AWS Migration Hub vs. Azure Migrate vs. Google Cloud: Which Tool to Use in 2026
AWS Migration Hub closed to new customers in 2025. Compare AWS Transform, Azure Migrate and Google Migration Center on cost, VMware and fit.

Last updated November 2026
Use AWS Transform if you are moving to AWS, Azure Migrate if you run Windows Server and SQL Server with Software Assurance, and Google Cloud Migration Center if you need the widest cross-cloud intake with no tool fee. AWS Migration Hub stopped accepting new customers on November 7, 2025, so it is no longer a starting point for a new program.
The native tool is rarely the expensive part of a migration. Assessment and replication tooling is free or close to free on all three clouds. What costs money is how long each server replicates, which licenses you can bring, and how well the assessment sizes what you move.
That matters because overspend is already the norm. Organizations report cloud budgets running 17% over and estimate 29% of cloud spend is wasted. A migration plan built on a bad assessment carries that waste straight into the new environment.
Every price, status and date below was checked against vendor documentation in November 2026.
AWS Migration Hub is closed to new customers: what replaced it
AWS Migration Hub stopped accepting new customers on November 7, 2025 (AWS). Existing customers can finish their projects and keep their data, and AWS still ships security updates, but the service gets no new features. AWS points every new project to AWS Transform, which became generally available in May 2025.
The closure covers the whole Migration Hub family, along with Application Discovery Service and the older Mainframe Modernization Service. Application Migration Service, the replication engine most teams actually used, survived. AWS renamed it AWS Transform MGN on June 8, 2026.
If you are mid-migration on Migration Hub, you do not need to move anything. AWS says no data migration or formal transition is required, and existing projects keep working until they finish. For your next wave, plan in AWS Transform so you are not building on a frozen tool.
Know Your Migration Context Before You Compare Tools
Before comparing features, understand your starting point. The best migration tool depends entirely on your specific situation.
Your Current Infrastructure
What are you migrating from? VMware environments need different tools than Hyper-V deployments. Physical servers require different approaches than existing cloud workloads. Legacy applications with complex dependencies demand more sophisticated discovery than modern microservices.
Take inventory of your application portfolio. How many workloads are you moving? What databases are involved? Which applications have tight coupling that makes migration risky?
Your Team's Capabilities
Be honest about your team's expertise. Do you have certified cloud architects on staff? Can your team handle complex migrations independently? Or do you need guided processes with built-in best practices?
Your team's bandwidth matters too. Are they already stretched thin with operational demands? Migration projects require sustained attention. Understaffed teams need tools with stronger automation and less manual configuration.
Your Organizational Constraints
Budget reality shapes every decision. What's your migration budget? Can you absorb unexpected costs? Migrations routinely run over plan. In McKinsey's 2021 survey of about 450 CIOs, migrations cost 14% more than planned each year and 38% ran more than a quarter late (McKinsey).
Timeline pressures create additional constraints. Do you have business-driven deadlines? Compliance requirements? Executive expectations that may not align with technical reality?
Risk tolerance varies by industry and organization. Can you afford downtime during cutover? What's your rollback strategy if something breaks? How much business disruption can you absorb?
Your Strategic Intent
Are you doing lift-and-shift or pursuing modernization? Simple rehosting moves workloads with minimal changes. Replatforming makes modest optimizations. Refactoring rebuilds applications for cloud-native architectures.
Your cloud strategy matters. Are you committed to a single cloud provider? Or do you need multi-cloud flexibility? Single-cloud commitment enables deeper ecosystem integration. Multi-cloud strategies require more portable approaches.
Think about your three-year vision. Are you planning to containerize applications? Adopt serverless architectures? Integrate AI and machine learning? Your migration tool should support your long-term direction, not just immediate needs.
Your Microsoft Licensing
Check your Windows Server and SQL Server licenses before you pick a cloud. Licenses with active Software Assurance can move to Azure through Azure Hybrid Benefit. Since October 1, 2019, Microsoft has barred licenses without Software Assurance or mobility rights from running on the dedicated hosts of its Listed Providers: Amazon, Google, Alibaba and Microsoft itself (Microsoft licensing FAQ).
