Editor's note: although the listed writer for this blog is 'AWS', and Werner Vogels works for Amazon Web Services, let's safely assume that these are his predictions and views. Virtified had the pleasure of a one-on-one discussion with Werner (probably eight years ago). Beyond his accomplishments and obvious intelligence, he appears to be a genuine person of high integrity. For this reason, along with the insight contained in his recent blog ( Tech predictions for 2026 and beyond ), we suggest you consider his big-picture thoughts. Namely: Companionship is redefined for those who need it most The dawn of the renaissance developer  Quantum-safe becomes the only safe
  Defence technology changes the world Personalised learning meets infinite curiosity We'll leave it to you to read and consider his predictions. As broad implications for technology, predictions 1 and 5 are of particular interest to Virtified. They are topics are of particular interest to Virtified and we will return to them over time.

Welcome to 2026, which will be a make-or-break year for many enterprise virtualisation initiatives. There will be untold technical obstacles but migration is ultimately an economic and operational risk management decision. Although AI is hyped, Virtified is confident that it can be the "accelerator" for modernisation of virtual environments ... and it aligns with where the money will be in 2026 IT budgets. Virtified has no doubt that the majority of VMware users are unhappy being Broadcom customers. The reasons are well-known and we won’t go into them here. Unfortunately, most enterprise-scale VMware migrations made minor progress in 2025. Why? There is no all-in-one solution for VMware migration. But there are many alternatives for particular environments. As examples: Hyperscale IaaS is a popular option, but it won't solve network latency, network reliability challenges or satisfy complex compliance mandates. 
 Software-based alternatives are either unproven or lack sufficient ROI. Some aren't even enterprise-ready. Containers aren't the answer for existing monolithic applications. Instead, they are the default infrastructure platform for modern custom applications. So, the implications for Broadcom customers are: Assume that multi-hypervisor environments are going to become the norm, and that you must dismantle your VMware-heavy management tooling (if you wish to reduce your dependence). This will take time and money … and it will not yield significant benefits in 2026 ... but it is essential for longer-term success.
 OR Accept the situation and move on to other things (i.e., 'suck it up'). Accept VMware Cloud Foundation and its full-stack value proposition. This goes beyond compute and management to storage and network. This does not mean that resistance is futile. Far from it. In either case, there are simple ways to reduce the exposure and they should be implemented. One example is in development environments, which in some cases can accrue up to 50% of licence costs. Moreover, development environments are a fertile place to build the next generation of on-premises IT infrastructure, where containers, virtual machines, physical instances and bare metal can coexist. Although your next renewal won't see any reduction, it will reduce your exposure. Keep your eyes on the AI prize Meanwhile, the backlog of other projects continues to increase, and the real prize is artificial intelligence. Addressing the skills gap will be critical, as always, but Virtified is confident that the rewards of AI will be worth the investment . Success will need a DevOps culture, platform engineering capabilities and a cloud-inspired infrastructure. Regardless of your server virtualisation strategy, Virtified recommends the following: Building the capability and capacity for on-premises inference Adding agentic AI to accelerate data integration, legacy refactoring and (re)documentation of existing environments Prototyping semi-deterministic business solutions. Such projects must not replace systems of record, but they will enable faster, more informed decision-making that delivers greater business value They will also reduce the dependence on traditional on-premises server virtualisation. If agentic AI can deliver easier data integration in less time then it will allow for the accelerated reduction of legacy environments. Why is AI the real prize? It enables radical modernisation. 
 The bottom line: don't waste 2026.

