Intel® MPI Library 4.0 Update 3 for Windows* OS
Release Notes

Contents

Overview
What's New
Key Features
System Requirements
Installation Notes
Special Features and Known Issues
Documentation
Technical Support
Copyright and Licenses
Disclaimer and Legal Information

Overview

The Intel® MPI Library for Windows* OS is a multi-fabric message passing library based on ANL* MPICH2* and OSU* MVAPICH2*.

The Intel® MPI Library for Windows* OS implements the Message Passing Interface, version 2 (MPI-2) specification.

To receive technical support and updates, you need to register your Intel® Software Development Product. See section Technical Support.

Product Contents

The Intel® MPI Library Runtime Environment (RTO) contains the tools you need to run programs including SMPD services and supporting utilities, dynamic libraries, and documentation.

The Intel® MPI Library Development Kit (SDK) includes all of the Runtime Environment components plus include files and modules, interface libraries, debug libraries and test codes.

Related Products and Services

Information on Intel® Software Development Products is available at http://www.intel.com/software/products.

Some of the related products include:

- The Intel® Software College provides training for developers on leading-edge software development technologies. Training consists of online and instructor-led courses covering all Intel® architectures, platforms, tools, and technologies.

What's New

The Intel® MPI Library 4.0 Update 3 for Windows* OS is an update release of the Intel® MPI Library for Windows* OS. This release includes the following updates compared to the Intel® MPI Library 4.0 Update 2 (see product documentation for more details):

• Performance and scalability improvements
  • Shared memory optimizations for platforms with Intel® Streaming SIMD Extensions 4.2 (Intel® SSE4.2) and Intel® AES New Instructions (Intel® AES-NI). This functionality is available for both Intel® and non-Intel microprocessors, but it may perform additional optimizations for Intel microprocessors than it performs for non-Intel microprocessors.
  • Dynamic connection mode for shared memory
  • Accelerated RDMA memory registration cache
  • Substantially accelerated and enhanced MPI tuning utility
• Usability improvements
  • Additional integrated performance monitoring (IPM) summary format
  • Extended debugging output control
  • Enhanced processor information utility cpuinfo
  • Bug fixes
• Extended interoperability
  • Intel® Composer XE 2011 Update 6 support

The Intel® MPI Library 4.0 Update 2 for Windows* OS is an update release of the Intel® MPI Library for Windows* OS. This release includes the following updates compared to the Intel® MPI Library 4.0 Update 1 (see product documentation for more details):

• Usability improvements
  • Secure DLL Loading mode. See more details in Special Features and Known Issues
  • Improved simple multi-purpose daemon (SMPD) and mpiexec
  • Signed Intel® MPI Library binaries
  • Improved static DAPL connections establishment in the wait mode
  • Bug fixes
• Extended interoperability
  • Intel® Composer XE 2011 Update 4 support
  • Ability to call MPI from the Co-Array Fortran programs

   

The Intel® MPI Library 4.0 Update 1 for Windows* OS is an update release of the Intel® MPI Library for Windows* OS. This release includes the following updates compared to the Intel® MPI Library 4.0 (see product documentation for more details):

• Performance and scalability improvements
  • Further optimization to several collective algorithms
• Usability improvements
  • Extended process pinning control for hybrid applications through I_MPI_PIN_DOMAIN and I_MPI_PIN_CELL environment variables
  • Improved mpitune for easier application tuning
• Extended interoperability
  • Intel® Composer XE 12.0 Beta support

   

The Intel® MPI Library 4.0 for Windows* OS is a new release of the Intel® MPI Library for Windows* OS. This release includes the following new features compared to the Intel® MPI Library 3.2 Update 2 (see product documentation for more details):

• New architecture for better performance and higher scalability
  • Optimized shared memory path for industry leading latency and on multicore platforms
  • New flexible mechanism for selecting the communication fabrics (I_MPI_FABRICS) that complements the classic Intel MPI device selection method (I_MPI_DEVICE)
• Updated MPI performance tuner to extract the last ounce of performance out of your installation
  • For a certain cluster, based on the Intel® MPI Benchmarks (IMB) or a user provided benchmark
  • For a certain application run
• MPI 2.1 standard conformance
• Experimental dynamic process support
• Backward compatibility with Intel MPI Library 3.x based applications

Example

Set the I_MPI_FABRICS environment variable to select a particular network fabric.

