#supercomputing

[larimar]

nasaames

[newstech24]

How Supercomputing Will Evolve, In line with Jack Dongarra

Quantum computing is attention-grabbing. It’s actually a beautiful space for analysis, however my feeling is we’ve got a protracted solution to go. Right now we’ve got examples of quantum computer systems—{hardware} at all times arrives earlier than software program—however these examples are very primitive. With a digital pc, we consider doing a computation and getting a solution. The quantum pc…

[govindhtech]

UK’s £2 billion Computer Roadmap 2025 In Quantum Computing

The UK Computer Roadmap 2025

To create a durable, autonomous, and future-ready computing environment, the UK invested £2 billion in a national compute infrastructure strategy. The 2025 UK Compute Roadmap plans to integrate quantum systems into the nation's essential infrastructure by the end of the decade.

Computing is a strategic enabler, and the plan comes at a crucial time. US, Canadian, and UAE rivals are investing billions, and hyperscalers alone will spend over $300 billion by 2025, disrupting global supply chains from energy infrastructure to semiconductors. The UK recognises the problem's enormity and the ramifications of outdated computer technology.

Quantum computing: UK Future Strategy

Despite its low adoption, quantum computing is listed in the roadmap as a technology that will transform the national compute architecture. This is one of the clearest evidence that the UK views quantum computing as a competitive computer paradigm rather than a research curiosity. A hybrid infrastructure approach by the government integrates quantum technology with cloud-based, supercomputing, and new computing paradigms.

The approach incorporates quantum into regular scientific processes rather than separate experimentation. UK National Supercomputing Centres (NSCs) will incubate quantum technologies before AI Growth Zones (AIGZs) scale deployments. Next-generation computing requires room and power, hence AIGZs were designed. Edinburgh's £750 million national supercomputer facility, the Edinburgh Parallel Computing Centre (EPCC), is the UK's first.

The “Compute Bridge” connects research to practice

Implementing a “Compute Bridge” is a key roadmap recommendation. This strategy helps British computational technologies, including quantum, go from study to implementation. This strategy will test new technologies at NSC testbeds. If practical, AI Growth Zones, which are expected to have 6 gigawatts of AI-capable processing by 2030, may expand these solutions into commercial settings. UK researchers and quantum enterprises can enter an environment governed by hyperscale cloud providers and well-established silicon hardware with this technique.

Over the Stack Sovereignty and National Priorities

Building autonomous, safe, and sustainable computing capabilities is important to the roadmap. A sovereign nation may allocate resources to national tasks, protect private information, and minimise overreliance on foreign infrastructure, according to the UK. One of the few areas where the UK can influence worldwide standards and leverage its capabilities in photonics, software optimisation, and chip design is quantum. The assets cited include ARM, Cambridge Quantum (now Quantinuum), and emerging hardware companies.

The National Quantum Technologies Program and Sovereign AI Unit will fund these enterprises. The Sovereign AI Unit will direct £500 million to quantum research and use AI Research Resource (AIRR) computational capacity to accelerate quantum technology development. Quantum technologies are well-positioned to support national agendas and create strategic leverage for the UK.

Beyond Infrastructure: Software, Benchmarks, Skills

Unlike many national strategies, the UK roadmap clearly explains quantum integration. NSCs will be funded and organised to support quantum R&D by hosting hardware warehouses, software development, data curation, and talent training. The strategy also recommends creating an open-source “Living Benchmarks Library” of compute workloads to aid procurement. Quantum's ability in modelling and optimisation could allow quantum ideas to be evaluated against real scientific issues.

Hybrid Future: 2030 Quantum-Ready

The government predicts AIRR's computing capability to expand twentyfold to 420 AI exaFLOPs by 2030. Although GPU-based infrastructure would fuel much of this increase, the strategy makes it clear that quantum computing will be in the national portfolio by then. By integrating quantum into its cloud and supercomputing infrastructure early on, the UK intends to pioneer hybrid compute environments.

In materials science, drug development, and climate prediction, quantum computers coexisting with classical systems will be the first economically effective quantum applications. The strategy's 2030+ vision includes “AI-hybrid systems” and predicts that quantum's role will shift as early specialised scientific deployments become broader AI integration.

TechUK Welcomes but Seeks Clarity

techUK Hello, but seeking Clarity techUK, the UK's technology trade association, sees the computing Roadmap as crucial to meeting the Spending Review's £1 billion AI and computing infrastructure allocation. They acknowledge the vision's ambition and the compute environment's fragmentation and capacity overload, caused by inconsistent policy. A united national computing environment is the roadmap's goal with centres of excellence, strong user support, and integrated data, software, and skills initiatives.

