hadibrais

From the blog post at: https://hadibrais.wordpress.com/2020/04/02/the-intel-x86-mic...

There are many (small) differences between AMD64 and Intel 64. But it's obvious that the Intel processor implements the Intel 64 version of x64 and the AMD processor implements the AMD64 version. This also goes for VT-x, VT-d, and AMD-V.

They should specify what benchmarks and input sets were used and the benchmarking methodology. It'd be nice to also see the standard deviation plotted on these bars.

Not yet. In the meanwhile, you can instead use the latest version of AMD μProf (https://developer.amd.com/amd-uprof/), which does support Zen 2.

I don't much about Java. But, for example, if you want to implement the technique from the link above in Java, you'd need LoadFloat8, BroadcastFloat8, and AdduFloat8. I'm not sure how they are implemented in Glow, but they can be implemented using inline assembly. You'd need something like SIMD intrinsics Java library to access these crucial x86 SIMD instructions.

Sparse matrix multiplication is generally considered to be a different problem. Sparse matrices need to be represented differently (See https://en.wikipedia.org/wiki/Sparse_matrix for an introduction). There has been recently a lot of research on how to do sparse matrix multiplication efficiently on manycore architectures. See this 2018 paper for example: https://arxiv.org/pdf/1804.01698.pdf. Note that sparse matrix-vector multiplication is also a different problem.