flame/blis on Hacker News

https://github.com/flame/blis#citations is the current set of references for various aspects of the approach. The article, or something similar, was posted recently with discussion.

comment by gnufx · 2024-08-01 · in Fast Multidimensional Matrix Multiplication on CPU from Scratch (2022)

BLIS is an interesting new direction in that regard: https://github.com/flame/blis >The BLAS-like Library Instantiation Software (BLIS) framework is a new infrastructure for rapidly instantiating Basic Linear Algebra Subprograms (BLAS)…

comment by sevagh · 2024-04-24 · in Faer-rs: Linear algebra foundation for Rust

…update to the blis alternative to BLAS that includes a number of RISC-V performance optimizations. https://github.com/flame/blis/pull/737

comment by j_not_j · 2024-03-01 · in Optimize sgemm on RISC-V platform

The real money shot is here: https://github.com/flame/blis/blob/master/docs/Performance.m... It seems that the selling point is that BLIS does multi-core quite well. I am especially impressed that it does as well as the highly optimized…

comment by quanto · 2024-01-24 · in BLIS: A BLAS-like framework for basic linear algebra routines

…2019. At the time in particular, OpenBLAS didn't properly use the AVX512 VPUs. (See thread in BLIS https://github.com/flame/blis/issues/352 ), It has made progress since then, still, on my current laptop perf is in the same range…

comment by mratsim · 2024-01-04 · in Benchmarking 20 programming languages on N-queens and matrix multiplication

https://github.com/flame/blis/ Field et al, recent winners of the James H. Wilkinson Prize for Numerical Software. Field and Goto both collaborated with Robert van de Geijn. Lots of TACC interaction in that broader team.

comment by RhysU · 2023-12-30 · in The art of high performance computing

…BLIS makes a good BLAS. It almost keeps up with MKL, according to the nice plots on their GitHub. https://github.com/flame/blis/blob/master/docs/Performance.m... They use C mostly, because the magic is in the GEMM (and possibly TRSM)…

comment by bee_rider · 2023-05-18 · in SciPy: Interested in adopting PRIMA, but little appetite for more Fortran code

Here's BLIS's object API: https://github.com/flame/blis/blob/master/docs/BLISObjectAPI... Searching "object" in BLIS's README ( https://github.com/flame/blis ) to see what they think of it: "Objects are relatively lightweight structs and…

comment by sfpotter · 2022-10-27 · in Column Vectors vs. Row Vectors

…blasfeo). Eigen might also be relevant. https://libxsmm.readthedocs.io/ https://blasfeo.syscop.de/ https://github.com/flame/blis

comment by gnufx · 2022-04-14 · in Doing small network scientific machine learning in Julia faster than PyTorch

https://github.com/flame/blis/blob/master/docs/Performance.m... but ignore the SKX results with the old OpenBLAS there.

comment by gnufx · 2022-04-04 · in Eigen: A C++ template library for linear algebra

…for the innermost block size of the non-trivial hierarchy you need. You can see examples under https://github.com/flame/blis/tree/master/config with an a priori procedure for determining them in https://www.cs.utexas.edu/users/flame/pubs/TO…

comment by gnufx · 2021-12-18 · in Matrix Multiplication Inches Closer To Mythic Goal

…them for GEMM; with a current microarchitecture implementation, if you only have one, you don't. https://github.com/flame/blis/blob/2d8ec164e7ae4f0c461c27309... GEMM is one of relatively few computations with sufficient computational…

comment by gnufx · 2020-10-11 · in AMD Zen 3/Ryzen 5000 announcement [video]

…There are results for AVX2 systems with older versions of OpenBLAS and BLIS, which is AMD's BLAS at https://github.com/flame/blis/blob/master/docs/Performance.m... I haven't seen or made measurements for EPYC2 yet (and don't have results…

comment by gnufx · 2020-09-01 · in Intel MKL on AMD Zen

You can see BLIS on Intel's home turf at https://github.com/flame/blis/blob/master/docs/Performance.m... (52-core SKX) and compare with OpenBLAS on 32-core Zen1. (Multithreaded BLAS isn't typically used in HPC, where the parallelism is…

comment by gnufx · 2020-08-29 · in Intel's “cripple AMD” function (2019)