For a Microsoft-heavy estate, that rule changes the cost of AWS and Google before you compare a single tool. Get your licensing team or reseller to confirm what you can bring before the business case goes to finance.
AWS Transform and AWS Transform MGN: AWS's Migration Tools in 2026
AWS Transform is AWS's migration and modernization service, and it is where new AWS migrations start. It runs discovery, assessment, wave planning, landing zone setup and rehosting as one agent-driven workflow. The replication itself runs on AWS Transform MGN, the renamed Application Migration Service. Both are built for organizations moving to AWS and nowhere else.
Best Features and Benefits
One Workflow From Discovery to Cutover
AWS Transform pulls together jobs that used to need four tools: Application Discovery Service, Migration Hub planning, Strategy Recommendations and MGN. An agent maps dependencies, groups servers into waves, builds the landing zone and network, then hands each wave to MGN for replication. Your team approves each step instead of stitching outputs together.
Assessment That Starts From What You Already Have
You do not need a fresh discovery run to get a business case. Since May 2026, AWS Transform assessments accept RVTools exports, CMDB data and third-party discovery files. They then model what-if scenarios for region, utilization and service mapping. That speeds up the first estimate, especially for VMware estates that already have RVTools output.
More Sources Than VMware
AWS Transform launched with VMware as its main use case. In July 2026, AWS extended the migration workflow to Hyper-V and other virtual and bare-metal sources, and renamed it Migrations (including VMware). MGN also lists physical servers, Hyper-V, vSphere and other cloud providers as sources.
Free Agents for the Core Work
The assessment, migration, mainframe and Windows agents carry no charge (AWS Transform pricing). Custom code transformations cost $0.035 per agent minute. AWS notes that future capabilities may be paid, so confirm the price of anything new before you plan around it.
Replication With a 90-Day Free Window
MGN gives each source server 2,160 free replication hours, which is 90 days of continuous replication starting when you install the agent. After that, AWS bills hourly for any server still replicating. Replication servers, EBS volumes and test or cutover instances are billed from day one.
Mainframe and Windows Modernization
AWS Transform for mainframe has been generally available since May 2025 for IBM z/OS and Fujitsu GS21 code. At re:Invent in December 2025, AWS launched a full-stack Windows agent that moves .NET applications and SQL Server databases to Aurora PostgreSQL on Linux. AWS claims Windows modernization runs up to 5x faster and removes up to 70% of maintenance and licensing costs; those are AWS's figures, not independent results.
Compliance Coverage for Regulated Teams
MGN kept its FedRAMP High, HIPAA and PCI DSS scope through the rename and runs in all commercial regions plus both AWS GovCloud (US) regions. AWS Transform added FedRAMP coverage in US East (N. Virginia) and GovCloud (US-West) in 2026.
Where It Might Lack
AWS-Only Target
AWS Transform and MGN move workloads into AWS. If your strategy keeps a second cloud in play, you will run a different tool for that side.
The 90-Day Clock
MGN's free window is half the length of Azure's. Waves that stall in testing or change control start costing money per server per hour. AWS does not list the hourly rate on its MGN pricing page, so price it in the AWS Pricing Calculator before you commit to a schedule.
Agents Still Changing Fast
AWS shipped more than a dozen changes to the migration workflow in 2026 alone. That is good for capability and hard on runbooks: screenshots, training material and partner playbooks age quickly. Some newer pieces, such as the full-stack Windows agent, carry no clear GA label in AWS's change log.
Cross-Cloud Sources Are Vague
MGN supports other cloud providers, but AWS does not name Azure or Google Cloud on its FAQ. If you are moving from another cloud, migrate one representative server end to end before you plan the rest.
How It Compares to the Others
vs. Azure Migrate
AWS's migration agents are generally available and free today. Azure's migration agent is still in preview and cannot run replication. Azure gives each server a 180-day free window against AWS's 90 days, and Azure Hybrid Benefit gives Microsoft-licensed estates a cost edge AWS cannot match.
vs. Google Cloud
AWS automates more of the path from discovery to cutover. Google charges no tool fee for VM migration at all and accepts more cloud sources by name. Google also publishes how its migration funding is calculated; AWS does not.
