Azure Migration Analysis for UK-Based VMware Environments

All private cloud to Azure migration projects to include a repatriation plan

VIRTUALISATIONCLOUD

Jack Jalali

7/9/202516 min read

Executive Summary

This analysis highlights the need for all private cloud to Azure migration projects to include a repatriation plan back to private cloud.

This report provides a comprehensive analysis of the growing trend of cloud repatriation among United Kingdom enterprises, with a specific focus on its implications for law firms and organizations with substantial deployments of Microsoft Windows virtual machines.

The research indicates a significant strategic shift in the broader UK market, moving from a dogmatic “cloud-first” approach to a more pragmatic and discerning “cloud-appropriate” model. Findings reveal that an overwhelming 97% of UK midmarket companies are planning to move at least some workloads out of public cloud environments, driven primarily by escalating costs, performance degradation, complex data sovereignty requirements, and risk management concerns. For enterprises managing large estates of around 1000 Windows virtual machines, the economic and performance rationale for repatriating stable, predictable workloads to on-premises or private cloud infrastructure is becoming increasingly compelling.

Conversely, the UK legal sector currently exhibits a contrasting trajectory, with strong and ongoing momentum towards cloud adoption, particularly Microsoft Azure. Data shows that 68% of law firms now host more than half of their core systems in the cloud, fuelled by the need for enhanced operational efficiency, robust security, and support for modern hybrid working models. However, this report identifies that the very factors driving repatriation in other sectors—such as unpredictable costs, com‐ plex regulatory compliance, and the need for absolute control over sensitive client data—are acutely pertinent to legal and professional services firms. Recent disclosures regarding the limitations of data sovereignty guarantees from major cloud providers present a significant latent risk for the sector. While no specific case studies of UK law firms repatriating from Azure were identified, analysis of challenges within the sector, such as unoptimised “lift-and-shift” migrations leading to excessive costs, suggests a future pivot towards optimization and, potentially, selective repatriation. The technical complexities and substantial costs associated with reverse migration, including punitive data egress fees and the need for specialized expertise, remain formidable barriers. This report concludes that for IT and business leaders, the strategic imperative is no longer a binary choice between on-premises and cloud. Instead, it is a continuous, nuanced evaluation of workload placement to create a resilient, cost-effective, and compliant hybrid IT ecosystem that aligns with long-term business objectives.

The New Imperative: VMware Licensing and Market Dynamics

The landscape for enterprise virtualization has been irrevocably altered by Broadcom's acquisition of VMware. This has created a powerful catalyst for UK organizations to reassess their long-term commitment to the platform and explore alternatives like Microsoft Azure. The once-predictable model of perpetual licensing has been replaced by a new paradigm that directly impacts operational budgets and strategic planning.

The Broadcom Licensing Impact

The core of the change is a mandatory shift from perpetual, socket-based licensing to a subscription-based, core-centric model. This has several critical implications for a company running 300 Windows servers:

Core-Based Subscription: Licensing is now tied to the total number of physical cores in a server's CPUs, with a minimum purchase requirement of 16 cores per CPU. This model can disproportionately increase costs for environments that use CPUs with lower core counts but high performance.
The 72-Core Minimum: Effective from April 2025, a new rule mandates a minimum purchase of 72 cores per license instance. This means a server with a single 16-core CPU must now be licensed for 72 cores, creating a significantly higher cost entry point, especially for smaller or remote office deployments.
Elimination of Perpetual Licenses: The inability to purchase new perpetual licenses forces all customers into the subscription model upon renewal, leading to a shift from capital expenditure (CapEx) to a potentially much higher operational expenditure (OpEx).
Bundled Offerings: VMware has simplified its portfolio into core solutions like VMware Cloud Foundation (VCF) and vSphere Foundation (VVF). This bundling, while simplifying offerings, removes the flexibility to purchase standalone products and may force customers to pay for features they do not need.
Substantial Cost Increases: Market reports and partner analysis indicate that these changes are leading to dramatic cost increases for customers, with some facing renewals that are 150% to over 1,000% higher than previous agreements. For a 300-server environment, this translates into millions of pounds in additional annual costs.

These changes have effectively transformed VMware from a predictable infrastructure component into a significant and escalating operational expense, compelling CIOs and IT leaders to actively seek more financially sustainable alternatives.

