affinity mask aligning data, example architecture support Automatic Offload Mode, concept BLAS    Calling LAPACK, BLAS, and CBLAS Routines from C/C++ Language Environments    Fortran 95 Interfaces to LAPACK and BLAS    Threaded Functions and Problems    calling routines from C    Fortran 95 interface to    threaded routines C interface to LAPACK, use of C, calling LAPACK, BLAS, CBLAS from C/C++, Intel® MKL complex types calling    Calling BLAS Functions that Return the Complex Values in C/C++ Code    Calling LAPACK, BLAS, and CBLAS Routines from C/C++ Language Environments    BLAS functions from C    CBLAS interface from C    complex BLAS Level 1 function from C    complex BLAS Level 1 function from C++    Fortran-style routines from C CBLAS interface, use of Cluster FFT, linking with    Linking with ScaLAPACK and Cluster FFTs    on Intel® Xeon Phi™ coprocessors cluster software, Intel® MKL cluster software, linking with    Linking with ScaLAPACK and Cluster FFTs    Examples for Linking with ScaLAPACK and Cluster FFT    Linking with ScaLAPACK and Cluster FFTs on Intel Xeon Phi Coprocessors    commands    linking examples    on Intel® Xeon Phi™ coprocessors code examples, use of coding    Coding Tips    Coding Techniques    data alignment    techniques to improve performance compilation, Intel® MKL version-dependent Compiler Assisted Offload    Examples of Compiler Assisted Offload    examples compiler run-time libraries, linking with compiler-dependent function complex types in C and C++, Intel® MKL computation results, consistency computational libraries, linking with computations offload to Intel® Xeon Phi™ coprocessors    Automatic Offload    Examples of Compiler Assisted Offload    Using Automatic Offload and Compiler Assisted Offload in One Application    Using Intel® Math Kernel Library on Intel® Xeon Phi™ Coprocessors    automatic    compiler assisted, examples    interoperability of offload techniques    techniques conditional compilation configuring Eclipse* CDT    Configuring the Eclipse* IDE CDT to Link with Intel MKL    Configuring Your Integrated Development Environment to Link with Intel® Math Kernel Library consistent results conventions, notational custom shared object    Building Custom Shared Objects    Composing a List of Functions    Specifying Function Names    building    composing list of functions    specifying function names data alignment, example denormal number, performance directory structure    Contents of the Documentation Directories    High-level Directory Structure    Appendix C: Directory Structure in Detail    documentation    high-level    in-detail documentation    Contents of the Documentation Directories    directories, contents Eclipse* CDT    Configuring the Eclipse* IDE CDT to Link with Intel MKL    Configuring Your Integrated Development Environment to Link with Intel® Math Kernel Library    configuring       Configuring the Eclipse* IDE CDT to Link with Intel MKL       Configuring Your Integrated Development Environment to Link with Intel® Math Kernel Library Enter index keyword environment variables, setting examples, linking    Examples for Linking with ScaLAPACK and Cluster FFT    Linking Examples    Linking on Intel Xeon Phi Coprocessors    for cluster software    general    on Intel® Xeon Phi™ coprocessors FFT interface    Coding Techniques    FFT Optimized Radices    Threaded Functions and Problems    data alignment    optimised radices    threaded problems FFTW interface support Fortran 95 interface libraries header files, Intel® MKL HT technology, configuration tip hybrid, version, of MP LINPACK ILP64 programming, support for include files, Intel® MKL installation, checking Intel® Hyper-Threading Technology, configuration tip Intel® Many Integrated Core Architecture    Linking on Intel Xeon Phi Coprocessors    Improving Performance on Intel Xeon Phi Coprocessors    Using Intel® Math Kernel Library on Intel® Xeon Phi™ Coprocessors    examples of linking on coprocessors, for Compiler-Assisted Offload    improving performance on    Intel® MKL usage on Intel® Xeon Phi™ coprocessors    Using Intel® Math Kernel Library on Intel® Xeon Phi™ Coprocessors    Linking on Intel Xeon Phi Coprocessors    Improving Performance on Intel Xeon Phi Coprocessors    Intel® MKL usage on    linking on    performance tips for Intel® Xeon Phi™ coprocessors    Concept of a Native Run    running Intel® Math Kernel Library in native mode on interface    Linking with Fortran 95 Interface Libraries    Using the ILP64 Interface vs. LP64 Interface    Fortran 95, libraries    LP64 and ILP64, use of interface libraries and modules, Intel® MKL interface libraries, linking with Java* examples language interfaces support language-specific interfaces    Using Language-Specific Interfaces with Intel® Math Kernel Library    interface libraries and modules LAPACK    Calling LAPACK, BLAS, and CBLAS Routines from C/C++ Language Environments    Fortran 95 Interfaces to LAPACK and BLAS    Coding Techniques    Threaded Functions and Problems    C interface to, use of    calling routines from C    Fortran 95 interface to    performance of packed routines    threaded routines layers, Intel® MKL structure libraries to link with    Selecting Libraries to Link with    computational    interface    run-time    system libraries    threading link tool, command line link-line syntax linking    Linking on Intel Xeon Phi Coprocessors    on Intel® Xeon Phi™ coprocessors linking examples    Examples for Linking with ScaLAPACK and Cluster FFT    Linking Examples    cluster software    general linking with    Linking with Compiler Run-time Libraries    Linking with Computational Libraries    Linking with Interface Libraries    Linking with System Libraries    Selecting the Threading Layer    compiler run-time libraries    computational libraries    interface libraries    system libraries    threading libraries linking, quick start linking, Web-based advisor LINPACK benchmark memory functions, redefining memory management memory renaming mixed-language programming module, Fortran 95 MP LINPACK benchmark multi-core performance native run, of Intel® Math Kernel Library on Intel® Xeon Phi™ coprocessors notational conventions number of threads    Changing the Number of Threads at Run Time    Setting the Number of Threads Using an OpenMP* Environment Variable    Threading Behavior of Intel MKL on Intel MIC Architecture    MKL_DYNAMIC    Setting the Number of Threads    Techniques to Set the Number of Threads    changing at run time    changing with OpenMP* environment variable    for Intel® Math Kernel Library on Intel® Xeon Phi™ coprocessors    Intel® MKL choice, particular cases    setting for cluster    techniques to set numerically reproducible results offload techniques, interoperability parallel performance parallelism, of Intel® MKL performance    Managing Multi-core Performance    Improving Performance on Intel Xeon Phi Coprocessors    Operating on Denormals    multi-core    on Intel® Many Integrated Core Architecture    with denormals    with subnormals results, consistent, obtaining results, numerically reproducible, obtaining ScaLAPACK, linking with    Linking with ScaLAPACK and Cluster FFTs    on Intel® Xeon Phi™ coprocessors SDL    Using the Single Dynamic Library    Dynamically Selecting the Interface and Threading Layer Segment Of Interest FFT, low-communication algorithm for Cluster FFT sequential mode of Intel® MKL Single Dynamic Library    Using the Single Dynamic Library    Dynamically Selecting the Interface and Threading Layer SOI FFT, low-communication algorithm for Cluster FFT structure    High-level Directory Structure    Appendix C: Directory Structure in Detail    Layered Model Concept    high-level    in-detail    model support, technical supported architectures syntax, link-line system libraries, linking with technical support thread safety, of Intel® MKL threaded functions threaded problems threading control, Intel® MKL-specific threading libraries, linking with uBLAS, matrix-matrix multiplication, substitution with Intel MKL functions unstable output, getting rid of usage information