- Infos im HLRS Wiki sind nicht rechtsverbindlich und ohne Gewähr -
- Information contained in the HLRS Wiki is not legally binding and HLRS is not responsible for any damages that might result from its use -

MKL: Difference between revisions

From HLRS Platforms
Jump to navigationJump to search
No edit summary
No edit summary
Line 9: Line 9:
}}
}}


== Fortran 95 Interfaces and Wrappers for LAPACK, BLAS, etc. ==
=== Using MKL on Nehalem cluster ===
The MKL fortran interfaces must be build seperately because different compiler return complex values differently. Due to the large number of combinations occuring from this the user has to do this on his own.
This example shows the basic steps when using MKL.


{{Command| command =<nowiki>
Load the necessary module. For example:
&#35; example how to build a fortran interface of the intel MKL
{{Command| command =
&#35; set up your environment
module load compiler/intel  
module load numlib/intel # set up MKL environment
module load numlib/intel/mkl # set up MKL environment
module load compiler/intel
module load mpi/impi # if MPI needed  
module load mpi/impi # MPI needed by some fftw interfaces
&#35; switch into the directory of the needed interface
cd $MKLROOT/interfaces/<interfacename>
&#35; display build options like compiler and mpi library
make
&#35; install the library into $HOME/lib
&#35; By default, for interfaces without MPI the ifort compiler is used.
&#35; You can change it by an additional parameter FC=<compiler>.
make INSTALL_DIR=$HOME/lib libem64t [OPTIONS]</nowiki>
}}
}}


=== Example ===
=== Example ===
Here is a simple example how to compile an application with the Fortran 95 interface and wrappers for BLAS95. You should have build the BLAS95 Interfaces already as described above.


{{File|filename = example_blas.f90
{{File|filename = example_blas.f90
Line 58: Line 48:
}}
}}


Compile this with
=== Compilation example ===
 
{{Command| command =
{{Command| command =
ifort example_blas.f90 -L$HOME/lib -lmkl_blas95 $MKLPATH/libmkl_intel_lp64.a -Wl,
ifort -c example_blas.f90 -o example_blas.o
: --start-group $MKLPATH/libmkl_intel_thread.a $MKLPATH/libmkl_core.a -Wl,--end-group <nowiki></nowiki>
ifort -L$MKLROOT/lib/intel64 -lmkl_intel_lp64 -lmkl_intel_thread -lmkl_core -openmp -lpthread  example_blas.o
: -L$MKLPATH -liomp5 -o example_blas
}}
 
To see what libraries are recommended for a particular use case, visit:
 
* [http://software.intel.com/en-us/articles/intel-mkl-link-line-advisor/ Intel® MKL Link Line Advisor]


}}
== See also ==
== See also ==
* [[Software Development Tools, Compilers & Libraries]]
* [[Software Development Tools, Compilers & Libraries]]
Line 70: Line 64:
== External links ==
== External links ==
* [http://software.intel.com/en-us/intel-mkl/ Intel® MKL homepage]
* [http://software.intel.com/en-us/intel-mkl/ Intel® MKL homepage]
* [http://software.intel.com/en-us/articles/intel-mkl-link-line-advisor/ Intel® MKL Link Line Advisor]


[[Category:Numerical Library]]
[[Category:Numerical Library]]

Revision as of 11:16, 2 May 2013

Intel® Math Kernel Library (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.
Intel-logo.png
Developer: Intel
Platforms: NEC Nehalem Cluster
Category: Numerical Library
License: Commercial
Website: Intel® MKL homepage


Using MKL on Nehalem cluster

This example shows the basic steps when using MKL.

Load the necessary module. For example:

module load compiler/intel

module load numlib/intel/mkl # set up MKL environment

module load mpi/impi # if MPI needed


Example

File: example_blas.f90
 program example_blas
   implicit none
   real, dimension(3) :: a = (/2.,1.,-1./), b = (/5., -2., 1.5/)
   real               :: c, sdot
   real(kind=8)       :: d, dnrm2
   ! *** skalar product ***
   ! sdot: s ... REAL, dot ... skalar product
   ! 1. Argument  ... dimensions of the vectors
   ! 2. und 4. A. ... the vectors
   ! 3. und 5. A. ... increment (here 1)
   c = sdot(3, a, 1, b, 1)
   write(*,*) c
   ! Result: 6.500000
   ! *** norm of the vectors ***
   ! dnrm2: d ... DOUBLE PRECISION, nrm2 ... (euklidian) norm
   ! 1. Argument: dimensions of the vectors
   ! 2. A.:       vektor
   ! 3. A.:       increment (here 1)
   d = dnrm2(3, dble(a), 1)
   write(*,*) d
   ! Result: 2.44948974278318
 end program example_blas


Compilation example

ifort -c example_blas.f90 -o example_blas.o ifort -L$MKLROOT/lib/intel64 -lmkl_intel_lp64 -lmkl_intel_thread -lmkl_core -openmp -lpthread example_blas.o


To see what libraries are recommended for a particular use case, visit:

See also

External links