Although the direction is promising, TechUK notes that nothing is known about how the plan will integrate the people, infrastructure, technology, and governance needed to make the UK competitive in the global AI market. In particular, they note that while the roadmap calls for substantial programs and testbeds to support quantum processing technologies, there are few precise strategies and timeframes for commercial ising emergent UK technology.

TechUK also emphasises that AIGZs need reliable chip supply planning and commercial, cloud-based compute to push computation beyond public infrastructure. They want to know exactly where AIGZs are in Wales and Scotland and where the Sovereign AI Unit focusses.

Uncertainties and Challenges Remai

There are substantial risks despite high goals. The plan relies on government, academic, and industry collaboration, which can be problematic in practice. The success of new quantum enterprises and early-stage technology that failed to scale are also important.

Quantum computing's practical utility in time to justify infrastructure funding is uncertain due to murky commercial ready timescales. Employee development is another concern. The strategy asks for upskilling public sector and industrial users, but it doesn't explain how quantum talent will be developed outside of research.

[techfinancehubplus]

OpenAI Strikes Back: The Real Story Behind Its Bold New Hires from Tesla, xAI, and Meta

If you thought the world of artificial intelligence was all about code and cold calculations, think again. There’s a very human drama unfolding in Silicon Valley right now—a high-stakes tug-of-war for the brightest minds in AI. And this week, OpenAI just made a power move that’s turning heads across the tech world.

The Talent War Heats Up

It all started when Meta (yes, Facebook’s parent company) began aggressively poaching some of OpenAI’s top researchers, reportedly offering eye-watering signing bonuses and perks. For a moment, it looked like OpenAI might be on the back foot. But then, in a move straight out of a tech thriller, OpenAI struck back—poaching four engineering heavyweights from Tesla, xAI, and even Meta itself.

Meet the New Recruits

So, who are these new OpenAI team members? Let’s put faces to the names:

David Lau: Until recently, he was Tesla’s Vice President of Software Engineering. If you’ve ever marveled at a Tesla’s self-driving features, you’ve seen his work in action.

Uday Ruddarraju: The mastermind behind xAI’s colossal supercomputer, Colossus. This guy knows how to build the digital brains that power tomorrow’s AI.

Mike Dalton: Another Colossus veteran, Dalton’s experience in large-scale computing is exactly what OpenAI needs as it gears up for its next big leap.

Angela Fan: Fresh from Meta, Fan brings cutting-edge AI research skills and a passion for building smarter, safer AI.

Why This Matters (And Why It’s So Personal)

This isn’t just about filling job openings. It’s about the future of AI—and who gets to shape it. These engineers aren’t just employees; they’re visionaries, builders, and leaders. Their expertise could tip the balance in the race toward artificial general intelligence (AGI)—the holy grail of AI that could change the world as we know it.

And let’s not forget the human side: behind every headline about “poaching” and “talent wars” are real people making big career moves, chasing bold missions, and sometimes even switching sides in long-standing tech rivalries.

The Ripple Effect

Elon Musk, who co-founded OpenAI and now leads xAI, hasn’t been shy about his feelings. He’s publicly called out OpenAI for “aggressively recruiting Tesla engineers with massive compensation offers.” Tesla, in turn, has bumped up pay to keep its own stars from jumping ship.

Meanwhile, Meta’s own hiring spree has forced everyone in the industry to up their game. It’s not just about who has the best algorithms anymore—it’s about who can build the most powerful, reliable, and ethical AI infrastructure.

What’s Next?

With these new hires, OpenAI is doubling down on building the backend systems and supercomputers that will power the next generation of AI. Projects like Stargate, their ambitious new AI infrastructure initiative, are now in even more capable hands.

But don’t expect the talent war to cool down anytime soon. As AI becomes more powerful—and more central to our lives—the race for the best minds in the business is only going to get hotter.

So, the next time you read about an AI breakthrough or a new supercomputer, remember: behind every innovation are people making bold moves, taking risks, and shaping the future—one job offer at a time.

[retrocompmx]

Efemérides informática: 8 de julio de 1957

El 8 de julio de 1957 se funda Control Data Corporation (CDC), una importante empresa informática, conocida principalmente por sus sistemas mainframe y supercomputadora.

Fue un actor clave en la industria informática durante la década de 1960, junto con empresas como IBM y DEC.

CDC era especialmente conocida por sus sistemas informáticos de alto rendimiento diseñados por Seymour Cray, incluyendo los CDC 6600 y 7600, que en su día fueron los ordenadores más rápidos del mundo.

En 1972 Cray deja CDC para formar su propia empresa (Cray Research) y CDC enfrentó serias dificultades financieras hasta que dejó el negocio de la informática a finales de los 80s. En 1999 Syntegra compró lo que quedaba de la fragmentada CDC.

Lo único remanente hoy en día de CDC es Dayforce, una compañía de software

A pesar de su cierre, CDC deja un legado importante en el área de la computación de alto rendimiento. Seymour Cray contribuyó a establecer el campo de la supercomputación e influyó en el diseño de las futuras generaciones de computadoras.

