Search this site

Note from editor: sharing because independence. By Aaron Tan ComfortDelGro (CDG) plans to start moving applications off the public cloud now an upgrade to VMware Cloud Foundation (VCF) 9.1 has given its private cloud the ability to run containerised workloads in production. Its IT team completed the first leg of the upgrade the night before Dirk Paul Sibiet, CDG’s group CIO, spoke to Computer Weekly. Work on the second availability zone follows over the coming weeks, and new development will now come back on-premise. https://www.computerweekly.com/news/366651901/ComfortDelGro-to-pull-workloads-back-from-public-cloud

Note from Virtified: We have a similar concern; "Build it and they will come" can be a path fraught with danger and wasted time, effort and budget. By Keith Townsend During an unrelated Buyer Room session, one of the delegates described what their platform team had built: a representative private cloud with the ability to burst to the public cloud. That’s the reference architecture we’ve seen for the past ten years. IT provides a cloud-like experience to developers through a platform team that normalizes the public and private experience behind one API. The project failed. Developers didn’t adopt the private API. We’re about to see the same motion with AI adoption and development. Read Keith's full article at https://thectoadvisor.com/blog/2026/10/07/private-ai-is-repeating-private-clouds-mistake/

By Anthony Macarayan Everland had about three months to leave its VMware environment without interrupting park operations. Western Union faced a different constraint: Its ageing applications and operations across more than 200 countries made the choice of what to migrate less straightforward. Both companies chose Nutanix, but their accounts of the move show why a deadline and a migration plan aren’t the same thing. Read the full article here

Full disclosure: I know many of the authors and vendors involved but I have not had any discussions about the MQ with them. I have zero 'insider knowledge' on this Magic Quadrant (MQ). Gartner has just published the 2026 Gartner Magic Quadrant for Server Virtualization Platforms. This is good news because it helps increase awareness of options in a mature market that has been turned completely upside down. Alas, the analysis would have been even better had it been published two years ago. Adding insult to injury, so to speak, I note that it was published more than two weeks late. I would bet my left arm that this means there was escalation from one or more vendors. It's not hard to guess who that would've been. If analysts can analyse markets then why can't other analysts analyse the analysis? So here's my first take, which gives me no great pleasure in several cases. Note: my analysis is based exclusively on the 'dot placement' that I have seen. I don't have access to the report. The first ever Virtified Microscope? Leaders There's too much to say about the hallowed top-right Leaders quadrant. Let's just say this is the box for the big dogs. Virtified's verdict: Gartner gets 2 right and 2 wrong. Overrated: Microsoft's vision is far from deserving a leadership position and their execution is poor. They are focused on Azure public for all things infrastructure, which is fine but Azure Local is not proven despite a decade of attempts. Windows Server has been forgotten. Microsoft has much to offer but struggles to maintain focus or remember how important these platforms remain to its traditional infrastructure customers. 'About right' Nutanix is no doubt the strongest alternative to VMware. They are the proven alternative and continue expanding their portfolio (containers, AI/inference, database) whilst addressing technical blockers (external storage). It has become a stack suited to the enterprise data centre. Customer support has always been positive. Cost is the main complaint. Concerns about an acquisition continue, but they are unfounded; an acquirer would not purchase Nutanix to replicate Broadcom's playbook. Red Hat is...probably about right but the offering is unproven (yet). If you're looking to build a container-first post-VMware environment then it makes sense. If you're trying to displace Broadcom then there's work that needs to be done. Underrated: Like it or not, VMware still has the highest functionality. They also have the largest revenue and market share. They are not making any friends but that's a—necessary—different discussion. Niche Players The niche category is poorly understood. It can contain solid offerings well suited to particular needs, as well as emerging vendors that may become future leaders. Virtified's verdict: Hard to call right from wrong but it doesn't do justice to some innovative options. 'About right' There's not much to say about Citrix Hypervisor. The most likely use case is to support Citrix on the desktop, which has its own (pricing/support) challenges. The engineering commitment seems patchy (at best). It feels slightly inaccurate to label Scale Computing as merely 'about right', because in a remote office environment they warrant serious consideration. Use cases matter. Underrated: They are undoubtedly the hero of the home lab but Proxmox should not be underrated. For too long, even I ignorantly dismissed them based on their unconventional approach to support [see note at bottom]. The team at VATES continues to develop XCP-ng as a cost-effective open-source hypervisor with enterprise support, while demonstrating ongoing innovation in areas such as new processor architectures (Arm, RISC-V). I confess I have a soft spot for VergeIO. Their product is fit-for-purpose and if your needs are storage-centric then they are worthy of consideration. Likewise, if you are a service provider then their multi-tenancy is worth a look. Challengers Once you get past the top right and bottom left, we head into confusion. Virtified's verdict: Gartner gets 1 right, 2 wrong. Overrated: It's fair to say that HPE is in a category of its own. Their virtualisation layer is rudimentary and won't win many head-to-head evaluations but the management layer is solid. Morpheus makes HPE worthy of consideration for a heterogeneous play, particularly for those with HPE hardware and/or a need to move large amounts of low-risk, non-prod workloads off VMware. 'About right' It always makes me chuckle to say that Oracle is one of the most cost-effective options. Their platform uses oVirt, which is what Red Hat used with their previous virtualisation product. Underrated: All the cool kids have moved on from OpenStack but Canonical has a competitive product. Visionaries This is the box for the 'not-quite-there-yet' vendors. Virtified's verdict: Gartner gets 2 right, 1 wrong. Overrated: There's just not enough evidence to put Platform9 in the visionary quadrant. The company's origins and leadership stem from VMware (in the good ol' days) but they are yet to find their place. 'About right' Historically SUSE has been hypervisor-agnostic, which means they are still building maturity. They deserve credit for their container capabilities and for Harvester. They have the pieces but they are not a proven server virtualisation platform today. I'm tempted to say they're slightly overrated. I've known Sangfor for a long time. They have been in the market for more than a decade and consolidate their expertise with deep understanding of networking, security and desktop infrastructure. They're also one of the first to offer memory tiering, which is a credible example of their innovation. Organisations across South-East Asia often ask about them. Others No comments from me I haven't done enough recently to comment on IDT SYSTEM or H3C Technologies. It's a gross generalisation but their main territory is mainland China and I don't feel that I can do that market justice. Missing If Scale Computing is included then why isn't StorMagic? If you have SUSE and Red Hat, which both use KubeVirt, then why not Spectro Cloud or Kubermatic? Gartner often includes 'honourable mentions' and perhaps the following are there. I can't say. The Bottom Line It's almost impossible to draw the line between containers, virtual machines and hybrid infrastructure today. Why bother? Even if you could, the need for inference infrastructure will change everything. It's not about the analyst's artificial and arbitrary boundaries—and endless definitional debates. Enterprises are looking for a modern hybrid infrastructure stack that suits their use case(s). So I look forward to seeing the Critical Capabilities companion doc some day (i.e., the "thinking person's Magic Quadrant"). It's a far more useful piece of research for a customer IMO. Use cases are critical; it's why Virtified Loops are based on use case. Note 1: Proxmox support It's good to see that my former employer has made note of Proxmox's pivot (see First Take: Proxmox Removes Objections With Enterprise 24/7 Support) but that problem wasn't really a problem. Proxmox already has a viable ecosystem of support providers that gives them the scale they need.

