Hi, I am wondering what is the current state of multi-threaded BLAS in Julia? I remember looking into several options about a year ago for accelerating dense matrix multiplication and sparse solving on some PDE code I was working on. At the time, the impression I had was that OpenBLAS was quite poo…

There is an environment variable to control BLAS threads. You can also use MKL. Also there is Octavian.jl

though do be aware that Octavian.jl is soon deprecated [image] Why is LoopVectorization deprecated? Performance Polyester.jl doesn’t depend on LoopVectorization.jl or VectorizationBase.jl, but I think one of the same problems Julia 1.11 introduces impacts it as well, so w…

AMD Zen4 has AVX512, while only Intel server CPUs do. If you’re restricted to desktop level parts, I’d go AMD if you’re interested in any sort of linear algebra. Although, their throughputs are going to match (64 bytes of fma/clock cycle), so it isn’t that big a deal. With server class parts (whic…

[image] Elrod: the Intel CPUs that support AVX512 can do 128 bytes of fma/clock cycle. But they reduce their clock frequency when doing so, so in most cases there is no advantage over AMD…

BLIS has good multi threading performance on AMD. Easiest way to trying it out is via GitHub - JuliaLinearAlgebra/BLISBLAS.jl: BLIS-pendant of MKL.jl or with more access to the C API through GitHub - JuliaLinearAlgebra/BLIS.jl: This repo plans to provide a low-level Julia wrapper for BLIS typed inte…

That’s not really true. Ice lake and newer don’t really drop clock speeds, while older CPUs like Skylake-X and Cascadelake, which do drop clock speeds, are far faster than AMD when doing a decent job of leveraging AVX512. MKL and other BLAS libraries do a good job leveraging AVX512. A lot of @turb…

Thank you all for your inputs. The situation doesn’t seem entirely clear regarding which option is best, although going with Intel + MKL seems like a pretty safe bet.

Ok, thanks for this advice. One thing is that I definitely want a ‘desktop’ CPU, as the workstation is going to be running mostly ‘small’ jobs that won’t be massively parallelizable. So, we want a CPU that will have excellent single-thread performance (i.e. not Xeons). If the jobs were highly parall…

AOCL.jl is now registered under JuliaLinearAlgebra at AOCL.jl refer: AOCL.jl v5.3.1

What is the current state of multi-threaded BLAS in Julia?

General Usage Performance

adienes March 21, 2024, 7:17pm 3

though do be aware that Octavian.jl is soon deprecated

Topic		Replies	Views
Regarding the multithreaded performance of OpenBLAS Performance blas , multithreading	7	6219	January 31, 2022
Is MKL performance on AMD no longer crippled? General Usage	4	3177	May 11, 2024
OpenBLAS is faster than Intel MKL on AMD Hardware (Ryzen) Performance blas , lapack	40	37728	June 19, 2020
Matrix vector multiplication Performance question	4	1022	September 27, 2020
Strange multithreaded behaviour on ARM Ampere A1 (Oracle Cloud) Julia at Scale multithreading , aarch64 , arm	10	1438	April 14, 2022

What is the current state of multi-threaded BLAS in Julia?

Related topics