Who Is Best Suited For
AWS Transform works best for:
- Organizations that have chosen AWS as the target cloud
- VMware, Hyper-V or bare-metal estates that want one workflow from assessment to cutover
- Teams that can keep each server's replication under 90 days
- Mainframe programs moving z/OS or Fujitsu GS21 code
- Regulated teams that need FedRAMP High or GovCloud coverage
If AWS is already your direction, start in AWS Transform even for small migrations. The core agents are free, and you avoid building on Migration Hub, which no longer accepts new customers.
Azure Migrate: Best for Windows Server and SQL Server Estates
Azure Migrate provides an integrated platform for discovering, assessing, and migrating workloads to Microsoft Azure. It's built for organizations with significant Microsoft investments who want a unified migration experience.
Best Features and Benefits
End-to-End Unified Platform
Azure Migrate covers the entire migration lifecycle from discovery to modernization. You don't need to stitch together multiple tools. Everything works within a single, integrated environment. This reduces complexity and learning curve for your team.
Flexible Discovery Options
The platform supports both agent-based and agentless discovery. This flexibility matters when you have diverse environments. You can assess large VMware deployments without installing agents on every server. For physical servers or complex environments, agents provide deeper visibility. Use agentless dependency analysis where you can. Agent-based dependency visualization has billed standard Log Analytics charges since July 1, 2024.
Intelligent Sizing Recommendations
Azure Migrate analyzes historical utilization data to recommend appropriate cloud instance types. This prevents over-provisioning that wastes money and under-provisioning that creates performance issues. The recommendations account for actual usage patterns, not just peak capacity.
Deep Azure Integration
Seamless connection with Azure security, governance, and monitoring services simplifies post-migration operations. Role-based access control, Azure Policy, Azure Monitor, and Microsoft Defender for Cloud work together out of the box. This integration reduces the operational overhead of managing migrated workloads.
Azure Hybrid Benefit
If you have Windows Server or SQL Server licenses with active Software Assurance, you can apply them to Azure instead of paying for the license again. Microsoft states SQL Server customers can save up to 85% against pay-as-you-go rates, and Windows Server customers up to 80% when they combine the benefit with reserved instances (Azure Hybrid Benefit). Those are Microsoft's best-case figures. Your saving depends on region, instance type and how many licenses you actually hold.
Cost-Effective Migration
Discovery, assessment and web app migration are free. Server migration is free for the first 180 days per machine, then costs $25 a month for each server still replicating (Azure Migrate pricing). Storage, transactions and data transfer during replication are billed throughout. Azure Database Migration Service is free on the Standard tier for offline migrations, and the Premium tier is free for the first six months.
AI Agents, Mostly in Preview
Microsoft added agent features to Azure Migrate through 2026: automatic application discovery, wave planning, VMware network assessment and reports with TCO and ROI. The Azure Copilot migration agent entered public preview on March 11, 2026. Its own documentation says it runs end to end only for VMware and cannot perform replication. On the code side, the GitHub Copilot modernization agent became generally available on June 2, 2026, for upgrading Java and .NET applications.
Where It Might Lack
Microsoft-Centric Approach
Azure Migrate is optimized for Microsoft technologies. If your environment is heavily Linux-based or uses non-Microsoft databases, you may not get the same level of optimization. The platform works best when migrating Windows Server, SQL Server, and Microsoft-stack applications.
Limited Multi-Cloud Capabilities
While you can use Azure Migrate for some multi-cloud scenarios, it's clearly designed for Azure-centric strategies. If you need to maintain workload portability across multiple clouds, this creates constraints.
Learning Curve for Non-Microsoft Teams
Teams without strong Azure or Microsoft expertise face a steeper learning curve. The platform assumes familiarity with Azure concepts, services, and management approaches. Plan for training if your team comes from different technology backgrounds.
Preview Agents Are Not Production Tooling Yet
Most of Azure Migrate's AI features are in preview, and the migration agent cannot run replication. Plan your waves on the generally available tooling and treat the agents as helpers for planning and reporting.
Other Clouds Migrate as Physical Servers
Azure Migrate treats AWS EC2 and Google Cloud VMs as physical servers. You need a replication appliance and a Mobility agent on each VM. Amazon Linux cannot be migrated as it is, and paravirtualized AWS VMs may not boot after migration.