Snippet of CISPE press release issued on 15 September 2026 Broadcom continues to extend practices that limit choice, raise costs and lock in customers whilst destroying the economic viability of many European cloud providers’ business models. As such, its Red status – denoting a critical situation with little or insufficient progress in addressing concerns – is reconfirmed. The Fourth ECCO Report details that Broadcom has: Excluded the vast majority of European CSPs from offering VMware services under any terms; Removed end customers’ Bring Your Own License (‘BYOL’) rights, restricting them to Broadcom’s own choice of a few hand-picked partners to host their workloads; Imposed relentless unilateral contractual changes and cost increases that are driving European CSPs out of the market. Meanwhile, Broadcom is waging ‘lawfare’ to obstruct investigations of its practices, including denying European Commission access to critical documents, whilst using marketing and PR to cast itself as a champion of European cloud and sovereignty. Recent licence changes at SAP have also raised concerns among customers and competitors in Europe, several of which the Fourth ECCO report details. ECCO has given SAP an Amber Status indicating that more must be done to protect choice and fair software licensing in Europe. Its concerns fall into two main areas: Anticompetitive licensing measures that direct customers to SAP’s own cloud offers to the detriment of independent European cloud vendors; and, Emerging concerns over restrictions to agentic AI data access. Read the full press release at https://www.cispe.cloud/broadcom-doubles-down-on-harm-to-european-cloud-and-sap-extends-unfair-licensing-behaviour/

By Simon Sharwood VMware's virtual disk development kit (VDDK) is an SDK that a Broadcom technical document describes as “a set of libraries and utilities that enable software developers to create applications that can access and manipulate virtual disks used” and also “provides APIs for backup and recovery, cloning, and other operations on virtual disks.” Those qualities mean VDDK is also useful when moving VMs from VMware to another virtualization environment. “A vendor is entitled to change how it distributes an SDK...Doing it without a word, while the Apache CloudStack administration guide and other projects’ install instructions still tell people to click a link that no longer resolves, is something else.” - ShapeBlue Read the full article here

