Today's tech conferences are a dangerous place. Vendor booths—spanning from has-beens to wannabes—attempt to lure attendees with everything from socks to games to LEGO. In exchange, you become a marketing lead and target for their AI pitch and how they stand at the forefront of the AI revolution. But any seasoned IT professional can smell that BS from a mile away. It's almost always raining in Sydney when there's a cloud conference Enter Google Cloud Summit Sydney 2026. With some notable exceptions—particularly in retail—Google Cloud had a tough fight competing against AWS and Azure to win medium and large enterprises during the last decade. There were many reasons for this (maturity/fit being at the top of this list). What's changing? Customer growth. For many years, and for good reasons, I was a tough critic. However, as we enter the AI era, the battleground is changing. The enterprise IT market is starting to come to Google. The proof is in the financial results (approximately 45% YoY growth in 2025). Why? Google is known for its strengths in data and analytics but it struggled against the IaaS-centric AWS juggernaut and Microsoft's reach. Today, along with Google's other AI assets, Gemini is transforming GCP into a more cohesive and attractive proposition. It's a modern (open) stack that focuses on information and insight rather than infrastructure for legacy workloads. One of the most notable takeaways from the event, for me, was the expanding set of enterprise customers up on stage. Alongside retailers such as Bunnings, K-Mart and Woolworths were IAG, Transurban and Macquarie Bank. "Do your research" Nobody is perfect...but when you operate at hyperscaler you need to be. In this light, who will forget Bard's awkward missteps or questionable Gemini demos that came at almost exactly the wrong moment for Google? In soccer parlance^, these were 'own goals' that distracted many from what has always been one of Google's strengths. If you want to understand the depth of AI expertise within Google then I recommend watching The Thinking Game, which introduces the viewer to Demis Hassabis and takes you inside the origins of DeepMind (seemingly available on YouTube and the link to the trailer is here). ^ It is World Cup time, after all Disclosure: Google is not a client of Virtified...and it wouldn't matter if they were. Our research and insights are self-funded and 100% independent of vendor influence.

By Simon Sharwood The roots of the dispute are a January 2021 contract that saw Tesco acquire perpetual licenses for VMware's vSphere Foundation and Cloud Foundation products, plus Tanzu subscriptions. Read the full story here.

By Joab Jackson The AI bills at Netflix may not be so eye-popping thanks to company senior engineer Tejas Chopra, who has created software to prune agent instructions, as measured in tokens, before they hit the LLM. Read the entire article from El Reg here.

At long last Microsoft has announced that Azure Local expands to sovereign-scale infrastructure with disaggregated deployments. Given how many times their on-premises cloudy offerings change names, hopefully we'll get a shorter name than "disconnected operations for Azure Local" (doAL). General availability is long overdue and has the potential to help close the gap to Nutanix, Red Hat and VMware (among others). Key use cases are likely to include: Remote locations: This deployment allows organisations to run workloads and utilise Azure Arc services without relying on active internet connectivity. Because network latency and reliability are frequent showstoppers for cloud adoption, this is a major win for sizable edge environments. Is it overkill for small remote locations? Yes. Why? doAL has a minimum of 3 nodes each requiring 512 GB memory, 24 cores and 8 that must be at least 2TB each. This beefy hardware footprint underscores the ongoing market need for traditional Windows Server deployments or non-Microsoft alternatives ( Linux ). High control / compliance: This addresses regulated industries, legislative mandates, high control and (some) sovereign infrastructure requirements. Virtified groups these together, and avoids buzzwords along the way, because they are effectively variations on the same theme...regardless of what most vendor marketing would have you believe. For the sovereign use case, digital public goods such as SUSE will tick more boxes (outside USA). Next-generation workloads: This includes emerging distributed, private, and local AI architecture—which is more accurately termed inference. The notable exception, for now, is bundled containers and native Kubernetes capabilities. No doubt this will be resolved before long, given that this functionality has been available from Microsoft elsewhere. Connected versus disconnected (source: Microsoft) Supported Azure services The offering supports these Azure services. It doesn't need to support all of Public Azure but it needs to deliver key management and reliable capability in a secure manner. From a management perspective, doAL includes Azure portal, Azure Arc, Azure Portal, Azure PowerShell and Azure Resource Manager (ARM). Of most note is Azure Local VMs. Notably absent at announcement was 'Azure Kubernetes Service (AKS) enabled by Arc for Azure Local', which is in preview (hence not considered by Virtified as being ready). In coming weeks, Virtified will publish a Virtified Loop that evaluates disconnected operations for Azure Local. We will explore the gamut of inclusions and exclusions in detail. Pricing and licensing Although $10 / core / month sounds appealing it's all the add-ons that will hurt (Azure Arc: $6 / server / month; Microsoft Defender for Servers $4.90 - $14.60 / server / month). Note: above prices are Central US region. Azure hybrid benefit may be a licensing option and you'll need a Microsoft Customer Agreement for enterprise (MCA-E). After that, you will need to fill in a form and meet one of the following criteria (or similar in the discretion of Microsoft):  zero / limited connectivity to the internet, "sovereign requirements", strict regulations or restrictions prohibiting sending any kind of data back to the cloud. As with most Microsoft products and services, pricing remains a mystery beyond the capability of Virtified. Seek expert Microsoft licensing advice. Always. Note: Don't forget hardware.

