Linear solver \(A, B) performance vs Matlab A\b

[Tecnezio]

I’ve done it !
Here the post about my test:

Call Intel MKL as external library

No, i haven’t set anything… But i loaded the wrong version of the script… and now it work! Here the working version: # Test MKL solve linear systems N = 10000; A = rand(N,N); AA = copy(A); b = rand(N,1); c = copy(b); function juliaSol(A,b) \(A,b); end @time sol = juliaSol(A,b); # Test MKL const global librt = Libdl.find_library(["libmkl_rt"], ["/opt/intel/mkl/lib"]) # Open librt Libdl.dlopen(librt) # LU FACORIZATION function luFactMKL(A::StridedMatrix{Float64}) m, n = size(A) lda = …