• To use shared memory for intra-node communication and the DAPL* layer for inter-node communication, do the following steps:

$ mpiexec -n 4 –env I_MPI_FABRICS shm:dapl ./IMB-MPI1

See more details in the Intel® MPI Library for Windows* OS Reference Manual.

Key features

This release of the Intel® MPI Library supports the following major features:

• MPI-1 and MPI-2.1 specification conformance with some limitations. See Special Features and Known Issues
  
• - Support for any combination of the following interconnection fabrics:
  • Shared memory
  • RDMA-capable network fabrics through DAPL*, such as InfiniBand* and Myrinet*
  • Sockets, for example, TCP/IP over Ethernet*, Gigabit Ethernet*, and other interconnects
• (SDK only) Support for IA-32 and Intel® 64 architecture clusters using:
  • Intel® C++ Compiler version 10.1 through 11.1 and higher
  • Intel® Fortran Compiler version 10.1 through 11.1 and higher
  • Microsoft* Visual C++* Compilers
• (SDK only) C, C++, Fortran* 77 and Fortran 90 language bindings
• (SDK only) Dynamic linking

System Requirements

The following sections describe supported hardware and software

Supported Hardware

Systems based on the IA-32 architecture:

• Intel® Pentium® 4 processor or higher
• Intel® Core™ i7 processor recommended
• 1 GB of RAM per core
• 2 GB of RAM per core recommended
• 1 GB of free hard disk space

Systems based on the Intel® 64 architecture:

• Intel® Core™2 processor family or higher
• Intel® Xeon® 5500 processor series recommended
• 1 GB of RAM per core
• 2 GB of RAM per core recommended
• 1 GB of free hard disk space

Supported Software

Operating Systems:

• Systems based on the IA-32 architecture:
  • Microsoft* Windows 7*
• Systems based on the Intel® 64 architecture:
  • Microsoft* Windows* HPC Server 2008
  • Microsoft* Windows* HPC Server 2008 R2
  • Microsoft* Windows* Server 2008
  • Microsoft* Windows* Server 2008 R2
  • Microsoft* Windows 7*

(SDK only) Compilers:

• Intel® C++ Compiler 11.1 for Windows* OS and higher
• Intel® Fortran Compiler 11.1 for Windows* OS and higher
• Microsoft* Visual Studio 2005*
• Microsoft* Visual Studio 2008*
• Microsoft* Visual Studio 2010*
• Microsoft* Visual C++* Compilers

Batch Systems:

• Microsoft* job scheduler
• Altair* PBS Pro* 9.2 and higher

Recommended InfiniBand Software:

• Windows* OpenFabrics* (WinOF*) 2.0 or higher

SUPPORTED LANGUAGES

• For Intel® Professional Edition Compilers: C, C++, Fortran 77, Fortran 90

Installation Notes

Launch the installer and follow the instructions. See Intel® MPI Library for Windows* OS Installation Guide for details.