On Monday, Simon Sharwood (Asia Pacific editor at The Register aka 'El Reg') published an article titled Quit VMware and you’ll emerge with more complex and less capable infrastructure. Up front I need to disclose a long-standing personal association with him but more so that I consider him one of the best IT journalists in the industry. His article reports on a Gartner presentation by Paul Delory (someone I worked with at Gartner for many years). The session was called How to Replace VMware in the Enterprise Data Center (see bottom for a copy of the session description; provided here in case it disappears at some point). 'El Reg' can be a little cheeky with their titles but that's not material. As an ex-Gartner analyst, I'm sufficiently well positioned to comment on how Gartner undertakes research. Likewise, over the years I've had many discussions with seasoned industry experts who voluntarily disclose their respect for The Register. I've been interviewed by them and have read their articles for decades. As a result of the article, many across the industry have asked my position. The above suggests I am sufficiently well placed to comment. Read on for my take. Summary (tl;dr) Almost every considered response to the article that I've seen has some merit Now that may seem to be fence-sitting but you need to look beyond the headline. A great analyst, Dale Kutnick, founder of META Group, taught his analysts: 'make a call'...so please be patient. As with many things, the truth is somewhere in the middle and context is critical. Context Let's start with the protagonists. Paul knows virtualisation. I wasn't at his session but the title and description suggest it covered the effort and challenges involved in replacing VMware. I trust Simon's reporting. By his own account, the session was not focused on alternatives. It was on replacement. Other sessions at the conference covered competitors. Virtified's position VMware reduction and replacement share some overlaps but will lead to fundamentally different outcomes. Until you eliminate VMware you will be continue to be subject to Broadcom's licensing and pricing. This applies equally, and often more so, for those using CA or any other Broadcom software. All existing evidence of Broadcom's behaviour suggests you will pay more each renewal and largely irrespective of how much you use. This negates much of the potential benefit of reducing. Organizations that decide to reduce their VMware footprints will emerge with more complex infrastructure. Much of the industry will end up in this position; Virtified's statistically-valid and self-funded primary research (n=450) shows that approximately one-half of medium and large organisations plan to reduce their VMware footprint before 2028. You can successfully mitigate the impact of the extra complexity. Any reduction will require heterogenous management because the result is a multi-platform environment. The alternative is siloed management and duplicated effort. If you acknowledge this: You will be in a better position to begin to mitigate any potential loss of capability. You can begin planning to successfully reduce your dependence. Heterogenous solutions exist and more will emerge to address this need. Morpheus, now part of HPE, is one such example. All of the above has precedent (example: approximately four out of five medium and large organisations are multicloud). It has and can be done. Recommendations If you acknowledge that complexity is the price of replacement or reduction, Virtified asserts that the following architectural sequence is required to reduce the impact: Start with the control plane. That means vCenter, vRealize, Aria etc. This is a critical step. Heterogenous management is essential. Eliminate Tanzu at any stage. Experimentation was moderate but large-scale production deployment was low. If necessary, remove NSX. This is the most 'sticky'. Determine the impact to existing storage. This includes external storage hardware compatibility (potentially fibre channel) and/or vSAN migration. The last step is the virtualisation layer (vSphere). This must be the finish not the start although experimentation and low-risk reduction is recommended. Similar to storage hardware, asset lifecycles / refresh and hardware certification will be key factors driving timeframes. However, steps (a) through (d) will keep you busy likely for 2 - 4 years. Alternatives