PARDISO Schur Complement BUG
Here's the following sparse CSR matrix.n = 10ia = {0, 3, 6, 9, 12, 15, 18, 19, 20, 21, 22 };ja = { 4, 7, 8, 0, 5, 7, 1, 6, 8, 0, 2, 7, 1, 2, 8, 2, 3, 4, 3, 5, 6, 4 };a = { -1, 1, 1, 1, 1, 5.5, 1, 1, 4,...
View Articlepossible documentation error for initialization of Philox4x32-10 BRNG in MKL VSL
There seems to be a discrepancy between the method of initialization of the Philox-4x32-10 random number generator in the following two web-pages when more than a single 32-bit integer is used for...
View ArticleSPMM fails when small matrix multiply big matrix
Hello,I am trying to multiply two matrixs.Matrix A : 1 row & 7180 columns & 1341 elements.Matrix B : 7180 rows & 10001 columns & 372623 elements.But segment fault occurs when calling...
View Articlemkl_sparse_convert_csr doesn't change rows & cols
Hello, I am using mkl_sparse_convert_csr to perform transpose function by calling with SPARSE_OPERATION_TRANSPOSE option.However, the values of the result matrix is transposed correctly with the row...
View ArticleWhy is AVX512 not autodetected and used in MKL?
HiI have binary build Intel C 19.0.1 that is a heavy user of MKL. When I doexport MKL_ENABLE_INSTRUCTIONS=AVX512time goes from 88s to 55s for 1 problem instance.That is HUGE savnings. How come MKL...
View ArticleInvalid floating when calling DGEQRF
I am using the DGEQRF subroutine to implement an orthogonalization using the QR factorization. Compiling and running the code using gfortan and MKL works well but when I try ifort and MKL I get the...
View Articlefloating invalid calling DGEQRF with IFORT and MKL
I have implemented an orthogonalization method using the DGEQRF subroutine to compute a QR factorization. The compilation and subsequently execution works well using gfortran/MKL but when I compile...
View Articlepardiso performance drop when decreasing the number of right hand sides
Hi thereI struggle a bit to understand (circumvent if possible) the pardiso performance drop measured as "seconds per right hand side" when decreasing the number of right hand sides.The sparse matrix...
View ArticleAny long integer packages ?
I need to work with REALLY LARGE integers. What would be ideal is something that allows any length of integer. The basic operations on those, of course. + - / and * Has anyone developed those ?
View ArticleHow does MKL Parallelize Operations?
I was wondering are individual mkl operation calls parallelized or do the separate mkl calls run in parallel? e.g. I have code similar to thisfor (int i=0; i<N; i++) mkl_dot_product_call()Is...
View ArticleHelp!
I have developed Dinrhiw2 machine learning library and neuromarketing software using MKL math library. I have M. Sc diploma (TKK.fi).People in Finland want to murder me and talks about torture.Police /...
View ArticleClik here to view.
Different results getrf/getrs, dss and intel pardiso
Dear all,I have a small nonsymmetric linear system that is represented by a matrix in csr format (file fort.106). The task is to solve the system. To this end, I applied three different approaches. At...
View ArticleIntel MKL ERROR: Parameter 4 was incorrect on entry to DSTEIN2
Hi, Some users are seeing the following error messages when running VASP, built against Intel 2018.0.3.Intel MKL ERROR: Parameter 4 was incorrect on entry to DSTEIN2.Intel MKL ERROR: Parameter 6 was...
View ArticlePerformance characteristics of cblas_gemm_s16s16s32
Hi,I'm interested to get more details on the performance characteristics of the function cblas_gemm_s16s16s32. In my application, the performance gain over cblas_sgemm is lower than I would hope.Here...
View ArticleRandom Number Generator
Hi,I have developed a function to generate a vector of random numbers. However, there are two problems:1. When the number of random numbers to be generated is too large (e.g. 50000^2), it will output:...
View ArticlePackage script bug: link_install.sh and tr -s [:blank:]
Having installed the package intel-comp-l-all-vars-19.0.1-144 via apt on Ubuntu, I've run into a problem with the script at /opt/intel/compilers_and_libraries_2019.1.144/linux/bin/link_install.sh which...
View Articlemkl/blas routine for C=AA'B
Hi there,is there any mkl/blas function which performs the operation C=AA'B in on go. Currently I use an intermediate array T and dgemm: T=A'B;C=AT'. I am wondering whether there is a more efficient...
View Articleeigen value and eigen vector
I'm working on quantum structures, so I need to calculate eigenvalues and eigenvalues in a very precise drawing, I want to work with values such as n = 10000 and above. Mkl library that I use it is...
View Articlemkl(dgemm) performance problems on "superlarge" processors
Hi,I was running two subsequent dgemm operations: T=AB and C=A'T with A=(56,000x400,000), B=(400,000x30), T=(56,000x30) and C=B.Conditional on the CPU I measured these wall clock times (for the dgemm...
View ArticleEingenvalue solver dfeast_syev does not find values (info=1)
I am trying to use dfeast_syev to find eigenvectors of 4x4 matrix. I used FEAST, since I'be found that other methods give incorrect values. I call the routine with the following...
View Article