How It Compares to the Others
vs. AWS Transform
Azure gives each server 180 free days of replication against AWS's 90, which suits slow, change-controlled waves. AWS's migration agents are generally available and free today, while Azure's are mostly in preview. For Microsoft-licensed estates, Azure Hybrid Benefit usually decides the comparison before the tooling does.
vs. Google Cloud
Azure Migrate gives you one integrated hub for discovery, assessment and migration. Google spreads the same jobs across Migration Center, Migrate to Virtual Machines and Database Migration Service, and charges no tool fee for VM migration. Choose Azure for Microsoft-stack estates; choose Google when you need to pull from several clouds at once.
Who Is Best Suited For
Azure Migrate is the right choice for:
- Organizations with significant Microsoft licensing investments
- Windows Server and SQL Server-heavy environments
- IT leaders leveraging Enterprise Agreements and Azure Hybrid Benefit
- Teams with strong Microsoft and Azure expertise
- Companies prioritizing a unified platform experience over specialized tools
- Teams whose waves need more than 90 days of replication per server
If you're already in the Microsoft ecosystem, Azure Migrate is the path of least resistance. The licensing benefits alone often justify the choice. Your team's existing Microsoft knowledge transfers directly.
Google Cloud Migration Center and Google Cloud Modernize
Google Cloud does not offer a single migration tool. Migration Center handles discovery, assessment and planning, Migrate to Virtual Machines moves servers, and Database Migration Service moves databases. On October 5, 2026, Google grouped these, plus VMware Engine and its mainframe tools, under a new portfolio name: Google Cloud Modernize (Google Cloud). The products underneath kept their names and pricing.
Best Features and Benefits
Purpose-Built Specialized Tools
Google provides dedicated tools for each workload type. Migrate to Virtual Machines (formerly Migrate for Compute Engine) moves VMs. Database Migration Service moves MySQL, PostgreSQL, SQL Server and Oracle databases. Migrate to Containers (formerly Migrate for Anthos) still exists but has run only as a command-line tool since May 2024. This specialization lets each tool fit its job closely.
Migration Center for Unified Visibility
Migration Center is free and sits on top of the specialized tools. It discovers on-premises servers, Azure VMs and Amazon RDS databases, imports RVTools exports, and produces TCO and dependency reports. Its Agentic Quick Estimator, generally available since October 2026, turns an RVTools export into a Compute Engine cost estimate and lets you test assumptions such as bring-your-own-license against pay-as-you-go.
No Tool Fee for VM Migration
Migrate to Virtual Machines charges nothing for migrations into Google Cloud (Google pricing). You pay for test-clone and cutover VMs, storage and network, and AWS or Azure bill you for source-side resources if you are moving from those clouds. Homogeneous MySQL and PostgreSQL database migrations are free. SQL Server or Oracle to PostgreSQL migrations are billed per GiB of data.
Published Migration Funding
New Google Cloud customers get $300 in credits, enough for a proof of concept. The larger lever is the Rapid Migration and Modernization Program (RaMP). Its terms set partner or professional-services funding at a baseline of 20% of projected annual recurring revenue, capped at $2 million per workload. Neither AWS nor Microsoft publishes a formula for its migration funding.
Modernization Capabilities
Google's modernization tooling now centers on Gemini. Database Migration Service has used Gemini for schema and code conversion since September 2025, and SQL Server to PostgreSQL migrations have been generally available since June 2025. The new Modernization Hub and App Modernization CLI analyze Java, .NET and mainframe code. An EKS-to-GKE migration agent for moving Kubernetes workloads from AWS entered public preview in 2026; it translates manifests through pull requests but does not move data.
Specialized Legacy Support
Google offers a Mainframe Assessment Tool, generally available since April 2025, plus Dual Run for running mainframe and cloud versions side by side during cutover. AWS also has a generally available mainframe agent, so mainframe support alone no longer separates the two. Google's SAP support remains strong, with X5 instances offering up to 43 TiB of memory on a single node.
Where It Might Lack
Tool Integration Complexity
Multiple specialized tools mean you need to select, configure, and integrate them appropriately. This adds complexity compared to single, unified platforms. Your team needs to understand which tool applies to which scenario.