VMware-to-Azure Migration: Key Challenges

While the financial motivation to migrate is clear, the process of moving a complex 300-server VMware environment to Azure is a major undertaking fraught with technical, operational, and strategic challenges.

Technical and Operational Hurdles

Application Modernization and Compatibility: Many legacy applications were not designed for the cloud. A "lift-and-shift" migration may lead to poor performance and high costs. True optimization often requires refactoring applications for a cloud-native architecture, a resource-intensive task demanding specialized skills in containerization (Docker, Kubernetes) and microservices.
Data Migration Complexity: The physical act of transferring the data associated with 300 servers—including virtual machine disks (VMDKs), databases, and file stores—is a significant logistical challenge. It requires high-bandwidth network connections (like Azure ExpressRoute) and carefully planned cutover windows to minimize business disruption.
Performance Parity: Ensuring that workloads perform as well or better in Azure than they did on-premises requires careful planning. This involves right-sizing Azure VMs based on actual performance data (not just allocated resources) and optimizing storage and network configurations.
Skills Gap: An IT team highly proficient in VMware vSphere, vCenter, and NSX may lack the necessary skills for Azure administration, security, and cost management (FinOps). This necessitates a significant investment in training and skill development to manage the new environment effectively.

Strategic Decision: Azure IaaS vs. Azure VMware Solution (AVS)

A critical decision for any VMware shop is the migration pathway.

Azure VMware Solution (AVS): This service allows organizations to run their native VMware workloads on dedicated bare-metal infrastructure within Azure. It provides a familiar environment (vSphere, vCenter, NSX-T) and enables a faster, lower-risk migration with minimal refactoring. It is an ideal choice for organizations with tight timelines, complex VMware dependencies, or a desire to retain existing skill sets.
Native Azure IaaS/PaaS: Migrating to native Azure Virtual Machines (IaaS) or Platform-as-a-Service offerings requires more upfront work to refactor or re-platform applications. However, this approach unlocks the full potential of the cloud, offering deeper integration with Azure services, greater opportunities for cost optimization, and a more direct path to innovation with AI and analytics.

The choice is not mutually exclusive; many organizations adopt a hybrid approach, using AVS for critical legacy systems while moving other applications to native Azure services over time.

The Critical Barrier: Data Egress Fees

While migrating to Azure is financially attractive, it is crucial to understand the costs of moving data out of a public cloud, a key factor in any repatriation or multi-cloud strategy. Azure, like other hyperscalers, charges for data egress.

Pricing in the UK (Europe Zone): After a free tier of 100 GB/month, internet egress from Azure's UK regions starts at approximately $0.087 per GB (£0.07/GB) and decreases with volume.
Inter-Region Transfer: Moving data between Azure regions within Europe (e.g., UK South to West Europe) costs approximately $0.02 per GB (£0.016/GB).
Strategic Impact: These fees can be a significant barrier to leaving the cloud, creating a form of vendor lock-in. This practice is currently under investigation by the UK's Competition and Markets Authority (CMA) for its anti-competitive nature. Any cloud strategy must account for potential egress costs in its exit plan.

Key VMware-Specific Challenges

Nested Virtualization: 15-25% performance penalty
Licensing Complexity: BYOL complications and audit risks
Management Overhead: Maintaining both VMware and Azure skills
Migration Complexity: P2V conversion challenges

The Evolving Cloud Landscape: The Rise of Repatriation

The paradigm of enterprise IT has been dominated for the past decade by the concept of a “cloudfirst” strategy. Organizations across all sectors, encouraged by the promise of unparalleled scalability, operational agility, and a shift from capital to operational expenditure, embarked on ambitious migrations to public cloud platforms such as Microsoft Azure and Amazon Web Services. This movement was significantly accelerated during the global pandemic, as businesses rapidly deployed cloud services to support a newly remote workforce. However, as the initial wave of mass adoption matures, a significant counter-trend has emerged: cloud repatriation. This strategic re-evaluation involves moving applications, data, and entire workloads from public cloud environments back to on-premises datacentres, private cloud infrastructure, or colocation facilities. This is not a wholesale rejection of the cloud, but rather a sophisticated recalibration of IT strategy, moving towards a more discerning “cloud-appropriate” model where workload placement is dictated by specific business, performance, and financial requirements.