Source / inspiration: Intelligence Per Watt: A Study of Local Intelligence Efficiency Stanford researchers dropped a research paper in mid-November that challenges the cloud-centric delivery model of AI inference due to good enough price performance versus multi-year construction costs. Here's my bottom line: The "Intelligence Per Watt" framework shows that appropriate AI models running on consumer devices are becoming viable for the majority of everyday queries. The paper indicates that consumer-grade tech such as the M4 Max is becoming viable for the vast majority of everyday queries ... and is improving at an impressive rate. Meanwhile, their TL;DR shows highlights the challenge facing hyperscalers: the laws of physics and construction. "AI demand is growing exponentially, creating unprecedented pressure on data center infrastructure. While data centers dominate AI workloads due to superior compute density and efficiency, they face scaling constraints: years-long construction timelines, massive capital requirements, and energy grid limitations". Although Virtified is a little tight with its capital expenditure in startup mode, experience from the Apple M4 up the back of Virtified Labs shows a real-world example supporting this assertion. We've used ollama quite successfully since July (albeit alongside more than 950,000 automated, multi-model API-initiated prompts run by our junior research assistants. Yes: 950,000+). The key insight is straightforward but game-changing and raises a big question in my mind: Does the centralised hyperscale model hit the wall when decentralised compute can meet the majority of AI demands? Rather than asking whether local models can match frontier AI capabilities head-to-head (spoiler: they can't for everything), the Hazy Labs team ask a different question: can consumer-grade hardware deliver sufficient accuracy within its power constraints? Their "Intelligence Per Watt" (IPW) metric— accuracy divided by power consumption —tracks exactly this tradeoff, and they claim it's improved by 5.3× over just two years (2023-2025). That's driven by 3.1× gains from smarter model architectures (like Mixture-of-Experts) and 1.7× gains from better hardware (accelerators). The practical implications are enormous. Testing against 1 million real-world chat and reasoning queries, local models delivered the following: Successfully handled 88.7% of queries/chats Jumping from 23.2% success rate in 2023 to 71.3% in 2025. Handled over 90% of creative and humanities tasks. Handled more than 65% of architecture and engineering tasks. The paper suggests that intelligent routing of queries might reduce the energy, compute, and cost of inference by 60-80% while maintaining accuracy (quality). It's not binary, of course: anything the local model can't handle automatically gets routed to a frontier model. Sure, cloud-delivered hyperscale still has efficiency advantages—and armies of engineers delivering superior innovation—but the real story here is about overall system-level wins. You don't need theoretical perfection when you've got the majority of queries running on your local workstation solved quick enough, good enough and cheap enough. With some boffins—not Virtified—suggesting centralised global AI infrastructure could potentially demand 50-100 GW of power by 2030, the distribution / decentralisation of everyday inference to laptops and phones isn't just clever engineering—it's becoming an essential infrastructure strategy. PS Bonus round: add the enterprise-grade hardware capabilities to the mix and imagine what might happen. Footnote: Virtified is not good at analogies, similes or metaphors but we will draw a comparison. In 2015 we wrote a Maverick research paper originally titled " Get off the grid before it goes off ". The eventual titled was The Grid Is Dead, Long Live the Grid! We looked at reductions in the levelised cost of energy (LCoE) for a range of energy sources including coal, nuclear, solar and wind. At the time, the missing piece was batteries. Look at what has changed since then: the levelised cost of storage (LCoS) for lithium-ion batteries has reduced by more than 80% .

[Editor: it's getting too easy to find incumbent software vendors trying to dictate strategy to their customers]. According to The Register: SAP insists customers wanting "innovation" such as AI must upgrade to its latest platform for ERP, using prescribed cloud migration plans. Kingfisher - which operates 2,000 European retail stores including UK brands Screwfix and B&Q - rejected that approach. By Lindsay Clark . We've been asking vendors for value. We're happy to have a conversation about your upgrade pathway… about new things you're doing. But show us value. Don't just give me a price list. Don't just present me with a new licensing module that somehow spikes my cost by 20x. - Chris Blatchford (Kingfisher Group CTO) Read the full article at https://www.theregister.com/2025/11/12/retail_giant_kingfisher_says_no/

In a world of Wall- Street-driven fixation with shareholder returns, it's pleasant to stumble upon a smidgen of corporate social responsibility. Namely that SUSE Linux Enterprise Server (SLES) has been verified as a Digital Public Good . SUSE claims to be the first and only commercial Linux product to receive this recognition from the Digital Public Goods Alliance Virtified's take: bravo SUSE. Let's hope this catches on. Read the full announcement here . View the gritty details here .