Special Features and Known Issues

• The Intel® MPI Library 4.0 Update 2 for Windows* OS provides enhanced security options. Use the HKEY_LOCAL_MACHINE\Software\Intel\MPI registry key to define the following registry entries:
  • The SecureDynamicLibraryLoading enables the dynamic library loading enhanced security mode. Set the value to enable|yes|on|1. This option is disabled by default.
  • The I_MPI_DAT_LIBRARY specifies the DAT library to be used in the DLL enhanced security mode.
    NOTE: the I_MPI_DAT_LIBRARY environment variable has no effect in the dynamic library loading the enhanced security mode. See the Intel® MPI Library for Windows* OS Reference Manual for more details on this variable.
• Intel® MPI Library 4.0 for Windows* OS is binary compatible with the majority of Intel MPI Library 3.x-based applications. Recompile your application only if you use:
  • MPI one-sided routines in Fortran (mpi_accumulate(), mpi_alloc_mem(), mpi_get(), mpi_put(), mpi_win_create())
  • MPI C++ binding
• Intel® MPI Library 4.0 for Windows* OS implements the MPI-2.1 standard. The functions of the following MPI routines have changed:
  • MPI_Cart_create()
  • MPI_Cart_map()
  • MPI_Cart_sub()
  • MPI_Graph_create()
  If your application depends on the strict pre-MPI-2.1 behavior, set the environment variable I_MPI_COMPATIBILITY to 3.
• The Intel® MPI Library supports the MPI-2 process model for all fabric combinations except the following case:
  I_MPI_FABRICS is set to <fabric1>:<fabric2>, where <fabric1> is not shm, and <fabric2> is not equal to <fabric1> (for example, dapl:tcp).
• If communication between two existing MPI applications is established using the process attachment mechanism, the library does not control whether the same fabric has been selected for each application. This situation may cause unexpected applications behavior. Set the same I_MPI_FABRICS variable for each application to avoid this issue.
• The following restriction exists for the DAPL-capable network fabrics because it relates to support of the MPI-2 process model: if the size of the information about the host used to establish the communication exceeds a certain DAPL provider value, the application fails with an error message similar to:
  [0:host1][../../dapl_module_util.c:397] error(0x80060028):….: could not connect DAPL endpoints: DAT_INVALID_PARAMETER(DAT_INVALID_ARG5)
• The Intel® MPI Library uses dynamic connection establishment by default for 64 and more processes. To always establish all connections upfront, set the I_MPI_DYNAMIC_CONNECTION environment variable to "disable". This option may improve stability of certain DAPL* providers.
• (SDK only) Provide the msvcr71.dll library to the end user if your product redistributes the mpitune utility.
• The Intel® MPI Library pins processes automatically. Use the I_MPI_PIN and related environment variables to control process pinning. See the Intel® MPI Library for Windows* OS Reference Manual for details.
• The Intel® MPI Library enhances message-passing performance on DAPL*-based interconnects by maintaining a cache of virtual-to-physical address translations in the MPI DAPL* data transfer path. To disable the translation cache completely, set the environment variable I_MPI_RDMA_TRANSLATION_CACHE to "disable".
• (SDK only) The nmake utility does not work correctly if the path to the Intel® MPI Library compiler drivers contains spaces. For instance, C:\Program Files (x86)\Intel\MPI\<version>\bin\. Copy the Intel® MPI Library compiler drivers to another location to avoid this issue.
• An MPI application inherits the current working directory of the mpiexec command at the time of its invocation. Use the -gwdir or -wdir options to override this functionality.
• To run a mixed Intel MPI/OpenMP* application, follow these steps:
  • Use the thread-safe version of the Intel® MPI Library by using the -mt_mpi compiler driver option
  • Set I_MPI_PIN_DOMAIN to select the desired process pinning scheme. The recommended setting is I_MPI_PIN_DOMAIN=auto

  See the Intel® MPI Library for Windows* OS Reference Manual for more details.

• Intel® MKL 10.0 may create multiple threads depending on various conditions. Follow these rules to correctly use Intel® MKL:
  • Use the thread-safe version of the Intel® MPI Library in conjunction with Intel® MKL
  • Set the OMP_NUM_THREADS environment variable to "1" to run the application if linked against the non-thread-safe version of the Intel® MPI Library
• Follow these rules to launch the Intel MPI Library application using the MPICH2* environment:
  • Use the -port mpiexec option to specify the MPICH2* SMPD* port number
  • Set the I_MPI_SMPD_VERSION_CHECK environment variable to "disable"
• Some operating systems have a limitation on the number of concurrent connections. This limitation may cause job failure with the following error message:
  "no more connections can be made to this remote computer at this time because there are already as many connections as the computer can accept".