CapGemini's Research Institute reveals four key themes shaping organizational approaches to digital sovereignty: Digital sovereignty has entered the mainstream as a strategic business priority Digital sovereignty focuses on resilient interdependence rather than absolute independence Organizations face significant technology dependency risks Organizations are taking a pragmatic approach to their pursuit of digital sovereignty Download the full report from Cap Gemini here (13MB) Note: Research found via an article by Carly Page on The Register here

By Joseph Gabriel Lagonsin Alibaba Cloud and Cambricon gain seats on the PyTorch Foundation Governing Board and Technical Advisory Council. Ant Group joins as the foundation expands the roster of companies involved in developing the PyTorch software ecosystem. For Alibaba Cloud and Cambricon, Platinum membership brings formal influence over governance and technical direction. Both are expected to work with other members on support for different hardware platforms and accelerators across the PyTorch ecosystem. Read the full article here

The 2026 ESOO Data Centre forecasting overview, from the Australia Energy Market Operator (AEMO) provides sober reading. Quote: "There are around 165 data centres [in Australia] that consumed approximately 5 terawatt hours (TWh) of electricity in financial year (FY) 2026, equivalent to around 3% of total NEM operational electricity consumption". If AEMO's forecast proves correct, by 2036, data centre electricity consumption in Australia (13%) will almost double that of electric vehicles (7%). This is occurring at the same time that Australia’s backdown on data centres allows them to use energy from coal and gas. Here’s what that means

Quote: "The [Australian] federal government’s previous position required new data centres to pay for grid upgrades, avoid pushing up household bills and source additional clean energy, rather than simply rely on existing supply. However, the new standards grant carve-outs...that must not compromise the government’s goals of lowering climate emissions and ensuring that energy remains affordable".

Fastly's Xcelerate conference occurred in Sydney on 20 August 2026. Why was Virtified there? You might wonder why we would attend a Fastly conference. The answer is simple: we're always looking for significant technology trends and developments. Distributed infrastructure: What trends might influence this 'mid-cloud' distributed architecture'? We look for trends in the space between centralised hyperscalers and 'the edge' (be that a device, an enterprise data centre or anything in between). Although the necessity is clear—for gaming, online retail, streaming and other content-heavy workloads—this architecture is yet to find a problem needing to be solved for most enterprises (examples: financial services and public sector). WebAssembly: Is anything changing with Wasm? We continue to track the potential of WebAssembly (Wasm). This includes Fastly's pioneering and continuing use. In broad strokes, Wasm is a technology that could rival containers whilst also abstracting the application runtime. But Wasm has limited adoption beyond the browser. Not an official mascot, but these cute plushies were offered to those attending the conference in Sydney's Sofitel hotel. Who Fastly serves today The event included speakers from Stan, New Relic, freelancer.com and amazee.io. These are solid examples within Fastly's area of core competence (and, admittedly, not our area of focus). Where Fastly could go Once we get through the hype, the future is AI. In that light, Fastly co-founder Tyler McMullen clearly differentiated what Fastly wasn't planning. Quote: "We're not doing GPUs at the edge". Reasons that we agree with include that GPUs are too expensive, are in high demand, are in short supply and they're not necessary in that mid-cloud architecture between frontier models and on-premises inference. He didn't discount that this plan may change; we assert that it will (and before 2030). What Fastly and its competitors have is fast connectivity to hyperscalers and distributed points of presence. In an anywhere-to-anywhere architecture, organisations need a control point The similarities to zero trust architecture (ZTA) are readily apparent (verify, don't trust; segment; assume the worst among other principles). But these are, by design, concepts related to secure networking. Determining intent and enforcing AI policy will be key to charting the path for inference beyond the hyperscale data centre. Call it an AI gateway, a policy agent. Other drivers include: The need for predictable AI costs through routing of queries originating outside the perimeter to appropriate models^. Targeted use cases that don't need frontier models. Small and micro models show the other end of today's AI world. Models such as Qwen and Phi. A simple demo showed the potential for mid-cloud-architecture to use natural language processing for language translation, for efficiently filtering inappropriate queries and for applying business logic. Physical AI wherein planet-scale deployment would demand a multi-tiered approach. A non-update on WebAssembly Virtified did not find any major changes in how Fastly uses Wasm. It remains intrinsic to their platform and formed part of the demo mentioned above, showing how Fastly could play the role of AI gateway. ^ It's arguable that public cloud suffered the same challenge yet found ways to sufficiently sidestep the issue through discounted bulk-buying. Perhaps experience with FinOps and cloud expense management will expand to cover the variable portion of AI expenditure. That would make sense...in theory.

By Simon Sharwood Linux kernel boss Linus Torvalds thinks the project has reached a new normal of supersized updates, thanks to AI. He’s previously railed against the idea that good documentation can explain the work of developers who use AI to contribute poor code to the kernel, and also complained that the Linux kernel security mailing list has become almost unmanageable due to a flood of AI-generated bug reports, many of them duplicates. Read the full story here.