==== Intel MKL ====


Intel Math Kernel Library (Intel MKL) is a library of optimized math routines for science, engineering, and financial applications. Core math functions include BLAS, LAPACK, ScaLAPACK, sparse solvers, fast Fourier transforms, and vector math. The routines in MKL are hand-optimized specifically for Intel processors.

**Static Libraries**
^File^Contents^
|libmkl_intel_lp64.a  | LP64 interface library for the Intel compilers |
|libmkl_intel_ilp64.a | ILP64 interface library for the Intel compilers |
|libmkl_blas95_lp64.a | Fortran 95 interface library for BLAS and Intel® Fortran compiler. Supports the LP64 interface.|
|libmkl_blas95_ilp64.a| Fortran 95 interface library for BLAS and Intel Fortran compiler. Supports the ILP64 interface.|
|libmkl_lapack95_lp64.a|Fortran 95 interface library for LAPACK and Intel Fortran compiler. Supports the LP64 interface.|
|libmkl_lapack95_ilp64.a|Fortran 95 interface library for LAPACK and Intel Fortran compiler. Supports the ILP64 interface.|
|libmkl_intel_thread.a|OpenMP threading library for the Intel compilers|
|libmkl_tbb_thread.a|Intel TBB threading library for the Intel compilers|
|libmkl_sequential.a|Sequential library|
|libmkl_core.a|Core computation library|
|libmkl_scalapack_lp64.a|Static library with LP64 versions of ScaLAPACK routines|
|libmkl_scalapack_ilp64.a|Static library with ILP64 versions of ScaLAPACK routines|
|libmkl_cdft_core.a|Static library with cluster FFT functions|
|libmkl_blacs_intelmpi_lp64.a|Static library with LP64 versions of BLACS routines for Intel MPI|
|libmkl_blacs_intelmpi_ilp64.a|Static library with ILP64 versions of BLACS routines for Intel MPI|

**Dynamic Libraries**
^File^Contents^
|libmkl_rt.so|Single Dynamic Library|
|libmkl_ao_worker.so|The Intel® MIC Architecture library to implement the Automatic Offload mode|
|libmkl_intel_lp64.so|LP64 interface library for the Intel compilers|
|libmkl_intel_ilp64.so|ILP64 interface library for the Intel compilers|
|libmkl_intel_thread.so|OpenMP threading library for the Intel compilers|
|libmkl_tbb_thread.so|Intel TBB threading library for the Intel compilers|
|libmkl_sequential.so|Sequential library|
|libmkl_core.so|Core computation library|
|libmkl_scalapack_lp64.so|Dynamic library with LP64 versions of ScaLAPACK routines|
|libmkl_scalapack_ilp64.so|Dynamic library with ILP64 versions of ScaLAPACK routines|
|libmkl_cdft_core.so|Dynamic library with cluster FFT functions|
|libmkl_blacs_intelmpi_lp64.so|Dynamic library with LP64 versions of BLACS routines for Intel MPI|
|libmkl_blacs_intelmpi_ilp64.so|Dynamic library with ILP64 versions of BLACS routines for Intel MPI|


===Usage on AHPCC clusters===

To load the Intel MKL library into your environment use the **''module''** command.
<code>
tres-l2:pwolinsk:$ module load mkl
</code>
This sets the the following environmental variables in your shell: ''MKLROOT'', ''LD_LIBRARY_PATH'', ''INCLUDE''.
<code>
/storage/apps/intel/composer_xe_2013_sp1.3.174/composer_xe_2013_sp1.3.174/mkl
tres-l2:pwolinsk:$ echo $LD_LIBRARY_PATH 
/storage/apps/intel/composer_xe_2013_sp1.3.174/composer_xe_2013_sp1.3.174/mkl/lib/intel64
tres-l2:pwolinsk:$ echo $INCLUDE 
/storage/apps/intel/composer_xe_2013_sp1.3.174/composer_xe_2013_sp1.3.174/mkl/include
tres-l2:pwolinsk:$ 
</code>

If you are using an Intel compiler to link your code against mkl libraries (''module load intel'') please refer to the [[https://software.intel.com/en-us/articles/intel-mkl-link-line-advisor|Intel MKL link advisor]] page.