Pragmatism or self-induced legacy? I was listening to DevOps and Docker Talk 's early October episode titled Move K8s Stateful Pods Between Nodes ^ and it reminded me of some pet peeves. None of them are better exemplified than: "95% of the customers I deal with don't do Kubernetes the right way" - Phil Andrews ( cast.ai ) To me, putting stateful workloads in containers is madness. It’s another example of a recurring pattern I’ve seen throughout my career: the tendency of enterprise IT professionals to reinvent the wheel. Their inner engineer can't resist. Containers are best suited to ephemeral, immutable, stateless workloads. This was elegantly summarised by my former colleague Tony Iams : P rogrammable R esilient I mmutable M odular E lastic D eclarative The podcast covers many of the challenges of stateful containers, including. Networking and connection state management : maintaining TCP/IP connections, session state, and preserving the pod’s IP address during migration is a major technical hurdle. Kubernetes deployments often use different CNIs (e.g., Calico, Cilium, AWS VPC). Reestablishing/replaying traffic flows (especially for long-running jobs and web sockets) demands precise ordering and coordination.​ Storage and data replication : replicating memory and disk state isn't easy (and was a big part of the magic of VM live migration. Migrating persistent volumes involves API-driven detach/reattach steps that add lag/delay. Local ephemeral storage or NVMe disks may require lengthy background replication. Managing database logs, mirroring, and file access safely during migration remains a difficult engineering problem — one that was solved by modern hypervisors 20 years ago.​ Workload complexity and statefulness : many clusters are a mix of poorly architected or fragile workloads, making bin packing and node reduction challenging without risk of interruption.​ Stateful containers are often more like pets than cattle Resource/time constraints : AI/ML/tuning jobs often have large memory/state requirements. There are physical limits to the speed of data transfer or replication. Migrating jobs with hundreds of GBs or terabytes of working data within short time frames can be a major challenge.​ Limits of Kubernetes orchestration : coordination between pod names, API calls, containerd states, and Kubernetes control plane is required so the pod is instantiated in a paused state, assigned the right resources, and then switched over to active. Order of cutover events (pod name uniqueness, IP address transfer, routing table updates, etc.) must be handled precisely—any discrepancy may cause failures.​ Most of these problems were already solved by other technologies. Use the right tool for the job--even if it's not shiny and new. Stateful is ideally scale-up and either physical or virtual. Modern apps are typically scale-out and better suited to containers, serverless (and/or AI). Whether you agree or not, the podcast provides plenty of food for thought...and I fully acknowledge that theory and practicality rarely intersect. For those facing these challenges, it might be a good idea to take a look at what cast.ai has developed. Footnote: I use Podcast Addict , which is free, so I reckon they deserve a free plug in return. Use of their free tool, and the content of the podcast, are the full sum of my compensation for publishing this article. I have no financial relationship with any of the parties involved,

As I prepare for a presentation later this week, it's a good time to put digital pen to digital paper and write down Warrilow's Laws. What are they? They are the key learnings from 30+ years in IT. And here's the first: If you don't know where you're going... ...then all roads will take you there. This one is both obvious and impactful (in fact, they all are). What I didn't consciously realise was the origin; in Chapter 6 of Alice's Adventures in Wonderland, Lewis Carroll wrote the following masterful piece of dialogue: “Would you tell me, please, which way I ought to go from here?” “That depends a good deal on where you want to get to,” said the Cat. “I don’t much care where—” said Alice. “Then it doesn’t matter which way you go,” said the Cat. “—so long as I get somewhere,” Alice added as an explanation. “Oh, you’re sure to do that,” said the Cat, “if you only walk long enough.” And this is exactly what many technology projects and initiatives become: a stronger desire to do something — anything — than to take the time to build a strategy (ie a plan of action). One variant of this is the following, which I learnt during leadership training in my first full-time job: Don't just do something, stand there. This embodies the same concept and is clearly a reversal of the classic " don't just stand there, do something ". But it doesn't start (to stop) there. In Moral Letters to Lucilius, Seneca the Younger wrote: "If a man knows not to which port he sails, no wind is favourable." What does all this prove? Like many things in IT, it's not the technology. Postscript: You'll see me reference this Law as it applies to contemporary examples of Alice's ongoing adventures in a world of wonder.

[Editor's note: I remember Dale Vile and Freeform Dynamics!] By Lindsay Clark A survey of 455 CIOs, senior-level IT roles, SAP specialists, and business managers found that SAP's push to change the way customers pay for software was also unsettling them. Eighty-three percent of respondents in the Freeform Dynamics research said they did not fully understand SAP's latest migration policies and deadlines, while 84 percent expressed concern about current messaging and how it would impact their operations. Read the full article here .

[Note from editor: This post is shared as part of Virtified's mission to keep vendors accountable. The content is not endorsed by Virtified in any manner]. If you dig deep into this Broadcom blog post , which seems almost entirely generated by a non-human, you'll eventually find this little chestnut: "To ensure our customers have consistent user experience, the fastest access to our latest innovations, and maximum flexibility to move VCF environments anywhere they choose, Broadcom is transforming its hyperscaler go to market to a license portability-only operating model beginning in the new fiscal year." Virtified's take: This is yet another example of Broadcom shutting down ways to use VMware without purchasing a fixed-term license. Moreover, it is yet another example that demonstrates Broadcom's disregard for VMware's pre-acquisition partner ecosystem. Rest assured that Virtified will be developing and publishing an official position on the customer impact of this concerning trend. Meanwhile, we will seek and publish responses from affected parties. Full URL: https://blogs.vmware.com/cloud-foundation/2025/08/29/vmware-cloud-foundation-cloud-on-your-terms/

Enables tens of applications per quarter to be modernised. By Ry Crozier The Commonwealth Bank of Australia (CBA) has built a multi-agent workflow that covers every step required to prepare an application for migration to the cloud, enabling the bank to move “20-to-30” applications a quarter. Read the whole article here . [Editor's note] I was fortunate enough to attend the same session. Fingers crossed I'll be able to get permission to write a public case study. What they have done has changed my perspective on what's possible when it comes to modernisation.

