Loss of performance for DGEMM with matrix dimensions ~1500 using MKL 11.1
Hello,I'm using DGEMM from MKL 11.1 and Intel Composer XE 14.0.0.080 for benchmarking purposes. My system is a Intel Core i5-2500 and I run Debian GNU/Linux 64 bits. Attached I send the plots showing...
View ArticleParallel Studio XE/MKL Installation on Linux Mint 15
Hi,I'm trying installing the MKL package as part of the parallel studio xe suite or by itself. When trying both routes, the installer says that I'm missing a number of optional prerequisites because...
View ArticleMKL sgemm executed serial
Hi people,I am struggling with the threading of SGEMM in MKL called from C++. I am working on a rather big software that is parallelized with OpenMP. In one of the functions, I first set up matrices A...
View ArticleUbuntu + gcc-4.8 problems
Hi,I have two sets of computers. One has Ubuntu 12.04, the other one 13.10. The former works fine with Intel MKL 11.0, the later gives some weird link errors. The command line is the same on both...
View Articlemkl_rt has no way to specify fortran interface
I am attempting to compile the python libraries numpy and scipy using the GNU compiler and using MKL for the blas library. Explicitly linking to the appropriate fortran interface library using...
View ArticleSymbol Lookup Error After Linking MKL with GCC
I am attempting to compile and link the following program with gcc.
View ArticleNonlinear optimizer
Does MKL have non linear optimizer functions with boundary constraints?I have a function(foo) that reads in parameters and calculates certain values. I have another function(foo2) that calculates the...
View Articleopenmp & mkl: getri inside $omp do trouble
Hello!I'm using IVF Compiler 11.0.066 and MKL 10.1So, I have this subroutine which is called from parallel $omp do cycle. In this subroutine I calculate inverse of a matrix, like this: call...
View ArticleBig performance differences of matrix inversion under windows and linux
I have one question about the performance of mkl functions under different platform, windows and linux. I wrote a small program using mkl to do matrix inversion. The...
View ArticleMemory growing when using MKL DFT descriptor in a C++ class
Hi,I've recently noticed an issue when using a MKL DFT_DESCRIPTOR_HANDLE in a C++ class where the memory usage continually increases as I create and dereference objects of the class. I believe I may...
View ArticleMKL license question
hiMy team would like to use the Intel MKL library, notably for its integration with Python-Numpy.What kind of license must be purchased for this case ? e.g. :- we would only need the 'precompiled'...
View Articledtrnlsp_init gives an error in 64 bit mode
The dtrnlsp_init method returns a 1503 or a 1502 with the following pseudo code:_TRNSP_HANDLE_t handle = nullptr; float64 epsilon[6] = {1e-5,1e-5,1e-5,1e-5,1e-5,1e-5}; float64 io_x1[2] = {1, 0};...
View ArticlePARDISO OOC: Value of MKL_PARDISO_OOC_MAX_CORE_SIZE
Is it possible to know the correct value of MKL_PARDISO_OOC_MAX_CORE_SIZE to factorize a matrix in OOC mode? I have seen the suggested value only after an error, and not from the analysis stage. Having...
View Article11.1 update 2 release timeframe?
Will MKL 11.1 Update 2 be released soon? I'm in the process of updating to 11.1.1, and would like to wait for 11.1.2 if it will be available soon.Thank you.
View ArticleMac OS X 10.9.1 + MPICH2/3 + Intel Fortran 14.0.1 + Intel MKL 11.1 cause...
Earlier versions of Intel MKL (such as 10.x) work perfectly fine. I haven't found a single version of MPICH that does not crash the following code. The command line used to compile the code ismpif90...
View Articlelapacke_dpptrf could contain a bug
Hello, using the intel MKL 11.1 for windows, i discovered what seems a bug in LAPACKE C interface to Lapack: calling LAPACKE_dpptrf (Cholesky decomposition) with a packed LT row major matrix does not...
View Articlecblas_dscal metdod
Hello,I implemented below code segment to compare "memset", "for loop" and "cblas_dscal (Intel MKL)". The test results are as:memset: G0 = 0.000000 6954 nsfor: G0 = 1.000000 38741 nsMKL: G0 = 2.000000...
View Articlemex and OSX, symbol not found for libmkl_intel_thread.dylib library
Hi,I have the following problem. My OSI compile a mex file with the following command:/Applications/MATLAB_R2013b.app//bin/mex myfile.cpp mylib.a \ CXXFLAGS="\$CXXFLAGS -pthread -fPIC -DNDEBUG...
View ArticleHow do I get the current number of MKL threads
Hi, I'm trying to make a section of code that runs with a certain number of MKL threads without affecting the number of MKL threads used in other sections of the code. To do this I want to store the...
View ArticleCan you use version 10.2 with an 11 license?
I'm using MKL 10.2 and and need a new license for another developer. I don't want to use version 11 yet for various reasons. If I buy a new license, can it be used for 10.2 while still allowing me to...
View Article