This strategic pivot is underpinned by compelling quantitative evidence. A recent Citrix study revealed that a quarter of all UK companies are in the process of transitioning the majority of their cloud-hosted capacities back to on-premises facilities. The trend is even more pronounced in the UK’s mid-market sector, which includes companies in the target profile of this report. Research commissioned by man‐ aged services provider Node4, focusing on companies with 500 to 5,000 employees, found that a staggering 97% are planning to migrate some workloads out of their public cloud environments within the next twelve months. While most of these companies, at 49%, are taking a targeted approach by moving specific applications, a notable 5% intend to repatriate their entire cloud estate. Globally, the trend is just as significant, with a Q4 2024 Barclays CIO Survey indicating that 86% of CIOs planned to move some public cloud workloads back to private or on-premises infrastructure, the highest figure on record for the survey.

The drivers behind this reverse migration are multifaceted and reflect a growing disillusionment with the one-size-fits-all promise of the public cloud. The most frequently cited reason is performance considerations. Many organizations that performed a rapid “lift-and-shift” migration found that legacy applications were not architected for the cloud, resulting in poor performance and latency issues that impacted user experience and operational efficiency. The second most common driver, cited by 30% of mid-market firms, is data sovereignty. In an era of tightening regulations like GDPR and the Digital Operational Resilience Act (DORA), and increasing geopolitical uncertainty, UK and European companies are expressing growing unease about entrusting sensitive data to US-based hyperscalers, particularly in light of legislation like the U.S. CLOUD Act. Other significant factors include risk management, cited by 29% of firms, technical limitations of public cloud platforms at 27%, and security concerns at 21%. Perhaps most critically, cost optimization, cited by 26% of firms, has become a major catalyst. The initial promise of cost savings has often been replaced by the reality of complex pricing models, punitive data egress fees, and spiralling operational expenses, with Gartner revealing that companies may be overpaying for cloud infrastructure by as much as 70%. This confluence of factors is forcing a strategic reckoning, compelling business and IT leaders to critically assess which workloads truly benefit from the public cloud and which are better served by a return to a more controlled, on-premises or hybrid environment.

Repatriation Trends in the UK Mid-Market

The UK’s mid-market, encompassing companies with approximately 2000 employees, has become a focal point for the cloud repatriation movement. These organizations are often large enough to have complex IT needs but may lack the vast resources of global enterprises to absorb unforeseen cloud costs or dedicate teams to constant optimization. The data from Node4’s research provides a clear and compelling narrative: the vast majority of these firms are actively re-evaluating and reversing aspects of their public cloud deployments. While the research addresses the “public cloud” market broadly, the dominance of Microsoft Azure and AWS, which collectively hold between 70% and 80% of the UK market share, means these findings are directly applicable to organizations heavily invested in the Azure ecosystem. The trend is not one of complete abandonment but of strategic refinement, with nearly half of the firms planning to move only a few specific applications, signalling a move towards optimized hybrid environments.

The primary impetus for this shift among mid-market firms is performance, particularly for those man‐ aging large estates of Windows virtual machines. Many workloads, especially those that were hastily moved to the cloud using a “lift-and-shift” methodology without being re-architected, have failed to perform as expected. User frustrations with latency in Software-as-a-Service applications and the unsuitability of the public cloud for certain high-performance, low-latency computing tasks are forcing a reconsideration. For a company running a large estate of approximately 1000 Windows virtual ma‐ chines, many of which may support stable, predictable, and performance-sensitive business applications, the logic of repatriation becomes clear. Constant, 24/7 workloads can often be run more cost-effectively and with better performance on dedicated on-premises hardware, where the variable costs of the public cloud are replaced by the predictable costs of owned infrastructure. Technical analysis shows that performance issues in Azure VMs can stem from numerous factors, including incorrect VM sizing, disk I/O bottlenecks on standard storage, and network latency, all of which can be mitigated with more direct control in an on-premises setting.