The Importance of Sustainable Energy Solutions The challenges of powering the planet's ever-increasing fleet of data centres are significant. Virtified has followed these developments for years. Recently, Equinix made headlines with its commitment to nuclear energy. This isn't the first announcement of its kind; for instance, Microsoft has also taken steps in this direction. However, Equinix's actions are noteworthy. The company has a long and demonstrated track record of taking environmental sustainability seriously. Nuclear energy is a hot topic, especially considering its prominence in the 2025 federal election in Virtified's home country. Virtified considers Equinix's move into nuclear as a (seemingly) necessary sign of things to come. Equinix's Partnerships and Innovations Equinix's strategy involves collaborating with several innovative companies to harness the power of nuclear energy. Here are some key partnerships: Oklo: Pioneering Small Modular Reactors In 2024, Equinix became the first data center operator to sign an agreement with a small modular reactor (SMR) company. This agreement allows Equinix to procure 500MW of energy from Oklo’s next-generation fission Aurora powerhouses. Oklo's fast reactors come with inherent safety features and can be fueled by nuclear waste. This innovative approach not only addresses energy needs but also contributes to waste management. Radiant: Microreactor Technology Equinix has also announced a preorder agreement for 20 of Radiant’s Kaleidos microreactors. These microreactors provide a reliable, long-lasting energy source. They can be transported anywhere, installed in days, and deployed safely alongside existing equipment. This flexibility is crucial for meeting the energy demands of modern data centers. ULC-Energy and Rolls-Royce SMR Equinix signed a Letter of Intent with ULC-Energy for a Power Purchase Agreement (PPA) of up to 250 MWe to power data centers in the Netherlands. ULC-Energy, based in Amsterdam, has selected Rolls-Royce SMR as its preferred technology solution. Rolls-Royce SMR is developing a 470 MWe light water small modular reactor. This partnership is a significant step towards sustainable energy solutions in Europe. Stellaria: The Future of Molten Salt Reactors Equinix has also secured a pre-order power agreement for 500 MWe to expand data centers across Europe. Stellaria, incubated by Schneider Electric and the CEA (French Atomic Energy Agency), is developing the world's first molten salt Breed & Burn reactor. This reactor will breed 100% of its liquid fissile fuel inside the reactor without the need for refueling. It also recycles spent fuels and burns long-life waste, making it an environmentally friendly option. Advanced Fuel Cells: A Complementary Technology In addition to nuclear energy, Equinix has been exploring advanced fuel cells for scalable, efficient, and cleaner onsite energy. The company has been using fuel cells for over ten years in collaboration with various partners. Bloom Energy: Expanding Fuel Cell Deployment Equinix has an agreement to expand its deployment of solid-oxide fuel cells to over 100MW across 19 data centers in six states. This onsite power generation is highly efficient and enables Equinix to avoid 285,000 MTCO2e emissions and 382 billion gallons of embedded water use. This commitment to reducing carbon emissions aligns with global sustainability goals. The Future of Energy in Data Centers Equinix's move into nuclear energy and its partnerships with innovative companies demonstrate a commitment to sustainable energy solutions. As the demand for data centers continues to grow, finding reliable and environmentally friendly energy sources is crucial. Embracing Innovation for a Sustainable Future The integration of nuclear energy and advanced fuel cells represents a significant shift in how data centers can operate. These technologies not only provide reliable energy but also contribute to reducing the environmental impact of data centers. Equinix's proactive approach sets a precedent for the industry. By embracing innovative energy solutions, the company is paving the way for a more sustainable future. Conclusion Equinix's commitment to nuclear energy and advanced fuel cells showcases its dedication to sustainability. As the world moves towards cleaner energy solutions, Equinix is leading the charge. The company's partnerships with Oklo, Radiant, ULC-Energy, and Stellaria highlight the importance of collaboration in achieving these goals. The future of energy in data centers is bright, and Equinix is at the forefront of this transformation. The shift towards sustainable energy is not just a trend; it's a necessity for the planet's future. For more insights, check out the related blog here . ---

By Simon Sharwood Supermarket giant Tesco sues VMware, warns lack of support could disrupt food supply. Goes after Computacenter too, seeks £100 million damages See the full article here .