…BLIS is used by AMD and is a good open alternative to MKL (for BLAS) across many platforms. https://github.com/flame/blis/blob/master/docs/Performance.m...

comment by jedbrown · 2020-08-28 · in Intel's “cripple AMD” function (2019)

…low bar [1], and not at all how one should go about writing a performance portable BLAS. BLIS ( https://github.com/flame/blis/ ) is a much better approach, and underlies vendor implementations on AMD (…

comment by jedbrown · 2020-02-15 · in Auto-vectorization for the masses (2011)

…pointless. Just use BLIS, like AMD do, and which is infinitely faster than MKL on non-x86 systems. https://github.com/flame/blis

comment by gnufx · 2019-12-08 · in Reducing the Performance Gap of Intel's MKL on AMD Threadripper

…around 2 TFLOPS. You can take a look at reproducing them if you want if you have a similar system: https://github.com/flame/blis/blob/master/docs/Performance.m... I ran into some compile issues, but I care more about numpy so I gave this…

comment by lhl · 2019-12-08 · in Reducing the Performance Gap of Intel's MKL on AMD Threadripper

…not an "obscene edge case" - MKL cripples AMD on every single matrix operation tested here: https://github.com/flame/blis/blob/master/docs/graphs/large/...

comment by lhl · 2019-12-08 · in Reducing the Performance Gap of Intel's MKL on AMD Threadripper

…optimizations (I know AMD has their own fork, but check out the 64 core results for "vanilla" BLIS: https://github.com/flame/blis/blob/master/docs/Performance.m... )

comment by lhl · 2019-12-08 · in Reducing the Performance Gap of Intel's MKL on AMD Threadripper

…BLAS, but I don't know what there is that doesn't have free rough equivalents.) BLIS performance: https://github.com/flame/blis/blob/master/docs/Performance.m...

comment by gnufx · 2019-12-02 · in Puget Systems:“Intel Has Every Right” to Make Intel MKL Run Slower on AMD CPUs

…Kernel Library (MKL) is very well implemented in open source projects such as OpenBLAS and BLIS: https://github.com/flame/blis Both are well optimized for AMD CPUs.

comment by ogrisel · 2019-09-20 · in AMD Ryzen Threadripper 3000 32-core CPU is more bad news for Intel

…vectorization, but also cache use with levels of blocking (and prefetch). See the material under https://github.com/flame/blis . Other levels -- not matrix-matrix -- are less amenable to fancy optimization, with lower arithmetic density to…

comment by gnufx · 2019-05-14 · in Numpy Clone in Common Lisp

…actually tells you how to write a fast production large-matrix GEMM, and the papers linked from https://github.com/flame/blis would be a better reference than the Goto and van de Geijn. For small matrices see the references from…

comment by gnufx · 2018-05-29 · in How to write efficient matrix multiplication

…using __builtin_prefetch). You can study the (micro-)architecture-dependent uses in e.g. https://github.com/flame/blis https://github.com/xianyi/OpenBLAS and https://github.com/hfp/libxsmm and it may be discussed in the papers describing…

comment by gnufx · 2018-05-01 · in Is software prefetching (__builtin_prefetch) useful for performance?

…to beat the compiler should start by taking something like a GEMM routine (for example from there https://github.com/flame/blis/tree/master/kernels/x86_64 ) and reimplement it to show how a mainstream optimizing compiler can do better…

comment by Joky · 2016-11-29 · in Beating the Compiler

…And I've not yet seen a high level language that does a good treatment of supporting simd shuffles. https://github.com/flame/blis/blob/master/kernels/x86_64/avx... is a very very good example of how shuffle/permutation heavy interesting…

comment by carterschonwald · 2014-12-20 · in What’s Wrong with Statistics in Julia? A Reply