More Technical Expertise Required
The specialized tool approach assumes higher technical sophistication. Your team needs to understand the nuances of different migration patterns and select appropriate tools. Less technical teams may struggle without guidance.
Agentic Execution Is Still Ahead of the Docs
Google's product pages promise migration agents that execute migration activities at scale, but its documentation does not yet describe such a feature. The EKS-to-GKE agent is in preview, and several modernization tools carry no GA label. Plan on the generally available tools.
A 100-Day Limit on Migrating VMs
Migrate to Virtual Machines expires a migrating VM after 100 days, extendable once by another 100. That does not cost money, but long-running waves need someone tracking the clock.
How It Compares to the Others
vs. AWS Transform
Google charges no tool fee for VM migration and accepts AWS and Azure sources by name. AWS automates more of the journey with generally available agents. If your estate sits across several clouds, Google's intake is wider; if you want one guided workflow into a single cloud, AWS goes further.
vs. Azure Migrate
Google's portfolio offers more specialized tools and published funding terms. Azure offers a single hub and the strongest licensing position for Microsoft workloads. Google suits diverse, multi-cloud estates; Azure suits Microsoft-stack migrations.
Who Is Best Suited For
Google Cloud Migration Center works best for:
- Organizations with diverse workloads needing specialized approaches
- IT leaders who want no tool fee for VM migration and published funding terms
- Teams pursuing aggressive modernization alongside migration
- Companies wanting to maintain multi-cloud optionality
- Technical teams comfortable integrating multiple specialized tools
If you have mainframe systems, SAP environments, or want to containerize during migration, Google's specialized tools and SQL Server to PostgreSQL path are worth evaluating. Migration Center's free TCO estimates give budget owners an early number to test.
How Much Do AWS, Azure and Google Cloud Migration Tools Cost?
The migration tools themselves cost little on any of the three clouds. AWS Transform's core agents are free, Azure Migrate charges nothing for 180 days per server, and Google charges no tool fee for VM migration. The money goes to replication infrastructure, test and cutover compute, data transfer, and any server that replicates past its free window.
The free window is the number to plan around. AWS gives each server 90 days of replication, Azure gives 180, and Google charges no fee but expires a migrating VM after 100 days unless you extend it. A wave that sits in testing for four months costs nothing extra on Azure or Google and starts billing per server per hour on AWS.
Two more costs sit outside the table. Licensing can outweigh every tool fee for a Windows Server and SQL Server estate, as covered in the licensing section above. Funding can offset part of the bill: AWS's Migration Acceleration Program and Microsoft's Azure Accelerate both fund migrations case by case, while Google's RaMP publishes a baseline of 20% of projected annual recurring revenue, capped at $2 million per workload. Ask for funding before replication starts, because it is negotiated per program and rarely applied after the fact.
Migrating VMware to AWS, Azure or Google Cloud
Every hyperscaler offers two VMware exits: keep running VMware on managed hosts, or move the VMs onto native cloud instances. The first path no longer escapes Broadcom. Amazon EVS, Azure VMware Solution and Google Cloud VMware Engine all require you to bring your own VMware Cloud Foundation subscription, and Microsoft and Google both stopped selling license-included capacity on October 15, 2025.
The pressure to move is real but slow. In a February 2026 survey of 302 North American IT decision-makers, 86% were actively shrinking their VMware footprint, but only 4% had fully moved off (CIO Dive, reporting a CloudBolt survey). CloudBolt sells cloud management software, so read the figures as directional. Most teams are running both paths at once, keeping some VMware while they move the rest.
I would size both paths before committing to either. Run your RVTools export through AWS Transform's assessment and Google's Quick Estimator; both accept it and both are free. Then price the managed-VMware option with your Broadcom quote attached. For on-premises alternatives such as Proxmox, Nutanix AHV and Hyper-V, see our guide to what happens when you migrate off VMware.
Cloud-to-Cloud Migration: Moving Between AWS, Azure and Google Cloud
Each provider's native tools migrate into its own cloud and nowhere else. To leave AWS for Azure, you use Azure Migrate; to leave Azure for Google, you use Migrate to Virtual Machines. No hyperscaler ships a native tool for moving workloads out.
