Is mkl speed dependend upon how contiguous memory is?
Does the speed of the mkl blas/lapack library routines change significantly when one has contiguous memory versus not?I have a strange problem that looks like a "Memory Cache Leak" (not a memory leak)...
View ArticleCan PARDISO routines exploit an initial guess?
Hi, I'm using PARDISO to find vector x in the equation Ax = b Where A is a real unsymmetric matrix. In the PARDISO manual, they mention the vector x is only accessed in the solution phase. If I...
View ArticleMeeting Access Conflict when trying to use mkl_sparse_convert_csr
Hi, has anyone encounterd the situation I met? All I did was try to convert a sparse matrix in CSC format to CSR format.At the foremost , I use mkl_sparse_z_creat_csc to create a matrix in CSC format,...
View ArticlePARDISO - different time in factorization step
Hi allWe use the PARDISO solver (Parallel Studio 2018) in our FEM code for a symmetric indefinite system and discovered that the time spent in the factorization step may vary significantly. We use the...
View ArticleIntel MKL FEAST
HiI am using feast extremel eigenvalue solver (dexample_extremal_ev_c1.c). for finding lowest eigenvalues. My matrix size is big 1 million cross 1 million. There are around 31 non zero values in each...
View ArticleLooking for a library or package that can read and write sparse matrices
When Intel visual Fortran call Pardiso in MKL to solve a linear equation group, you need to read the sparse matrix from the file. But there are many ways to store sparse matrices in files (such as:...
View ArticleSchur complement for asymmetric matrix
Hi all,I am using mkl pardiso to do Schur complement. So far I got it working for symmetric matrix. But for asymmetric matrix, I got crash at phase = 11. The error message is "... Access violation...
View ArticleCan I use MKL_DIRECT_CALL JIT on a Core i5 CPU?
The MKL Developer's Guide says JIT can be used on Xeon processor with AVX2, but can it be used on a Core i5 CPU with AVX2?
View ArticleWhat LAPACK function is available to calculate 2-norm of a matrix?
What LAPACK function is available to calculate 2-norm (or spectral norm) of a matrix? Thank you.
View ArticleQuestions & Puzzled about mkl_sparse_spmm
Hi, I`m very very very anxious now because of the using of mkl_sparse_spmm function always feedback me an error beyond my understanding. When I trying to compute the product of two sparse matrix...
View ArticleIs there any service function that helps us form the needed matrix storage...
Although the various storage schemes are explained in detail in MKL Reference Manual, I believe human users are still willing to present their matrix in conventional full storage form. Is there any...
View Articleusing MKL's fft in fixed point dsp codes
Hi,We've started exploring MKL's FFT in our dsp programs. We were able to set up FFT and get correct output. One issue we have now is that: the data flows before and after FFT are all in fixed point,...
View ArticleFortran MKL matrix operations in pure CSR format
Hi folks, I'm an enthusiastic Intel-MKL user employing parallel Fortran and C code. I need to perform specificCSR matrix-dot-matrix operations like MKL_SPARSE_SYRK without resortingto dense matrices at...
View Articlememory leakage(内存泄漏）
vslsConvNewTask1D will add 1M memory when running in first time，vslConvDeleteTask(&t) can not free the memory, why?(vslsConvNewTask1D 这个函数会增加1M左右的内存，运行vslConvDeleteTask后并不能释放增加的内存，请问为什么？）
View ArticleIntel® MKL version 2019 Update 1 is now available
Intel® Math Kernel Library (Intel® MKL) is a highly optimized, extensively threaded, and thread-safe library of mathematical functions for engineering, scientific, and financial applications that...
View ArticleProblem with Compiling HPL using MKL&MPI
Hello everyone~I've installed Intel MKL & Intel MPI for linux to run HPL, however, I have problem while compiling it. I use virtual machine to run CentOS, imformation of my computer and virtual...
View Article[icpc Warning]: Specifying -lm before files may supersede the Intel(R) math...
I just notice this warning while compiling my C++ code. My sIntel version is compilers_and_libraries_2018.3.222My compile options from the MKL advisor is in MKL advisorThen it will say a...
View ArticleSum along specific matrix axis
I am working on a project where I want to accelerate numpy element-wise multiplication and sum.What I am doing is transfer the numpy array to C pointer and use MKL function to accelerate them(through...
View ArticleClik here to view.
Strange bus error in Pardiso
Hi, all !I just met a strange bus error while using Pardiso. The following is the detailed information.First, the purpose of my code: As I study finite element method (FEM), I would like to use Pardiso...
View ArticlePARDISO - huge memory requirement, slow calculation
I implemented PARDISO solver to solve a linear system of equations with a sparse matrix. I tested it using a model that yields 1 million equations, each of them containing typically 7 nonzero...
View Article