cluster_sparse_solver Schur complement matrix - can't get it to work
Hi all,Has anyone had luck getting the new Schur Complement matrix option (iparm(36)) to work with MKL 2018 release 2? I am not having luck, and can't find any example programs that use this option. If...
View ArticleMPICH3 and libmkl_blacs_mpich_
While attempting to run siesta-4.1-b3 using MPICH3 (3.1.4) with intel mkl 2018 libmkl_blacs_mpich I received the following error:[Jesses-iMac-2:98886] *** An error occurred in...
View ArticleClik here to view.
Double Precision SVDs returns incorrect (and inconsistent) results for some...
Hi,We upgraded our MKL a couple of months ago from 11.3.3 and have been plagued by inconsistent results in some runs of our software. For most things we analyze, results are good, but for some small...
View ArticleMultithreading with MKl Performance Drop
Hi all, Im first time user of MKL library and I thought a good place for me to get the hang of it is to replicate the results on this intel blog...
View Articlecblas_zsyrk returns wrong results
I have the following function and I'm trying to compute C = A*A' using zsyrk from blas, but when I print the C matrix the results are wrong. What am I missing?double* zsyrk(int N, double *A) { int i,j;...
View ArticleMKL License
Hi,Some intel pages mention a commercial license and evaluation/trial period for MKL, while others say that commercial use is simply allowed. Can you please clarify? Thanks!
View Articlepardiso transpose solve vs non-transpose solve
Hello,I am wondering what differences to expect between loading in a CSR3 matrix and doing a solve, and loading in a CSC3 matrix and doing a transpose solve. The ia and ja matrices are swapped...
View ArticleLink faiure
Hello,I'm trying to link MKL with the clang compiler (on Linux), and I'm using the following link-line:clang-5.0 -std=c++17 -O3 -Wall -Wextra -Wno-missing-braces -Wunreachable-code -stdlib=libstdc++...
View ArticleQRDecomposition using LAPACK package
Hello,I am doing QR decomposition using LAPACKE_dgeqrf() function. Strangely it works fine on my windows Desktop but when I run same code on other windows desktop it gives error as 5th parameter is...
View Articleintel pardiso iparm[12] or iparm(13)
Hello,I have a question about the parameter iparm[12] (zero index, weighted matchings) of intel pardiso solver.Actually the problem was in a solution of a non linear problem by Newton Raphson...
View ArticleClik here to view.
Calling dgetrf_ before MPI_Init causes HANG with MKL_CBWR=COMPATIBLE
Hi,I have a mwe with the bug. I reproduced with the following setup:export MKL_CBWR=COMPATIBLEexport MKL_VERBOSE=1./fooOutput:Foo is called MKL_VERBOSE Intel(R) MKL 2018.0 Update 1 Product build...
View ArticleHow to force AVX-2 vs AVX-512
Hello,I'm running benchmarks of my code on test hardware (Intel Xeon Gold 5115), and i’m trying to isolate the impact of avx-512 vs avx-2 instructions on overall runtime. My issue is, I don’t know...
View ArticleAVX512 slower than AVX2? What I am doing wrong?
Hello All, I was so excited to test new the new Intel Xeon Silver 4114 CPU just to find out that with AVX512 enabled the performance of the matrix multiplication is the same as with legacy SSE4. If I...
View ArticleJoin the Intel® Parallel Studio XE 2019 Beta Program today
Join the Intel® Parallel Studio XE 2019 Beta Program today and—for a limited time—get early access to new features and get an open invitation to tell us what you really think.We want YOU to tell us...
View ArticleClik here to view.
Pardiso does not scale well in reordering step
Hi eveyrone,I have been solving a large symmetric indefinite sparse matrix using Pardiso symmetric solver, with property as below:< Linear system Ax = b > (Upper half) number of...
View ArticleMulti-dimensional function minimization
I have a non-linear function in many variables, whose minimum (local) I want to find. Is there an appropriate routine in the MKL for such a scenario?I can only find linear solvers and least-square...
View Articlecluster_sparse_solver Schur complement - how to distribute the Schur matrix?
We need to find the Schur complement matrix of a sparse matrix A of formA11 A12A21 A22I.e., we want the Schur block defined by S = A22 - A21 A11-1 A12, which can be done by the new sparse solver...
View ArticleClik here to view.
Pardiso does not scale at all, and possibly a memory leak
hiI had a problem with Pardiso in the past (https://software.intel.com/en-us/forums/intel-math-kernel-library/topic/...) and thanks to Alex, we were able to come out a solution in 2015.Now I am running...
View ArticleOptimizing matrix multiplication algorithm on Intel Xeon Gold (DevCloud)
Hi, I am working on Case #03357624 - Benchmarking algorithms on Intel Xeon Gold (DevCloud):https://communities.intel.com/thread/124090 Summary:The concern is on time overhead while running compiled...
View ArticleLINPACK with multiple MPI ranks
Hello,to benchmark our new Skylake cluster consisting of two and four socket machines together with a Broadwell system, I want to be able to run LINPACK with a different amount of MPI ranks per node....
View Article