The depth of support differs. Google names AWS and Azure as sources for VM migration and discovers Azure VMs and Amazon RDS in Migration Center. Azure migrates AWS and Google VMs as physical servers, with an agent on each one. AWS says MGN supports other cloud providers without naming them.
Pick where you are migrating from and where you are going. The path finder shows the native tool for each step, its status as of November 2026, the free window, and the catch to plan around.
Worked example: moving Windows VMs from AWS EC2 to Azure. Assess them in Azure Migrate, which treats EC2 instances as physical servers. Deploy a replication appliance and a Mobility agent on each VM, and check for Amazon Linux or paravirtualized instances, which will not migrate cleanly. Replication is free for 180 days per server, then $25 a month.
AWS vs Azure vs Google Cloud Migration Tools: Side-by-Side Comparison
AWS leads on generally available migration agents, Azure on free replication time and Microsoft licensing, and Google on cross-cloud intake and published funding terms. All three need your own Broadcom license to keep running VMware. Use the tabs to compare tooling and status, cost and funding, and sources and licensing.
Which Cloud Migration Tool Fits Your Environment? Take the 5-Question Test
Answer five questions about your target cloud, Microsoft licensing, VMware plans, replication timeline and special workloads. The questionnaire scores AWS, Azure and Google Cloud and returns the native toolchain that fits best, why, what to watch out for, and the runner-up.
Worked example: a company with no target cloud decided yet, heavy Windows Server and SQL Server licensing with Software Assurance, VMware it plans to leave, waves expected to take 90 to 180 days per server and no special workloads scores Azure 9, AWS 3 and Google Cloud 2. Azure Migrate is the best match, with AWS Transform as the runner-up.
Ask Yourself These Questions:
About Your Environment:
- What percentage of your workloads run on Microsoft technologies?
- Do you have mainframe or legacy systems requiring specialized migration approaches?
- How complex are your application dependencies?
- What's your current cloud footprint, if any?
- Which Windows Server and SQL Server licenses carry active Software Assurance?
About Your Organization:
- What's your realistic migration budget including contingency?
- What timeline are you actually working with, not just the executive expectation?
- How much business disruption can you absorb during migration?
- What's your tolerance for vendor lock-in?
About Your Team:
- What cloud platforms does your team have experience with?
- Can you dedicate resources full-time to migration, or is this on top of existing responsibilities?
- Do you need extensive vendor support, or can you largely self-execute?
- What's your team's capacity to learn new tools and platforms?
About Your Strategy:
- Are you committed to a single cloud provider, or keeping options open?
- Is this pure lift-and-shift, or are you modernizing during migration?
- What's your three-year application architecture vision?
- How important is maintaining workload portability?
These questions help you move from abstract comparison to concrete decision-making based on your reality.
Real-World Scenarios: Which Cloud Migration Tool Wins?
Let's examine specific scenarios that IT leaders commonly face.
Scenario 1: Mid-Sized Financial Services Company
The Context: 300 virtual machines, predominantly Windows Server and SQL Server. Strict compliance requirements including SOX and PCI-DSS. Limited IT team of 12 people with strong Microsoft background. Budget constraints require clear ROI justification.
The Decision: Azure Migrate
Why It Makes Sense: The licensing leverage is immediate. Azure Hybrid Benefit lets the firm reuse its Windows Server and SQL Server licenses with Software Assurance, which Microsoft says can cut SQL Server costs by up to 85% against pay-as-you-go. The 180-day free replication window also suits change-controlled waves. The team's Microsoft expertise transfers directly, minimizing training needs. Deep integration with Azure security and compliance tools addresses regulatory requirements. The unified platform reduces complexity for a small team managing everything from discovery to optimization.
The Implementation: Start with agentless discovery to map the environment quickly. Use Azure's assessment tools to build the business case with accurate cost projections. Migrate in waves, starting with development environments to build team confidence. Leverage Azure Migrate's integration with Microsoft Defender for Cloud and Azure Policy to maintain compliance throughout migration.
Scenario 2: Large Retail Enterprise
The Context: 1,500 servers supporting e-commerce, point-of-sale, inventory management, and supply chain systems. Multi-year, three-phase migration program. Multiple teams across IT, operations, and business units. Mix of Windows and Linux workloads. Existing small AWS footprint.
