Christos Argyropoulos

@ChristosArgyrop@mast.hpc.social
60 Followers
73 Following
373 Posts

Interested in #HPC for clinical (electronic health records) and bioinformatic (sequencing) research.
In perpetual student mode (in 2024 this means lots of #OpenMP, #OpenMPI and #Fortran!)

Associate Professor and Chief of Nephrology Department of Internal Medicine,
Faculty Lead, Hub Research Capacity, Clinical and Translational Science Center, University of New Mexico Health Sciences Center, Albuquerque, NM,USA

Redditu/ReplacementSlight413
Google Scholarhttps://scholar.google.com/citations?user=Hj15fvUAAAAJ&hl=en
mstdn.science@ChristosArgyrop
Programming LanguagesR | Perl | C | Mathematica | SQL | Python | Fortran
@ChristosArgyrop @openmp_arb depending on the GPU hardware that you use, single precision execution may utilise tensor cores or produce different residual loop behaviour. With Nvidia compiler, options -Kieee and -gpu=uniform (and not using fastmath) may restore a more coherent behaviour.
The problem with AGI/ASI and LLMs (thinking or not) is that *linear algebra* (the matrix multiplications from high school),can never be sentient in the same manner as differential equations of the transport processes inside the kidney can produce homeostasis (or make urine).

In an opinion piece in The Conversation, #ACMTuringLaureate Jack Dongarra takes a stand for traditional high-performance computing (#HPC), opining that AI's reliance on HPC necessitates greater focus from the U.S. government.

"US national strategy should include funding new machines and training for people to use them," he writes. Read more: https://theconversation.com/challenges-to-high-performance-computing-threaten-us-innovation-255188

This is really bizarre an error with @openmp_arb (or gcc) when offloading to a #gpu. Double precision version of the code that uses trigonometric functions offload and executes flawlessly. Single precision version of the same code dies with an illegal memory access. I suspect it has to do with the lm library (or rather with sinf cosf). Is there an alternative math library for gpus to try?
Fighting the HOA surveillance drones one desert baby bunny a time
Learning Perl – Introduction

Perl has long been known as the “duct tape of the Internet,” or "the Swiss Army chainsaw of scripting...

DEV Community
Since we can't use the cloud to automate our #EHR analysis projects , we tried #nextflow (traditionally used in bioinformatics https://www.nextflow.io/), and it worked as a charm coordinating #duckdb #R #python and within node tasks.
Next in line is (R)?ex https://www.rexify.org/
A DSL for parallel and scalable computational pipelines | Nextflow

Nextflow enables scalable and reproducible scientific workflows using software containers. It allows the adaptation of pipelines written in the most common scripting languages.

If my C code runs like a tortoise when compiled with -O0 , but as a rocket when compiled with -O3 shouldn't the compiler engineers get part of the credit ?
(The corollary is that microbenchmarks should either be written in straight assembly or conducted across a spectrum of optimization flags)
How this billion-dollar London startup backed by Microsoft made 700 engineers sitting in India pose as AI | - The Times of India

Tech News : Builder.ai, once valued at $1.5 billion, has filed for bankruptcy after being exposed for misleading investors about its AI capabilities. The company,

The Times of India
Based