www.tomshardware.com/software/clo...

Infrastructure upgrade week. Fiber came to the neighborhood in June. Upgraded some infrastructure in July and BAM!

I'm not sure what to make of this backyardbrains.com/products/rob... On the one hand, ew cockroaches on the other hand, an army of indestructible little bug tanks, hmm...

RoboRoach Kit | Build a Cyborg Cockroach | Backyard Brains

The RoboRoach Kit by Backyard Brains teaches students real neuroengineering by wirelessly controlling a cockroach's movements.

backyardbrains.com

I am returning today after switching from managing the High Performance Computing Center at Texas Tech to full-time as Research Professor and co-director of the repacss.org project, focused on energy and resource efficiency in scientific research computing. Find me here or at nsfcac.org for details!

Wrote up the Satochi paper on HPCwire. I tried to pull the important points. HPC Precision Wars: Satoshi Matsuoka Plants the Ozaki Flag www.hpcwire.com/2026/06/15/h...

HPC Precision Wars: Satoshi Matsuoka Plants the Ozaki Flag - HPCwire

A recent paper submitted to ArXiv by famed HPC scientist Satoshi Matsuoka, Director of the RIKEN Center for Computational Science in Kobe, Japan, has shaken the tried-and-true FP64 HPC relationship to...

hpcwire.com

History of CentOS: How a biochemist's Linux hobby project became the enterprise world's default operating system

History of CentOS: How a biochemist's Linux hobby project became the enterprise world's default operating system

When a community came together after Red Hat said Windows was 'probably the right product'

theregister.com

#HPC must read, "FP8 is All You Need (Part 1): Debunking Hardware FP64 as the HPC Holy Grail" from Satoshi Matsuoka arxiv.org/pdf/2606.06510

Phrase of the day "genesis content" www.hpcwire.com/2026/05/22/g...

Genesis Content: Where Does GenAI Get Its Next Meal? - HPCwire

Something about building GenAI LLMs bugs me. Before I begin, let me be clear: I am a supporter of AI technologies, particularly in science. Lately, however, a question keeps surfacing that I find hard...

hpcwire.com

Tomorrow (Thursday, May 21), Brad Chamberlain and Jade Abraham will be giving an overview, update, and demo of Chapel at the Northwest C++ Users’ Group at 7pm PT. Attend in person in Bellevue WA, or online using Microsoft Teams. nwcpp.org/May-2026.html

In light of all the new potential Linux kernel "bugs," Jasper Nuyens created a simple script to place all unused modules (not currently loaded) in "module jail" i.e.the kernel cannot load them. Good temporary hygiene until things get cleaned up. Read more here: github.com/jnuyens/modu...

GitHub - jnuyens/modulejail: Proactively shrink a Linux host's kernel-module attack surface by blacklisting every module not currently in use.

Proactively shrink a Linux host's kernel-module attack surface by blacklisting every module not currently in use. - jnuyens/modulejail

github.com

#hpc This Friday's exploit: ssh-keysign-pwm github.com/0xdeadbeefne...

GitHub - 0xdeadbeefnetwork/ssh-keysign-pwn: Steal SSH host private keys and /etc/shadow via the ptrace_may_access mm-NULL bypass + pidfd_getfd. Pre-31e62c2ebbfd kernels.

Steal SSH host private keys and /etc/shadow via the ptrace_may_access mm-NULL bypass + pidfd_getfd. Pre-31e62c2ebbfd kernels. - 0xdeadbeefnetwork/ssh-keysign-pwn

github.com

Well that is wrap, Spring 2026 Big Data Management (aka scalable tools) at Lehigh University is done, grades submitted, 29 students now know about Spark, Hadoop, Hive, Kafka, ... and even a bit of Linux command line 😉

Read how we designed and deployed Multipath Reliable Connection (MRC), a multi-company step towards Ultra Ethernet, at 800G in Microsoft's datacenters. Demonstrated with a 75k GPU pretraining job running stably through multiple faults. OpenAI blog: buff.ly/X0KL4Xy Paper: buff.ly/I3OcXWn

Supercomputer networking to accelerate large scale AI training

OpenAI introduces MRC (Multipath Reliable Connection), a new supercomputer networking protocol released via OCP to improve resilience and performance in large-scale AI training clusters.

buff.ly

At the Salishan HPC conference last week, I gave a talk called "AI doesn't need massive supercomputers after all!" A couple people asked me for the slides (as crappy as they were), so here they are in hastily written blog format. blog.glennklockwood.com/2026/05/ai-d... #AI #HPC