The Decision: AWS Transform with AWS Transform MGN
Why It Makes Sense: The scale demands one coordinated workflow, and the existing AWS footprint gives the program a landing zone to build on. AWS Transform's migration agents are generally available and free, and they handle discovery, wave planning, landing zone and network setup before handing each wave to MGN. Since September 2026, AWS Transform can extend an existing Control Tower landing zone instead of building a new one.
The Implementation: Start with an AWS Transform assessment from existing RVTools or CMDB data rather than a fresh discovery deployment. Group waves by business function, and size each wave so its servers finish replicating inside MGN's 90-day free window. Negotiate Migration Acceleration Program funding with AWS before replication starts. Do not plan around AWS Migration Hub or Application Discovery Service; both stopped accepting new customers in November 2025.
Scenario 3: Tech-Forward Manufacturing Company
The Context: Legacy mainframe systems running critical production scheduling. Modern containerized applications for customer portal and analytics. 400 servers total. Priority on modernization, not just migration. Technical team comfortable with multiple cloud platforms. Strong focus on cost control.
The Decision: Google Cloud Migration Center and Migrate to Virtual Machines
Why It Makes Sense: Google's Mainframe Assessment Tool and Dual Run address the legacy assessment and cutover risk. Its App Modernization CLI and Gemini-based tooling analyze code for containerization. AWS Transform for mainframe is a credible alternative here, so assess both before committing. The technical team can handle integrating multiple specialized tools. Migration Center's free TCO estimates support the budget focus. Multi-cloud approach maintains flexibility as the company evolves.
The Implementation: Use Migration Center to assess the estate and Migrate to Virtual Machines for standard VM migrations. Run the Mainframe Assessment Tool before any code conversion, and use Dual Run to compare outputs before cutover. Apply for RaMP funding before the program starts. Gartner predicts more than 70% of mainframe exit projects started in 2026 will miss their intended benefits because teams overestimate generative AI tooling (Gartner), so budget for manual review of converted code.
Scenario 4: Healthcare Organization
The Context: 600 servers supporting electronic health records, patient portal, and clinical systems. Hybrid approach maintaining some on-premises for compliance. Gradual migration over 24 months. Small IT team with limited cloud experience. Strict uptime requirements.
The Decision: Azure Migrate (with careful planning)
Why It Makes Sense: The Windows-heavy healthcare environment aligns with Azure's strengths. Hybrid capabilities support the partial migration strategy. Azure's healthcare-specific compliance certifications reduce regulatory risk. The unified platform reduces complexity for a small team. Integration with Microsoft security tools addresses healthcare's stringent security requirements.
The Implementation: Start with comprehensive assessment to identify dependencies. Use Azure's hybrid capabilities to maintain some workloads on-premises while migrating others. Migrate in very small waves to minimize risk. Leverage Azure Site Recovery for business continuity during migration. Work closely with Azure support given limited internal expertise. Keep Azure Migrate's preview agents out of decisions that touch production patient data until they reach general availability.
Which Cloud Migration Tool Should You Choose in 2026?
Your target cloud decides the tool. Once you know where you are going, the native tool for that cloud is the default, and the real decisions are licensing, wave length and how much of the work you trust to agents.
- Choose AWS Transform if AWS is your target. Its migration agents are generally available and free, and MGN replicates each server free for 90 days.
- Choose Azure Migrate if you run Windows Server and SQL Server with Software Assurance, or if your waves need more than 90 days per server.
- Choose Google Cloud Migration Center if your workloads sit across several clouds, or if you want published funding terms and no tool fee for VM migration.
- Stop planning around AWS Migration Hub. It no longer accepts new customers, and AWS directs new work to AWS Transform.
No tool eliminates migration risk. They manage it differently. Your people and processes matter more than the tool. The best technology can't compensate for poor planning or inadequate change management.
Free assessment phases often lead to expensive execution lock-in. Understand the total cost of ownership, not just the initial tooling costs. Factor in training, consulting, extended timelines, and post-migration optimization.
Migration timelines always take longer than projected. Build in contingency. Unexpected dependencies emerge. Integrations prove more complex than anticipated. Business priorities shift mid-migration. Plan for these realities.
