • Intel® Xeon® Platinum 9282/9242/9222/9221 family optimizations and platform recognition
  • ILP64 support (see Release Notes for details)
  • impi_info tool extensions (-e|-expert option)
  • NetworkDirect API support (technical preview)
Intel MPI Library 2019 Update 7
  • Added performance optimizations for Intel® Xeon® Platinum 9200 (formerly Cascade Lake-AP). Added debug version of Fortran ILP64 interface with parameters truncation checks.
  • Added IBM* Platform LSF support to Hydra process manager.
  • Added support for IPv6.
Intel MPI Library 2019 Update 6
  • Added
    I_MPI_PMI_VALUE_LENGTH_MAX
    to Other Environment Variables.
  • Added support for non-blocking collectives, more blocking collectives, and
    HCOLL
    collectives to Autotuning.
  • Improved Mellanox* Infiniband* EDR/HDR interconnect support.
  • Improved Amazon* Elastic Fabric Adapter (EFA) support.
  • Added performance optimizations for Intel® Xeon® Platinum 9200 (formerly Cascade Lake-AP).
Intel MPI Library 2019 Update 5
Intel MPI Library 2019 Update 4
  • Added new Autotuning functionality description and environment variables to Environment Variables for Autotuning.
  • Added new variables
    I_MPI_TUNING
    ,
    I_MPI_TUNING_BIN
    , and
    I_MPI_TUNING_BIN_DUMP
    to Tuning Environment Variables.
  • Added arguments for
    I_MPI_PLATFORM
    in Other Environment Variables.
  • Added new
    -tune
    ,
    -hosts-group
    options to Global Options.
  • Added new environment variables
    I_MPI_JOB_STARTUP_TIMEOUT
    ,
    I_MPI_HYDRA_NAMESERVER
    to Hydra Environment Variables
  • Added new transports to
    I_MPI_SHM
    in Shared Memory Control.
  • Removed
    -unmask
    and
    -gumask
    options.
Intel MPI Library 2019 Update 3
Intel MPI Library 2019 Update 2
  • Bug fixes.
Intel MPI Library 2019 Update 1
Intel MPI Library 2019
  • Document overhaul to align with supported functionality.
  • Removed the
    I_MPI_HARD_FINALIZE
    ,
    I_MPI_MIC
    ,
    I_MPI_ENV_PREFIX_LIST
    ,
    I_MPI_TUNE
    *,
    I_MPI_ENV_PREFIX_LIST,I_MPI_JOB_FAST_STARTUP,I_MPI_FALLBACK
    ,
    I_MPI_DAPL
    *,
    I_MPI_LARGE_SCALE_THRESHOLD
    ,
    I_MPI_OFA
    *,
    I_MPI_TCP
    *,
    I_MPI_TMI
    * environment variables.
  • Removed the
    -hostos
    option from Local Options.
  • Added the
    I_MPI_OFI_LIBRARY_INTERNAL
    environment variable to OFI-capable Network Fabrics Control.
  • Added an option for setting
    MPI_UNIVERSE_SIZE
    to Global Options.
  • Added new collective operations to I_MPI_ADJUST Family Environment Variables.
  • Added new variables
    I_MPI_SHM_CELL_EXT_SIZE
    and
    I_MPI_SHM_CELL_EXT_NUM_TOTAL
    to Shared Memory Control.
  • Added impi_info utility.
  • Updated mpitune utility.
Intel MPI Library 2018 Update 3
  • Added new algorithms for
    I_MPI_ADJUST_ALLREDUCE
    to I_MPI_ADJUST Family.
Intel MPI Library 2018 Update 2
  • Improved shm performance with collective operations (
    I_MPI_THREAD_YIELD
    ).
  • Bug fixes .
Intel MPI Library 2018 Update 1
  • Minor changes.
Intel MPI Library 2018
  • Removed support of the Intel® Xeon Phi™ coprocessors (formerly code named Knights Corner)
  • Changes in environment variables:
    • I_MPI_DAPL_TRANSLATION_CACHE
      is now disabled by default
Intel MPI Library 2017 Update 2
  • Added the environment variable
    I_MPI_HARD_FINALIZE
    in Other Environment Variables.
Intel MPI Library 2017 Update 1
  • Topology-aware collective communication algorithms support (I_MPI_ADJUST Family).
  • Added a new algorithm for
    I_MPI_ADJUST_GATHER
    and related environment variable
    I_MPI_ADJUST_GATHER_SEGMENT
    (I_MPI_ADJUST Family).
  • Added the environment variable
    I_MPI_PORT_RANGE
    in Hydra Environment Variables.
Intel MPI Library 2017
  • Document layout changes.

Product and Performance Information

1

Intel's compilers may or may not optimize to the same degree for non-Intel microprocessors for optimizations that are not unique to Intel microprocessors. These optimizations include SSE2, SSE3, and SSSE3 instruction sets and other optimizations. Intel does not guarantee the availability, functionality, or effectiveness of any optimization on microprocessors not manufactured by Intel. Microprocessor-dependent optimizations in this product are intended for use with Intel microprocessors. Certain optimizations not specific to Intel microarchitecture are reserved for Intel microprocessors. Please refer to the applicable product User and Reference Guides for more information regarding the specific instruction sets covered by this notice.

Notice revision #20110804