  To reduce the probability of this issue, decrease the auto disconnect time.

  From the command line do the following:

  net config server /autodisconnect:time_before_autodisconnect

  The default value of the auto disconnect time is usually set to 15 minutes.

• Use the mpiexec -mapall option if your application is located or accesses any network drive mapped to your local system.
• Disable User Account Control (UAC) to use domain-based authorization with the delegation ability.
• The following MPI-2.1 feature is not supported by the Intel® MPI Library:
  • Passive target one-sided communication when the target process does not call any MPI functions

NOTE: Many routines in the libirc.lib library (linked with the Intel® MPI Library) are more optimized for Intel microprocessors than for non-Intel microprocessors.

Documentation

The Intel® MPI Library for Windows* OS Getting Started Guide contains information on the following subjects:

• First steps using the Intel® MPI Library
• First-aid troubleshooting actions

The Intel® MPI Library for Windows* OS Reference Manual contains information on the following subjects:

• Command Reference describes commands, options, and environment variables
• Tuning Reference describes environment variables that influence library behavior and performance

The Intel® MPI Library for Windows* OS Installation Guide contains information on the following subjects:

• Obtaining, installing, and uninstalling the Intel® MPI Library
• Getting technical support

Notation Conventions

Release Notes and user guide documentation use the notation conventions listed in the following table:

Technical Support

Your feedback is very important to us. To receive technical support for the tools provided in this product and technical information including FAQ's and product updates, you need to register for an Intel® Premier Support account at the Registration Center.

NOTE: Registering for support varies for release product or pre-release products (alpha, beta, etc) - only released products have support web pages on http://support.intel.com/.

To register for an account, please visit the Intel® Registration Center web site. If you have forgotten your password, please email a request to: quad.support@intel.com. Please do not email your technical issue to this email address.

The product support web site, http://www.intel.com/software/products/support/mpi, provides top technical issues, frequently asked questions, product documentation, and product errata.

There is an HPC and Intel® Cluster Tools Forum for HPC experts and enthusiasts to share their knowledge, resources, and insights for the advancement of HPC solutions, cluster solutions, and the computing architectures that implement them.

Submitting Issues

Before submitting a support issue, see the Intel® MPI Library for Windows* OS Getting Started Guide for details on post-install testing to ensure that basic facilities are working.

When submitting a support issue to Intel® Premier Support, please provide specific details of your problem, including:

• The Intel® MPI Library package name and version information
• Host architecture
• Compiler(s) and versions
• Operating system(s) and versions
• Specifics on how to reproduce problems. Include makefiles, command lines, small test cases, and build instructions.
  Use /test sources as test cases, when possible.

You can obtain version information for the Intel® MPI Library package in the file mpisupport.txt.

Steps to submit an issue

1. Go to https://premier.intel.com/
2. Log in to the site. Note that your username and password are case-sensitive.
3. Click on the "Submit Issue" link in the left navigation bar.
4. Choose "Development Environment" from the "Product Type" drop-down list.
5. If this is a software or license-related issue, choose "Intel® MPI Library, Windows" from the "Product Name" drop-down list.
6. Enter your question and complete the fields in the windows that follow to successfully submit the issue.

NOTE: Please notify your support representative prior to submitting source code where access needs to be restricted to certain countries to determine if this request can be accommodated.

Copyright and Licenses

The Intel® MPI Library is based on MPICH2* from Argonne National Laboratory* (ANL) and MVAPICH2* from Ohio State University* (OSU).

See the information below for additional licenses of the following 3rd party tools used within the Intel® MPI Library: Python*, Windows Installer XML* (WiX), and AVL Trees*.