Post-migration optimization requires ongoing investment. Moving to the cloud is just the beginning. Right-sizing, cost optimization, and performance tuning require sustained attention. Budget for this from day one.
If you need a migration partner or a third-party platform rather than native tooling, see our guide to the best cloud migration solutions for IT leaders in 2026. Also read: AWS Bedrock vs Azure OpenAI vs Google Vertex AI: Enterprise AI Comparison Guide
Looking for cloud migration vendors?
Help us understand your environment and priorities. We'll match you with the vendors worth talking to. Plus a $50 gift card for your time.
FAQ
What replaced AWS Migration Hub?
AWS Transform replaced it. AWS Migration Hub stopped accepting new customers on November 7, 2025, along with Application Discovery Service, Strategy Recommendations, Orchestrator, Journeys and Refactor Spaces. AWS says their planning, recommendation and orchestration features now live in AWS Transform. Existing Migration Hub customers can finish their projects but get no new features.
Is AWS Transform free?
AWS Transform's assessment, migration, mainframe and Windows agents are free. Custom code transformations cost $0.035 per agent minute. You still pay for the AWS resources you create, and AWS Transform MGN bills hourly for any server that replicates longer than its 2,160 free hours, which is about 90 days.
What is the best cloud migration tool for enterprises in 2026?
The best tool is the native tool for the cloud you are moving to. Use AWS Transform for AWS, Azure Migrate for Azure and Google Cloud Migration Center with Migrate to Virtual Machines for Google Cloud. Azure Migrate is usually the cheapest path for Windows Server and SQL Server estates with Software Assurance, because of Azure Hybrid Benefit.
How much does cloud migration cost with AWS, Azure and Google Cloud tools?
The tools cost little. AWS Transform's core agents are free and MGN replicates each server free for 90 days. Azure Migrate is free for 180 days per server, then $25 a month per server. Google charges no tool fee for VM migration. Your real costs are replication infrastructure, test and cutover compute, data transfer, licensing and any server that replicates past its free window.
Which cloud migration tool is best for Windows Server and SQL Server?
Azure Migrate is the strongest fit when your Windows Server and SQL Server licenses carry active Software Assurance. Azure Hybrid Benefit lets you reuse those licenses, and Microsoft states savings of up to 85% for SQL Server against pay-as-you-go. Microsoft's licensing rules restrict bringing licenses without Software Assurance to AWS and Google dedicated hosts.
What is the difference between AWS Transform and Azure Migrate?
AWS Transform runs discovery, wave planning, landing zone setup and rehosting as a generally available, agent-driven workflow, with free replication for 90 days per server. Azure Migrate is a free hub for discovery, assessment and migration, with 180 free days of replication per server. Its AI agents are mostly in preview, and the migration agent cannot run replication.
AWS Migration Evaluator vs Azure Migrate: which should I use?
Use the assessment tool for the cloud you are considering. AWS Migration Evaluator is still open and free for building an AWS business case, and AWS Transform now offers assessments too. Azure Migrate's assessment is free and covers servers, SQL Server and web apps for Azure. To compare clouds, run the same RVTools export through both.
Can I migrate between AWS, Azure and Google Cloud with native tools?
Yes, but only into each provider's own cloud. Google's Migrate to Virtual Machines accepts AWS and Azure VMs as sources. Azure Migrate moves AWS and Google VMs as physical servers, with an agent on each VM. AWS Transform MGN supports other cloud providers, but AWS does not name Azure or Google Cloud, so test one server first.
Do Google Cloud migration tools handle mainframe migrations?
Yes. Google's Mainframe Assessment Tool has been generally available since April 2025, and Dual Run lets you run mainframe and cloud versions side by side before cutover. AWS also offers a generally available mainframe agent in AWS Transform. Gartner predicts more than 70% of mainframe exits started in 2026 will miss their intended benefits, so budget for manual review of converted code.
Does moving VMware to AWS, Azure or Google Cloud end my Broadcom licensing?
Not if you keep running VMware. Amazon EVS, Azure VMware Solution and Google Cloud VMware Engine all require your own VMware Cloud Foundation subscription. To leave Broadcom, migrate the VMs onto native instances with AWS Transform, Azure Migrate or Migrate to Virtual Machines.