vSphere Enterprise Plus was deployed in most enterprises whether it was needed or not. Why? They were paying for it. Too many VMware customers became fat and lazy, conditioned by the dicates of their enterprise licence agreement (ELA). They chose the premium virtualisation platform and settled on the 'highest common denominator'. But the situation has now become worse. Many organisations tell Virtified that VCF was thrust upon them whether they wanted it or not This was among Broadcom's most cunning acts. As can be inferred from our recommendations, VMware's portfolio evolved to become four different solutions (management, network, storage and compute). But, pre-Broadcom, VMware never managed to attract sufficient adoption beyond vSphere, vCenter and (in some cases) vSAN. Broadcom doesn't give VMware customers a choice. VCF compels them to deploy everything. Admittedly, Broadcom has invested to improve the integration between what were disparate and loosely-integrated products. Competitors are right to call out that there are credible VMware alternatives. Virtified Loops suggest that there are serious contenders. Alongside VMware we have evaluated virtualised and hyperconverged infrastructure (VHCI) offerings from Canonical, Microsoft, Nutanix, Oracle, Platform9, Proxmox, Red Hat, VATES and Verge. With proper due diligence, all can reduce the dependence on VMware. There are valid software-based alternatives. Their suitability will depend upon your needs and goals. Unfortunately too many organisations remain blissfully unaware of how their success will depend upon this. Virtified research shows that 76% of organisations agree that alternative VHCI software vendors can meet most of their current needs There are cloud-based alternatives. Their suitability will depend upon your needs (including the ability for you to use cloud, which can be a function of technical and compliance requirements). Hyperscalers have been successfully migrating VMware workloads for over a decade. The biggest competitor for VMware alternatives is organisations that 'do nothing' - Anonymous (see note b at bottom) The challenge, as Paul pointed out, is that finding and successfully migrating to a replacement is more difficult. You need to take a step-by-step approach and accept the pain along the way. The good news is that I've started to see the first successful examples of large-scale replacement. There will be more. Final note: the importance of independence In no set order, and broader in scope than the above, but important nonetheless: I refute any assertion that Simon’s reporting is unduly influenced by commercial interests. While The Register’s business model relies on advertising, their editorial strength lies in maintaining a fiercely independent voice—an increasing rarity as other revenue sources for journalism evaporate. Similarly, while I have legitimate concerns at some aspects of Gartner’s published research and process, I refute the 'cash-for-comment' narrative. The 'cash-for-comment' allegations leveled at Gartner and The Register within the comments section of the article are, in my view, misplaced. No doubt these comments reflect exposure to vendor spin and industry hype. This is understandable but their opinions are misplaced in this situation. There were also cheap shots at Gartner and no doubt Paul would scream out about all the nuances that couldn't be covered in a 45 minute group presentation. Bottom line: the IT industry needs strong independent voices backed by fact-based research with sufficient diversity Both Gartner and The Register serve this purpose (not exclusively but certainly among the best). Virtified aspires to do and be the same (perhaps adding a little oversight along the way). Notes: a. The details on Paul's session are copied verbatim below. How to Replace VMware in the Enterprise Data Center Paul Delory, VP Analyst,Gartner Broadcom's acquisition of VMware has increased prices, uncertainty and risks. Many VMware customers are looking for alternatives. We discuss a detailed, step-by-step plan for migrating off VMware. This includes scoping the move, selecting alternatives, designing and sizing, replacing missing capabilities, migrating VMs, optimizing both new and any remaining VMware environments, and translating operations. b. I'm anonymising the source of this quote only to avoid potential blowback (but I have let them know)