AI doesn't need giant supercomputers after all

I attended the 2026 Salishan Conference on High Speed Computing last month, and it was a week well spent in coastal Oregon hearing what man...

blog.glennklockwood.com

Wow. Another HPC acceleration by AI paper. This is the way

Underfox@underfox3.bsky.social · 6mo ago

In this paper, researchers have presented the first exascale multi-task workflow for atomistic graph foundation models, trained on 544+ million structures from 16 datasets using 16,384 GPUs on Frontier supercomputer. arxiv.org/pdf/2604.15380

Let me see if I got this, "We trained our AI on all your text and code, so your job is not needed, bye. We'll continue to scrape the new AI generated interwebs for even more content because we put the original creators out of work. Profit!" Did anyone read the papers about about model collapse?

a black and white silhouette of a cat laying on its back on a white background

ALT: a black and white silhouette of a cat laying on its back on a white background

media.tenor.com

First week of June, east coast AI for science! Registration is OPEN for #TPC26! Join us for one of the most forward-thinking conferences in tech, policy, and leadership. Where brilliant minds come together to shape the future. Early-bird pricing here: ow.ly/azv950Ymum0 #artificialintelligence

The Trillion Parameter Consortium

Convening AI leaders globally from industry, academia, and national laboratories

ow.ly

Reason number 251 why you should read hpcwire.com supercarblondie.com/tech/superco...

Home - HPCwire

hpcwire.com

At this point in the game, has anyone noticed that effective use of GenAI requires actual wet intelligence. Just say'n

Bit windy today on the east coast

Please share to help spread the word! We have already have some applications for the #HPC system administrator open position in our team at Ghent University (Belgium), but it's not too late yet! Deadline to apply is Sun 1 Feb 2026. jobs.ugent.be/job/Ghent-Sy...

Show floor safari from sc25 (BTW, this is obviously AI generated because, I would never ask such goofy questions and try and explain QC) ow.ly/mhHC50Y0F1m

What started as a few paragraphs to help explain AI to a friend grew into this long article (with reference links). If you have not been watching AI and LLM progress closely, like I have, you may find this background primer helpful. www.hpcwire.com/2025/12/16/a...

A (Mostly) Non-Technical AI Primer - HPCwire

I have studied and watched artificial intelligence grow over the last forty years. Like many, back in 1968, I was inspired by HAL 9000 in Stanley Kubrick’s 2001: A Space Odyssey. The question on my mi...

hpcwire.com

I just got saw Nvidia acquired ShedMD (Slurm), hmm blogs.nvidia.com/blog/nvidia-...

NVIDIA Acquires Open-Source Workload Management Provider SchedMD

NVIDIA will continue to distribute SchedMD’s open-source, vendor-neutral Slurm software, ensuring wide availability for high-performance computing and AI.

blogs.nvidia.com

IBM just purchased Kafka (Confluent). Smart move because building data pipelines is the first step in model building If you want to learn about Kafka, guess what, yours truly has created a video course to get you started (complete with serious looking head shot). www.oreilly.com/videos/kafka...

Kafka Essentials LiveLessons: A Quick-Start for Building Effective Data Pipelines

8+ Hours of Video Instruction Learn how to manage high-performance data pipelines, streaming analytics, data integration, and mission-critical applications with Apache Kafka. Apache... - Selection fro...

oreilly.com

As I understand the situation, the price of DDR5 memory has tripled (or more) because memory vendors have switched to making more lucrative HBM used on the GPUs that will be placed in unpowered AI systems that are waiting for a data center.

If you are interested in HPC and AI, take the time to read this piece. Plus, the best vendor analysis charts ever!

Glenn K. Lockwood@glennklockwood.com · 10mo ago

I wrote up my notes from #SC25. Have a look: blog.glennklockwood.com/2025/12/sc25... I’ll keep picking away at the editing, but would love to hear more from others about what stood out to them. I wasn’t at the conference itself as much this years as in the past, so I know I missed a lot. #HPC

Bonus fun, who can you identify

A few #SC25 observationS www.hpcwire.com/2025/11/19/s...

SC25 Observations: More Pumps than Processors - HPCwire

Each year at SC, I like to report on some of the experiences and observations I find as the show progresses. There is so much more going on, and I’m sure I missed a lot of news and events. There is a ...

hpcwire.com