Python*
PYTHON SOFTWARE FOUNDATION LICENSE VERSION 2

1. This LICENSE AGREEMENT is between the Python Software Foundation ("PSF"), and the Individual or Organization ("Licensee") accessing and otherwise using this software ("Python") in source or binary form and its associated documentation.
2. Subject to the terms and conditions of this License Agreement, PSF hereby grants Licensee a nonexclusive, royalty-free, world-wide license to reproduce, analyze, test, perform and/or display publicly, prepare derivative works, distribute, and otherwise use Python alone or in any derivative version, provided, however, that PSF's License Agreement and PSF's notice of copyright, for example, "Copyright (c) 2001, 2002, 2003, 2004, 2005, 2006 Python Software Foundation; All Rights Reserved" are retained in Python alone or in any derivative version prepared by Licensee.
3. In the event Licensee prepares a derivative work that is based on or incorporates Python or any part thereof, and wants to make the derivative work available to others as provided herein, then Licensee hereby agrees to include in any such work a brief summary of the changes made to Python.
4. PSF is making Python available to Licensee on an "AS IS" basis. PSF MAKES NO REPRESENTATIONS OR WARRANTIES, EXPRESS OR IMPLIED. BY WAY OF EXAMPLE, BUT NOT LIMITATION, PSF MAKES NO AND DISCLAIMS ANY REPRESENTATION OR WARRANTY OF MERCHANTABILITY OR FITNESS FOR ANY PARTICULAR PURPOSE OR THAT THE USE OF PYTHON WILL NOT INFRINGE ANY THIRD PARTY RIGHTS.
5. PSF SHALL NOT BE LIABLE TO LICENSEE OR ANY OTHER USERS OF PYTHON FOR ANY INCIDENTAL, SPECIAL, OR CONSEQUENTIAL DAMAGES OR LOSS AS A RESULT OF MODIFYING, DISTRIBUTING, OR OTHERWISE USING PYTHON, OR ANY DERIVATIVE THEREOF, EVEN IF ADVISED OF THE POSSIBILITY THEREOF.
6. This License Agreement will automatically terminate upon a material breach of its terms and conditions.
7. Nothing in this License Agreement shall be deemed to create any relationship of agency, partnership, or joint venture between PSF and Licensee. This License Agreement does not grant permission to use PSF trademarks or trade name in a trademark sense to endorse or promote products or services of Licensee, or any third party.
8. By copying, installing or otherwise using Python, Licensee agrees to be bound by the terms and conditions of this License Agreement.

Windows Installer XML* (WiX)
http://www.opensource.org/licenses/cpl1.0.php

AVL Trees*
Copyright (c) 1989-1997 by Brad Appleton, All rights reserved.

This software is not subject to any license of the American Telephone and Telegraph Company or of the Regents of the University of California.
Permission is granted to anyone to use this software for any purpose on any computer system, and to alter it and redistribute it freely, subject to the following restrictions:

1. Neither the authors of the software nor their employers (including any of the employers' subsidiaries and subdivisions) are responsible for maintaining & supporting this software or for any consequences resulting from the use of this software, no matter how awful, even if they arise from flaws in the software.
2. The origin of this software must not be misrepresented, either by explicit claim or by omission. Since few users ever read sources, credits must appear in the documentation.
3. Altered versions must be plainly marked as such, and must not be misrepresented as being the original software. Since few users ever read sources, credits must appear in the documentation.
4. This notice may not be removed or altered.

The Intel MPI library includes altered AVL Trees* source codes.

