@HPC Podcast Archives - OrionX.net - HPC News Bytes – 20260629

Episode Date: June 29, 2026

- IBM demonstrates sub-1nm chip tech - IBM as we see it - Executive orders for quantum innovation and risk mitigation - US, UK, EU roadmaps for PQC readiness - ORNL next-gen supercomputer Discovery��s Center for Accelerated Application Readiness (CAAR) [audio mp3="https://orionx.net/wp-content/uploads/2026/06/HPCNB_20260629.mp3"][/audio] The post HPC News Bytes – 20260629 appeared first on OrionX.net.

Welcome to HPC Newsbytes, a weekly show about important news in the world of supercomputing, AI, quantum computing, and other advanced technologies. Hi, everyone. Welcome to HBC Newsbytes. I'm Doug Black, and with me is Shaheen Khan. IBM is not a commercial chip manufacturing company, but it has maintained very advanced research prowess in the microprocessor field and licenses its technology to chip manufacturers like Rapidus, Samsung, and Intel, although not TSMC. Last week, it announced a serious semiconductor milestone. IBM says its new 0.7 nanometer, or 7-Angstrach,

architecture, can nearly double the density of its earlier two-nanometer research chip, packing close to 100 billion transistors onto a fingernail-sized chip, with projected gains of up to 50 percent more performance, or 70% better energy efficiency versus IBM's 2-nanometer class technology. In 2021, IBM was first to produce a 2-nanometer test chip on a 300-millimeter wafer through what they called nanosheets. Then in 2022, it partnered with the Japan-backed, high-end, and well-funded, advanced foundry startup rapidness to help transfer and jointly develop two-nanometer technology

for Japan's emerging advanced foundry effort. IBM also continues to work with Samsung Foundry, which manufactures IBM's recent mainframe chip, Telem 2, and Accelerator Chip, Spire, on Samsung's 5-nometer technology. So IBM's role is unusual but important. It sold its global microelectronics and fab business to global foundries in 2015, but kept the roadmap.

New York State's Albany Nanotech, now expanding with a publicly accessible high-numerical aperture EUB center, gives IBM a place to shape future chip technology without owning high-volume manufacturing. Strategically, we should continue to look at IBM as a research anchor in the U.S. Allied semiconductor ecosystem. You could say IBM has always been about being a trusted enterprise technology partner. These days, we could see seven strategic areas of focus. First, it's IBM Z and power systems in what I call, and maybe they should too, the high-trust infrastructure market.

Mainframes and power systems carry on as important infrastructure for banking, payments, governments, and mission-critical transaction processing. They also create IBM's needs for secure, reliable, high-performance silicon. Second is quantum computing. IBM has one of the most visible quantum roadmacks in the industry with hardware, software, cloud access, ecosystem development, and now deeper fabrication ties to Albany Nanotech. Quantum remains long term,

but it reinforces IBM's image as a deep computing company with a head start on the next wave. Third is Enterprise AI, now centered on Watson X, all lowercase, and agents with a strategy focused on governed, auditable, enterprise-grade AI for banks, governments, insurers, health care, and regulated industries. Fourth is hybrid cloud with Red Hat and OpenShift at its center. If you see Enterprise AI running across public cloud, private cloud, on-prem systems, edge environments, and regulated infrastructure, then a hybrid cloud as the deployment layer, and as provided by Red Hat,

makes perfect sense. Fifth is cybersecurity, governance and resilience, and especially as AI moves into business processes. Companies need identity, access control, auditability, data protection, model governance and operational resilience, and this fits IBM's traditional strength with big professional services and cybersecurity expertise. Speaking of which, the sixth is consulting and global enterprise execution. IBM consulting is the way many of these technologies reach customers. AI governance, cloud modernization, security, automation, and regulated industry transformation, all requires services and integration. The seventh is what we discussed, advanced chip research research as strategic leverage in a broader research agenda. IBM develops frontier chip

technologies such as two nanometer nanosheets and now sub one nanometer nanostack structures and partners with Samsung Rapidis, Albany Nanotech tool vendors and big customers. All of the above is also important in geopolitics and sovereign technology which perhaps forms an eighth area. IBM's global presence and focus on many of the places governments now care about can give it an advantage in AI governance, cybersecurity, quantum readiness, semiconductor, research, hybrid cloud, and critical infrastructure. So the significance of the sub 1 nanometer news is ultimately the reinforcement of being a trusted enterprise technology provider. The U.S. government has issued a pair of quantum focused executive orders, one aimed at accelerating quantum