#retrocomputingmx #CDC #seymourcray #supercomputing #computerhistory

[alltimeupdating]

Quantum computers just crossed a milestone — solving problems no traditional machine ever could. From faster drug discovery to smarter traffic systems, the future is now being powered by qubits. 🌍 Explore what this means for healthcare, finance, and our everyday lives.

👉 Read the full story and see how this breakthrough changes everything.

[impact-newswire]

NVIDIA Brings AI Supercomputer Manufacturing to the U.S. for the First Time

NVIDIA Blackwell chip production starts in Arizona — NVIDIA opens first US factories. NVIDIA is working with its manufacturing partners to design and build factories that, for the first time, will produce NVIDIA AI supercomputers entirely in the U.S. Together with leading manufacturing partners, the company has commissioned more than a million square feet of manufacturing space to build and…

[jade-malay]

Dallas Innovation: Where AI Shapes the Future

AI and supercomputing aren't just the future —they're the new foundation of innovation right here in Dallas! Where cutting-edge technology, top talent, and bold vision come together to build a smarter, brighter world. #JadeMalay #InnovationHub #DallasTech #FutureReady

[newslink7com]

Nvidia Hit With $5.5 Billion Loss After U.S. Government Bans H20 AI Chip Exports to China Over National Security and Supercomputing Risks

The U.S. just blocked Nvidia’s most advanced AI chip for China, citing fears of supercomputing misuse. The result? A staggering $5.5B charge and global tech ripple effects.

👉 Read the full story at NewsLink7.com

[newstodays1]

Nanotechnology and Supercomputing: A Synergy for the Future

Introduction The intersection of nanotechnology and supercomputing is driving unprecedented advancements in computational power, energy efficiency, and miniaturization. As traditional silicon-based computing approaches its physical limits, nanomaterials and quantum-scale innovations are paving the way for next-generation supercomputers capable of solving complex global challenges. This blog…

[nerdontour]

NVIDIA’s Project DIGITS brings #AI supercomputing to your desk! Powered by the GB10 Grace Blackwell Superchip, it delivers 1 petaflop of performance, runs 200B-parameter models, and fits digital nomads’ needs with portability and seamless cloud scaling. Launching in May at $3,000

[sentivium]

Project Digits: How NVIDIA's $3,000 AI Supercomputer Could Democratize Local AI Development | Caveman Press

[karannnnn69]

Supercomputing: The Key to National Leadership

[ai-network]

Elon Musk is Breaking the GPU Coherence Barrier

In a significant development for artificial intelligence, Elon Musk and xAI has reportedly achieved what experts deemed impossible: creating a supercomputer cluster that maintains coherence across more than 100,000 GPUs. This breakthrough, confirmed by NVIDIA CEO Jensen Huang as "superhuman," could revolutionize AI development and capabilities. The Challenge of Coherence Industry experts previously believed it was impossible to maintain coherence—the ability for GPUs to effectively communicate with each other—beyond 25,000-30,000 GPUs. This limitation was seen as a major bottleneck in scaling AI systems. However, Musk's team at xAI has shattered this barrier using an unexpected solution: ethernet technology. The Technical Innovation xAI's supercomputer, dubbed "Colossus," employs a unique networking approach where each graphics card has a dedicated 400GB network interface controller, enabling communication speeds of 3.6 terabits per second per server. Surprisingly, the system uses standard ethernet rather than the exotic connections typically found in supercomputers, possibly drawing from Tesla's experience with ethernet implementations in vehicles like the Cybertruck. Real-World Impact Early evidence of the breakthrough's potential can be seen in Tesla's Full Self-Driving Version 13, which reportedly shows significant improvements over previous versions. The true test will come with the release of Grok 3, xAI's next-generation AI model, expected in January or February. Future Implications The team plans to scale the system to 200,000 GPUs and eventually to one million, potentially enabling unprecedented AI capabilities. This scaling could lead to: More intelligent AI systems with higher "IQ" levels Better real-time understanding of current events through X (formerly Twitter) data integration Improved problem-solving capabilities in complex fields like physics The Investment Race and the "Elon Musk Effect" This breakthrough has triggered what experts call a "prisoner's dilemma" in the AI industry. Major tech companies now face pressure to invest in similar large-scale computing infrastructure, with potential investments reaching hundreds of billions of dollars. The stakes are enormous—whoever achieves artificial super intelligence first could create hundreds of trillions of dollars in value. This development marks another instance of "Elon Musk Effect" in which Musk's companies continue to defy industry expectations, though it's important to note that while Musk is credited with the initial concept, the implementation required the effort of hundreds of engineers. The success of this approach could reshape the future of AI development and computing architecture. Read the full article