Anyone happen to have historical data comparing the Extended #BLAS implementation and performance across different #BLAS packages?

I'm back on grid, re-reading the spec (https://netlib.org/blas/blast-forum/chapter4.pdf) and I *think* there's nothing stopping me from having a #EGEMM routine using the same underlying techniques as @enp1s0 pointed out in their (his?) recent paper.

Partially because @steve gave a subtle nod of "it's not insane", I think it might workout well?

#HPC #BLAS #HPL

In the context of #HPC? It’s #BLAS

In the context of Raytracing? It’s 😩LAS and 😈LAS

🖥️💪 Do you struggle with numerical efficiency in your work? 🤔👨‍💼 Marcin Błażejowski has the answer! 💡👨‍🏫

Presented at the #gretl conference 2023 work on "Which C compiler and #BLAS/ #LAPACK library should I use?" to learn expert insights on configuration that boosts your work efficiency.

#economics #econometrics #gc2023 #econtwitter

🖥️💪 Do you struggle with numerical efficiency in your work? 🤔👨‍💼 Marcin Błażejowski has the answer! 💡👨‍🏫

Presented at the #gretl conference 2023 work on "Which C compiler and #BLAS/ #LAPACK library should I use?" to learn expert insights on configuration that boosts your work efficiency.

#numeralanalysis #economics #econometrics #gc2023 #econtwitter

Matrix-matrix multiplications in BLIS. James H. Wilkinson Prize for Numerical Software at #siamcse23
#gemm #hpc #blas #arm

Mixed precision #BLAS things:

#AVX512 foundation gave us conversion instruction for fp16->fp32->fp16 instructions.

Overtime, these instructions got faster and faster. That's great!

From SkylakeX to ICL, we went from Latency7, CPI 1 to Latency7 CPI 0.5!

Great.

But with SPR having dedicated FP16 math extensions, the importance of datatype conversion was "downgraded"

from 7 and 0.5 to 8 and 1

@sri @hipsterelectron

@karolherbst is already working towards SYCL support (don't remember if it's 2020 or 1.2.1?) with the RustICL project within mesa for opensource, OpenCL drivers.

I'd be looking at first towards a Rust written , #SYCL compatible #BLAS library that also provides classical #Fortran Bindings.

As of now there are BLAS bindings for rust, and rust written libraries that provide BLAS functionality, but there isn't a rust BLAS project that caters to a general audience (AFAIK)

I made a new blog post on helping speeding up a very slow matrix computation.
https://thomasburgess.github.io/blog/2023/02/02/Matrix_Multiplication.html #python #mcmc #matrix #optimization #blog #numpy #blas

Just compiling up a version of NCBI Blast for use on a large node. Anyone know why it looks for Boost/LaPack/GSL but doesn't require them? Will it be faster if I make them available?

#bioinformatics #hpc #ncbi #blast #blas

La ilustración navideña que dibujé en 2018 con los personajes de Barrio Sesamo.
Sigue siendo una de mis favoritas (sino la que más) y mira que ya tiene tiempo.

¡Feliz Navidad a todos!

#art #arte #christmas #holiday #dibujante #cartoon #comicartist #sesamestreet #elmo #epi #blas #coco #groomer #cookiemonster #cartoonart #fanart #artwork

Time for an #introduction!
I'm a young Canuck with interests/experience in #HPC, #Linux, #BLAS, #SYCL, #C, #AVX512, #Rust, heterogeneous compute & other such things.

Currently my personal projects are bringing #FP16 to the #OpenBLAS library, working to standardize what Complex domain BLAS FP16 kernels/implementations should look like, and making sure #SYCL is available everywhere.

I also write every now and again. Here's the tail of AVX512 FP16 on Alderlake
https://gist.github.com/FCLC/56e4b3f4a

Was going through the Risc-V Vector ISA spec (as you do) and noticed this little gem:

Specifically the line "When 16-bit and 128-bit element widths are added, they will be also be treated as IEEE-754/2008-compatible values. "

Unless I'm miss interpreting this, is Risc-V indicating future *native* support for 128 bit integer and floating point?

On the other hand, because I'm that guy: GOSH DARN IT, WHY NOT SHIP FP16 AS PART OF V.1 😭
https://github.com/riscv/riscv-v-spec/releases/download/v1.0/riscv-v-spec-1.0.pdf

#HPC #BLAS #RiscV #FP16 #ASM

#ai #DeepLearning #optimization #BLAS #deepmind #tensor

#圖靈獎 #TuringAward
#超級電腦 #LINPACK #BLAS #LAPAC
#平行計算 #MPI
#TOP500

https://mp.weixin.qq.com/s/OJUcpxgZp9GfsjjBdHiHdw

@rmhogervorst @rstats@gup.pe

Sources for using OpenBlas multi-threading numerical libraries in CRAN R:

[1] https://csantill.github.io/RPerformanceWBLAS/ (including Benchmarks and HowTo for Linux)

[2] http://brettklamer.com/diversions/statistical/faster-blas-in-r/

[3] https://www.r-bloggers.com/why-is-r-slow-some-explanations-and-mklopenblas-setup-to-try-to-fix-this/

[4] https://github.com/xianyi/OpenBLAS (binary packages for Windows)

#rstats #cran #multicore #multithreading #parallelization #linux #windows #OpenBlas #BLAS #LAPACK

Good news especially for Debian scientific computing users: BLAS/LAPACK Ecosys massive update https://lists.debian.org/debian-science/2019/10/msg00038.html #LAPACK #BLAS #LinearAlgebra