By Simon Sharwood [Virtified editor: see disclosure at bottom. Also: minimal tweaks made to following synopsis] A Gartner analyst at their IT Infrastructure, Operations & Cloud Strategies Conference in Sydney stated there is no technical reason for VMware users to adopt a rival hypervisor, and that no vendor offers a one-for-one replacement for the virtualization pioneer’s flagship Cloud Foundation (VCF) suite. The analyst went to say Broadcom’s licensing policies, which see it only sell VCF, mean VMware users’ licensing bills typically rise by 300 to 400 percent Read the full article here Disclosure: It's a small world; Virtified has longstanding relationships with both the analyst and the journalist in this article.

Disclosure: the following is offered purely for informational purposes. It is 100% independent analysis without any form of remuneration (noting that Virtified was an invited guest). Over the years I had been invited to attend Nutanix .NEXT many times. At long last, in 2026 I was able to attend the event for the first time. Here's my trip report, for your information. Momentum Over 5,000 onsite attendees and more than 100 sponsors^. A high level of audience engagement within the breakout sessions I attended...with large groups of attendees seeking extra details from the speakers. Expanding partner and alliance ecosystem. Only a few years ago it would have been unthinkable to have NetApp and Nutanix onstage together but in 2026 it builds on top of Nutanix's expanding relationships with the likes of Dell and Everpure (previously Pure Storage). It is also demonstration of the evolution of Nutanix's strategy. As a life long fan of the Blues Brothers, it was great to get to the House of Blues alongside admiring the local architecture. Announcements I wouldn’t necessarily say that any of the announcements were revolutionary...but I would say several demonstrated gradual evolutionary advancement. Top takeaways from me: Expanding hardware compatibility list, which will make it easier for new customers to migrate. By supporting additional servers, depreciation schedules and hardware refresh become less of an obstacle for organisations wishing to migrate to Nutanix. Incremental product improvements across the portfolio including Nutanix Enterprise AI (NAI), Nutanix Database Service (NDB) and Nutanix Unified Storage (NUS). These showcase an IT infrastructure vendor that has moved from solving a specific problem toward offering a ‘stack’. Platform philosophy: Virtified’s position is that the majority of enterprises are hybrid and multicloud. They will continue to have on- and off-prem workloads, running physical and virtual, operating from the most remote locations all the way to centralised hyperscale. Nutanix calls this ‘run anywhere’; Virtified calls it a pragmatic necessity. The big surprise: open source I may be cynical sceptical on occasion, I confess, so it was a surprise to rectify some incorrect assumptions. Although not an open source company, I was surprised to learn of Nutanix’s contributions to projects such as KVM, QEMU and Libvirt. You could argue that that’s the least Nutanix could do…but the reality is that it isn’t. You can argue that isn't a high bar to jump and that it 'damns them with faint praise', as the saying goes. I'm not an open source lawyer but they are not obligated to contribute and could do less if they wanted—as is the behaviour with some vendors. Perhaps I'm overcompensating for having assumed they didn't contribute. Irrespective, and even though that was not even the reality in the past, acquiring the assets of D2iQ makes Nutanix more intertwined in the open source community, including the CNCF ecosystem and projects such as Kubernetes, Cilium and others. I got the feeling there is some interest within product teams to make Nutanix’s position clearer. I’d encourage that – a moderate search found nothing substantial that I could refer to (beyond some of my crappy photos taken in breakout sessions). Nutanix should work to make their position on open source clearer. One more thing... Just like the conversations about open source, there was general acknowledgement of the fact that Nutanix Community Edition can be more of a detriment than a benefit. I was pleasantly surprised to hear that the problem is known, acknowledged and should be addressed in coming months. The sooner the better, in my opinion. Notes: Keynotes are available at https://www.nutanix.com/next/on-demand. I imagine that (some) breakout sessions will be published online at some point. A cursory scan didn't find any yet. ^ I didn’t count attendees or sponsor booths but it looked about right so I’ll take Nutanix’s word for it.

