Hello. I’m trying to compare BLAS and CUBLAS performance with Julia. For example, I want to compare matrix multiplication time. Let A, B, C will be [NxN] matrices. Which function, should I use to get something like C = AB? Will standard AB implementation be the fastest one (using BLAS)? Is it …

It really depends on what you want to measure. If you care about the time it’ll take to allocate memory for the output, you can just use @btime $C = $A*$B and @btime CUDA.@sync $C = $A * $B. If you actually just care about the performance of the actual matrix multiplication and don’t want your resul…

I wanted to measure actual calculation time (without allocation). Thanks!

Will this calculate only multipilcation time? Looking at https://juliagpu.gitlab.io/CUDA.jl/ I was wondering if compilation before calling kernel will be OK (to avoid compiling time in actual results).

Yes, everything should be compiled here before the benchmark loop is run.

[image] Mason: Here’s an example: Faster than my CPU, but at least it’s in the same ballpark: julia> using LinearAlgebra; BLAS.vendor() :mkl julia> M = K = N = 10^3; julia> A = randn(Float32, M, K); B = randn(Float32, K, N); C = Matrix{Float32}(undef, M, N); julia> @benchmark mul!($C, $A…

@jpsamaroo Any ideas? [gemmFloat64_100_4000_skylake-avx512_AVX512] CPU was a 7980XE, and GPU a Vega 64. I think the GPU should manage more than a couple GFLOPS. FWIW, I also have these problems: (@v1.5) pkg> build AMDGPU Building AMDGPU → `~/.julia/packages/AMDGPU/nnddY/deps/build.log` ┌ Err…

I haven’t benchmarked rocBLAS at all, but have you tried increasing the matrix size even further? Memory transfer costs could be dominating at such a small size. Also, what AMDGPU.jl version are you on? You should test on 0.2.0 at a minimum, to get the ROCArray changes. I’m not sure what’s up with …

It turns out after building AMDGPU, isnothing(AMDGPU.librocblas). It must’ve been running a fallback. If I instead include the build script from the REPL (after using BinaryProvider), suddenly it’s defined and seems to run much faster. But I’m getting approximate equality failures between the CPU…

I posted a PR to at least make the builds quieter. [image] Elrod: It turns out after building AMDGPU, isnothing(AMDGPU.librocblas) . It must’ve been running a fallback. That’s probably GPUArrays generating a matmul kernel. [image] Elrod: But I’m getting approximate equality failures be…

BLAS vs CUBLAS benchmark

General Usage Performance

Mason September 7, 2020, 8:01pm 3

Here’s an example of how I’d bencmark the scaling of OpenBLAS vs CuBLAS

Topic		Replies	Views
CLBlast, a tuned OpenCL BLAS library GPU gpu , gpuarrays	6	1609	August 9, 2018
OpenBLAS is faster than Intel MKL on AMD Hardware (Ryzen) Performance blas , lapack	40	37523	June 19, 2020
CUDA v2 - performance regression on matrix multiplication GPU	14	1883	November 10, 2020
Poor openBLAS performance for large matrix multiply? New to Julia openblas	17	1809	April 4, 2025
CUDA matmul performance GPU question , performance	11	1708	August 21, 2020

BLAS vs CUBLAS benchmark

Related topics