Data sovereignty and compliance represent another critical driver, with 30% of mid-market firms citing it as a reason for repatriation. For professional services and legal firms, which handle vast amounts of sensitive client data, the physical location of that data and the legal jurisdiction it falls under are not trivial matters. The complexities of GDPR and other regulations, combined with a growing desire for digital sovereignty, are making the perceived simplicity of public cloud data storage less attractive. The fact that major providers like Microsoft and Amazon have been compelled to announce sovereign cloud services for European customers is a direct acknowledgment of this widespread concern. Furthermore, the research suggests that organizations running primarily on-premises infrastructure feel more confident in their ability to prevent and respond to cyberattacks. This perception stems from a greater sense of control and visibility compared to the shared responsibility model of the public cloud, where misconfigurations and skill gaps can create significant security vulnerabilities. The combination of performance degradation, escalating and unpredictable costs, and heightened concerns around security and compliance is creating a powerful business case for UK mid-market firms to strategically repatriate key workloads from public cloud platforms like Azure.

A Contrasting Trajectory: The UK Legal Sector’s Cloud Adoption

While the broader UK mid-market is actively engaged in cloud repatriation, the UK legal and professional services sector presents a contrasting and compelling narrative of accelerated cloud adoption. The predominant trend for law firms in 2023-2024 is a robust digital transformation characterized by a strategic migration of core systems and infrastructure to cloud platforms, with Microsoft Azure being a particularly favoured destination. This movement is not a tentative step but a decisive shift, as evidenced by a benchmarking study from Legal IT Insider and Philips, which found that 68% of surveyed law firms, primarily in the UK and North America, now have more than half of their core systems based in the cloud. In stark contrast, only 9% remain predominantly on-premises. The strategic intent is clear, with 12% of firms reporting they are already entirely cloud-based and a further 23% planning to achieve this within the next 18 months.

The drivers for this enthusiastic adoption are rooted in the unique demands of the legal profession. Migrating to Azure allows firms to enhance operational efficiency, offload the burdensome management of on-site technology, and gain access to advanced tools. A key benefit of a “cloud-first strategy” is that the provider manages the support, maintenance, resilience, and security of the underlying infra‐ structure. This is particularly crucial for mitigating risks associated with legacy applications, which are a major vector for malware. Security and compliance, far from being a deterrent, are often a catalyst for moving to the cloud. Law firms are increasingly confident in the ability of major cloud providers to protect their data from sophisticated cybersecurity threats, a confidence that extends to their clients. Furthermore, the shift to hybrid and remote working models has made cloud solutions, especially SaaS products and platforms like Azure Virtual Desktop, indispensable. These technologies provide the “anytime, anywhere” access that modern legal practice demands, enabling seamless collaboration and greater agility.

Case studies from the sector reinforce this trend. Fisher Jones Greenwood Solicitors, a prominent UK law firm, migrated its entire aging in-house infrastructure of over 20 servers to a cloud solution to enhance resilience, agility, and flexibility. Similarly, numerous UK public sector legal teams have migrated to cloud platforms to modernise legacy systems and strengthen compliance. However, it is crucial to note a critical nuance that signals future challenges. A case study of a Top 20 UK law firm that migrated its core financial system to Azure via a “lift-and-shift” approach to meet an urgent deadline subsequently faced “excessive” costs due to an unoptimised architecture. This situation did not lead to repatriation but to a project focused on optimization within the Azure environment. This highlights that while the legal sector is still in a strong adoption phase, it is beginning to encounter the same challenges of cost and performance that have driven more mature cloud users towards repatriation. Moreover, recent disclosures that Microsoft cannot guarantee absolute data sovereignty for all data processed in its cloud, as some data may be transferred internationally for processing, create a significant latent compliance risk for law firms bound by strict client confidentiality and UK data protection laws. The current focus may be on optimization, but the underlying pressures are identical to those driving repatriation elsewhere.

Strategic and Technical Realities of Azure Repatriation

For any organization, particularly one with a substantial IT estate of around 1000 Windows virtual ma‐ chines, the decision to repatriate workloads from Microsoft Azure is a significant strategic and technic‐ al undertaking. It is far more complex than simply reversing the initial migration process. The journey requires meticulous planning, a deep understanding of the technical hurdles, and a clear-eyed assessment of the associated costs, which can be substantial. The process begins with a comprehensive audit of the existing cloud infrastructure to identify which workloads are suitable candidates for repatriation. This involves analysing resource consumption, performance metrics, application dependencies, and the total cost of operation in the cloud versus a projected on-premises model. Workloads that are stable, predictable, and performance-sensitive, or those that handle highly sensitive data subject to strict compliance mandates, are often the prime candidates.