Virtified’s goal is clearly posted on our website here . Over the course of the last year, a key part of kickstarting that goal has been forged at my desk, in my lab and in the cloud. And it's finally launched. A Virtified Loop is a thorough, 100% independent analysis of an offering and provider in a specific use case. You can read all about it in the FAQ . More simply, what I'd also say is that Virtified Loops have been a labour of love built to kickstart achieving our goal. They build on more than 25 years of analysing technology offerings. Some fun facts about Virtified Loops: They are an automated, next-gen analysis tool They are currently throttled such that in less than 15 minutes they can analyse over 100 technical features from 10 offerings (faster if I wanted) Over 180,000 automated AI prompts were run to perfect the process and the tool It's AI-powered... and that's OK ...there's almost no way it could do what it does before LLMs (and agentic web search) It has been tested and operated on five different AI platforms (to introduce resilience, variety and reduce bias) No animals were harmed in the process...but one graphic designer almost went mad listening to me try to describe my concept for the key visual (the 'Loop Device') And here's the kicker: There's no charge to you to access Virtified Loops They've taken me a lot of time and effort (most of it fun!) I consider them my gift to an industry that has given so much to me over the course of my career. I decided it's my time to give something back. As a result, there's no 'emailwall', collection of personal data, cookies or any other insidious form of lead generation. As a wise analyst once said to me "Give something away to get something back" . What I want in return is getting closer to Virtified's goal...and for that I'll need feedback (good, bad, ugly). Have at it.

AI-generated portrait of Duncan Epping (taken from photo on Yellow Bricks website) Duncan Epping  and Jad El-Zein recently did a podcast ( here ) based on Jad's July 2025 blog post ( here ). The content puts forward an argument for VCF in the new world. A brief summary, courtesy of one of Virtified's army of junior researchbots , follows. Listen and/or read at your own choice. They discuss VMware Cloud Foundation (VCF) 9 as an integrated infrastructure platform for hybrid workloads, combining Virtual Machines (VMs) and Kubernetes containers on a single virtualisation stack The content explains the Supervisor and VM Service, which expose VMs and Kubernetes clusters through a common declarative Application Programming Interface (API), namespaces, and projects. This enables shared networking, security, and automation constructs The authors describe NSX-based networking—including Virtual Private Cloud (VPC) constructs, load balancing, distributed firewalls, and Source/Destination Network Address Translation (SNAT/DNAT)—and vSAN Express Storage Architecture (ESA) as transparent, policy-driven services for cloud-native workloads The discussion details VCF Automation and VCF Operations as the control layer for multi-tenancy, blueprints, GitOps (e.g., Argo CD, Harbor), cluster lifecycle (VMware Kurbernetes State Manager (VKSM), formerly Tanzu Mission Control), and full-stack observability The authors advocate for running Kubernetes on hypervisors rather than bare metal, citing two decades of VMware performance, availability features such as vMotion and High Availability (HA), and alignment with the virtualised model of public clouds Virtified will share its opinion on the material separately (focusing on the broader aspects of items 1 and 5). Footnotes: Virtified is 100% independent and does not  do 'cash for comment'. Never. The purpose for sharing the article aligns with our belief that not all value comes purely from analyst opinion. The referenced article provides technical insight into recent initiatives in and around VCF, including the ideal of VMs and containers living happily together (something we once called container-VM convergence). In this regard, the following is provided purely for informational value (and, thankfully, is largely marketing-free). Anyone who knows the technical side of VMware probably knows Duncan Epping as a person of integrity; he is someone with a long history of involvement with the platform and ecosystem. Virtified was not aware of Jad El-Zein before now...but he seems similar.

By Simon Sharwood The Python Software Foundation (PSF) has an extra $1.5 million heading its way, after AI upstart Anthropic entered into a partnership aimed at improving security in the Python ecosystem. See the full article here Virtified notes the following: Python is a critical component of many technology initiatives and plays an essential role in many (most?) of today's AI platforms. We keep our politics to ourselves but note the importance of what Anthropic has done in the context of PSF's recent funding 'challenge' . Bravo Anthropic .

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.