innovation and the other aimed at securing federal systems against future quantum-enabled cryptographic attacks. Roiders described the orders as targeting a powerful quantum computer by 28 and moving key government systems toward post-quantum cryptography by 2030 or 2031. The Innovation Order directs the government to push toward useful quantum systems, including a quantum computer for application development and discovery science, along with quantum sensing, networking, supply chain work, workforce development, and international cooperation. The cybersecurity order is the more immediate operational mandate. It pushes federal agencies to migrate high-value and high-impact systems to NIST-approved

post-quant cryptography. The U.S. deadlines are roughly 2030 for quantum-resistant key establishment and 2031 for digital signatures. It also pushes federal contractors and vendors toward PQC readiness through procurement and compliance pressure. In practice, this means companies that sell into federal markets, support critical infrastructure, or provide cryptographic products, will need crypto inventories, migration plans, vendor roadmaps, and eventually quantum safe implementations. This aligns with what Europeans are doing,

and it tightens what the U.S. had in place. The UK roadmap is formulated as guidance and says organizations, not just government agencies, should plan by 2028, complete high priority migration by 2031, and complete broader migration by 2035. The EU issued a coordinated PQC roadmap in June 2025, following a 2024 recommendation that member states develop comprehensive PQC strategies. timelines for French policies are similar. Of course, quantum computers continue to look far away from being able to break encryption, but the actions show that the risk has moved from research concern to government timetable. Yes, the U.S. policy really is shifting from research and readiness to procurement and compliance.

Governments increasingly view quantum computing as an imperative and believe they must cover both the opportunity, and the risk. On one side, quantum computing, sensing, and networking could matter for science, defense, materials, chemistry, navigation, and secure communications. On the other hand, a future cryptographically relevant quantum computer could undermine much of today's public key encryption. But quantum is not the only technology that has this kind of dynamics, and governments cannot afford to buy unlimited insurance against every possible black swan, so to say. AI accidents, biotech misuse, cyber collapse, IoT, critical infrastructure insecurity, grid failure, climate shocks, semiconductor disruption, and now quantum code breaking.

So they have to ask which risks have severe downside, which require long lead times, which mitigations are useful even if the worst case does not happen, which sectors or are critical enough to move first, and PQC scores high on those tests. Even if Q Day, so to say, is not near, Harvest Now, Decrypt Later has been a topic for years. Encrypted data is believed to be already stolen to be decrypted later. Migration also has important benefits, better vendor discipline, more resilient security architecture, industry agility or cryptoagility, which is the term of art, and generally highlighting the general cybersecurity issues that are not going to go away.

And then NIST has finalized its first post-quantum standards in 2024. So there is a practical basis for implementation and for procurement and compliance. The geopolitical angle is equally important. The U.S., the U.K. and the EU are converging around the 2030 to 2035 post-quantum transition window. That reflects concern about China and other strategic competitors, but also recognizing that cryptography is foundational infrastructure. If governments move together, they shape standards, supply chains, certification, procurement, and vendor behavior. And equally important, any migration needs to be high quality. A rushed migration could break systems, waste money, create new vulnerabilities, etc.

But a slow wait-and-see model is also dangerous because the work takes years. The prudent model is phased, inventory first, protect the most sensitive systems, require new products to be crypto-agile, use procurement to move markets, and complete broad migration over time with high quality. Oak Ridge National Lab has selected the first nine research projects that will run on Discovery, the next U.S. flagship, supercomputer that is slated for deployment in 2008. According to published reports, Discovery is projected to be roughly three to five times more powerful than Frontier the current flagship system at Oak Ridge,

and Discovery will play a central role in the Genesis Mission, DOE's Scientific Platform Initiative. The website for the Oak Ridge Leadership Computer Facility has all the details. Really looking forward to this system. And of course, also looking forward to continue the action in the top 500 and encouraging our listeners to go listen to our coverage of the last very newsworthy type top 500 results. The selected applications in this case span a diverse set of scientific domains, including astrophysics, molecular biology, quantum chemistry, aerospace engineering, and likewise the teams behind them reflect a broad cross-section of the research community with participants from the industry, universities, national laboratories, NASA, and other federal agencies, quite a multidisciplinary

approach. All right, that's it for this episode. Thank you all for being with us. And thank you, Doug, for dialing in from an airport. Okay, very good. HPC Newsbytes is a production of OrionX. Shaheen Khan and Doug Black host the show. Every episode is posted on OrionX.net.

If you like the show, please rate and review it. Thank you for listening.