The technical execution of a reverse migration from Azure to an on-premises environment is fraught with challenges. Unlike the well-defined forward migration path supported by dedicated tools like Azure Migrate, the process of moving back is less straightforward. While tools like Azure Site Recovery (ASR), primarily designed for disaster recovery, can be repurposed to replicate Azure VMs back to an on-premises VMware or Hyper-V environment, it is a complex procedure requiring specialist knowledge. A critical and often prohibitive barrier is the cost of data egress. Public cloud providers, including Microsoft, charge significant fees for transferring data out of their networks. For an organization moving terabytes of data associated with 1000 VMs, these egress fees alone can run into tens or even hundreds of thousands of pounds, creating a powerful financial disincentive to leave and reinforcing vendor lock-in. This very issue is a key focus of the UK Competition and Markets Authority’s (CMA) investigation into the cloud market, highlighting its impact on competition and customer choice.

Beyond egress fees, organizations must contend with the practical challenges of data transfer, which requires significant network bandwidth and can lead to extended periods of downtime if not managed carefully. Application reintegration is another major hurdle; applications may need to be redeveloped or adapted to function correctly on on-premises architectures, especially if they were modified to use cloud-native services that are not available outside the Azure ecosystem. This entire process demands a high level of technical expertise in both cloud and on-premises systems, often necessitating the engagement of specialist professional services firms or consultants to manage the migration and mitigate risks. The significant upfront capital expenditure required to build or scale up a high-performance on-premises infrastructure, including servers, storage, and advanced cooling solutions, must also be factored into the business case. These formidable technical and financial obstacles underscore why re‐ patriation is a strategic decision that cannot be taken lightly and why many firms, like the Top 20 law firm mentioned previously, may first opt for intensive optimization within the cloud before considering a full-scale reverse migration.

Conclusion and Strategic Recommendations

The landscape of enterprise IT in the United Kingdom is at a strategic inflection point. The era of un‐ questioning adherence to a “cloud-first” mandate is giving way to a more mature, pragmatic, and analytical “cloud-appropriate” strategy. This research confirms that cloud repatriation is no longer a niche concept but a mainstream trend, particularly within the UK’s mid-market sector. For companies with approximately 2000 employees and a significant virtual machine footprint, the drivers for moving select workloads out of public cloud environments like Microsoft Azure are compelling and clear, focusing on tangible concerns over performance, escalating costs, data sovereignty, and security. The evidence strongly suggests that for many stable and predictable workloads, the economic and operational benefits of on-premises or private cloud infrastructure are being rediscovered.

In stark contrast, the UK legal and professional services sector is currently on a different trajectory, remaining in a phase of accelerated cloud adoption. This is driven by an urgent need to modernize legacy systems, enhance security, and enable the flexibility required for modern hybrid work. However, this report concludes that this is likely a matter of timing and maturity rather than a fundamental di‐ vergence in strategy. The challenges that legal firms are beginning to encounter—such as the “bill shock” from unoptimized migrations and the ever-present burden of regulatory compliance, amplified by recent revelations about the limits of data sovereignty guarantees—are the very same catalysts that have spurred repatriation in other industries. The drivers for repatriation are deeply aligned with the core values of the legal profession: cost control, risk management, and the uncompromising protection of sensitive client data

Therefore, the key recommendation for IT decision-makers and business leaders in law firms and professional services organizations is to pursue a proactive and strategic approach to hybrid cloud management. Rather than viewing cloud as an all-or-nothing proposition, leaders should foster a culture of continuous assessment. This involves rigorously evaluating each workload based on its specific performance, security, compliance, and cost profile to determine its optimal placement. Organizations should develop a clear exit strategy and understand the potential costs of repatriation, including egress fees, even as they migrate new workloads to the cloud. By embracing a hybrid model and avoiding vendor lock-in, firms can harness the scalability and innovation of public cloud for the work‐ loads that benefit most, while retaining the control, performance, and cost-effectiveness of on-premises infrastructure for their most critical and sensitive operations. This balanced, strategic approach will be the hallmark of a resilient and successful IT strategy in the years to come.