Disclaimer and Legal Information

INFORMATION IN THIS DOCUMENT IS PROVIDED IN CONNECTION WITH INTEL PRODUCTS. NO LICENSE, EXPRESS OR IMPLIED, BY ESTOPPEL OR OTHERWISE, TO ANY INTELLECTUAL PROPERTY RIGHTS IS GRANTED BY THIS DOCUMENT. EXCEPT AS PROVIDED IN INTEL'S TERMS AND CONDITIONS OF SALE FOR SUCH PRODUCTS, INTEL ASSUMES NO LIABILITY WHATSOEVER, AND INTEL DISCLAIMS ANY EXPRESS OR IMPLIED WARRANTY, RELATING TO SALE AND/OR USE OF INTEL PRODUCTS INCLUDING LIABILITY OR WARRANTIES RELATING TO FITNESS FOR A PARTICULAR PURPOSE, MERCHANTABILITY, OR INFRINGEMENT OF ANY PATENT, COPYRIGHT OR OTHER INTELLECTUAL PROPERTY RIGHT.
UNLESS OTHERWISE AGREED IN WRITING BY INTEL, THE INTEL PRODUCTS ARE NOT DESIGNED NOR INTENDED FOR ANY APPLICATION IN WHICH THE FAILURE OF THE INTEL PRODUCT COULD CREATE A SITUATION WHERE PERSONAL INJURY OR DEATH MAY OCCUR.
Intel may make changes to specifications and product descriptions at any time, without notice. Designers must not rely on the absence or characteristics of any features or instructions marked "reserved" or "undefined." Intel reserves these for future definition and shall have no responsibility whatsoever for conflicts or incompatibilities arising from future changes to them. The information here is subject to change without notice. Do not finalize a design with this information.
The products described in this document may contain design defects or errors known as errata which may cause the product to deviate from published specifications. Current characterized errata are available on request.
Contact your local Intel sales office or your distributor to obtain the latest specifications and before placing your product order.
Copies of documents which have an order number and are referenced in this document, or other Intel literature, may be obtained by calling 1-800-548-4725, or go to:  http://www.intel.com/design/literature.htm

Intel processor numbers are not a measure of performance. Processor numbers differentiate features within each processor family, not across different processor families. Go to:
http://www.intel.com/products/processor_number/

MPEG-1, MPEG-2, MPEG-4, H.261, H.263, H.264, MP3, DV, VC-1, MJPEG, AC3, AAC, G.711, G.722, G.722.1, G.722.2, AMRWB, Extended AMRWB (AMRWB+), G.167, G.168, G.169, G.723.1, G.726, G.728, G.729, G.729.1, GSM AMR, GSM FR are international standards promoted by ISO, IEC, ITU, ETSI, 3GPP and other organizations. Implementations of these standards, or the standard enabled platforms may require licenses from various entities, including Intel Corporation.

BlueMoon, BunnyPeople, Celeron, Celeron Inside, Centrino, Centrino Inside, Cilk, Core Inside, E-GOLD, i960, Intel, the Intel logo, Intel AppUp, Intel Atom, Intel Atom Inside, Intel Core, Intel Inside, Intel Insider, the Intel Inside logo, Intel NetBurst, Intel NetMerge, Intel NetStructure, Intel SingleDriver, Intel SpeedStep, Intel Sponsors of Tomorrow., the Intel Sponsors of Tomorrow. logo, Intel StrataFlash, Intel vPro, Intel XScale, InTru, the InTru logo, the InTru Inside logo, InTru soundmark, Itanium, Itanium Inside, MCS, MMX, Moblin, Pentium, Pentium Inside, Puma, skoool, the skoool logo, SMARTi, Sound Mark, The Creators Project, The Journey Inside, Thunderbolt, Ultrabook, vPro Inside, VTune, Xeon, Xeon Inside, X-GOLD, XMM, X-PMU and XPOSYS are trademarks of Intel Corporation in the U.S. and other countries.

* Other names and brands may be claimed as the property of others.

Microsoft, Windows, Visual Studio, Visual C++, and the Windows logo are trademarks, or registered trademarks of Microsoft Corporation in the United States and/or other countries.

Java is a registered trademark of Oracle and/or its affiliates.

Copyright (C) [2003]–[2011], Intel Corporation. All rights reserved.