VMware to AWS: A clear explanation of virtual machine migration procedures, tools, and configuration examples

An increasing number of companies are migrating VMware virtual machines running in on-premises environments to AWS. Cloud migration offers many benefits, including reducing hardware obsolescence, rising operational costs, BCP measures, and ensuring scalability. However, when it comes to actual migration, challenges arise, such as "Which tools should I use?", "Which procedures are safe?", and "How much can I reduce downtime?"

This article focuses on migrating from a VMware environment to AWS, and provides a clear explanation of the features and usage of representative migration tools such as AWS MGN and VM Import/Export, as well as an overview of the migration procedure, configuration patterns, and points to note on the ground.

Why VMware is moving to AWS

Challenges faced by on-premise environments

On-premise system operations have supported corporate IT infrastructure for many years, but in recent years, many issues have arisen in maintaining and operating them.

The first major problem is the aging of hardware and rising maintenance costs. Servers and storage devices require updating every few years, but the update cycle is becoming shorter and parts and maintenance contract costs are rising every year. Update work requires system downtime, so planning and staffing to avoid impacts on business operations are also costly.

Next, there are the limitations of disaster prevention measures and business continuity plans (BCPs). Duplicating systems in an on-premises environment, backing them up in remote locations, and building redundant configurations requires a large initial investment and ongoing maintenance costs. In reality, many companies are unable to take sufficient measures, even while recognizing the importance of BCP measures.

Lack of scalability and operational burden are also serious issues. Even if the number of accesses or data volume increases sharply, on-premise systems cannot keep up with the increase in hardware, forcing IT departments to constantly secure excess resources to handle peak demand. As a result, excess resources pile up as unnecessary fixed costs, and IT departments are burdened with operational burdens such as troubleshooting and patch application.

Benefits of migrating to AWS

Migrating to Amazon Web Services (AWS) is gaining attention as a way to resolve these issues. By utilizing AWS, you can gain flexibility and efficiency that would be difficult to achieve in a traditional on-premises environment.

First, "reducing initial investment and optimizing costs."

There's no need to purchase the servers and network equipment required for on-premises deployments, significantly reducing initial investment. AWS uses a pay-as-you-go model, meaning you only pay for what you use, allowing you to flexibly increase or decrease resources and reducing unnecessary fixed costs. By utilizing pricing options such as Reserved Instances and Savings Plans, you can also efficiently reduce ongoing costs.

Next is "improved high availability and fault tolerance."

AWS makes it easy to build a redundant configuration using multiple data centers (AZs: Availability Zones), and automatically switches over in the event of a system failure, providing a foundation for continuing business operations even in the event of a disaster.

Another major attraction is the "improved operational efficiency through the use of the latest services."

By utilizing managed services such as RDS, ECS, and Lambda, you can reduce tasks such as server maintenance and patch application. Monitoring and automation can also be integrated with AWS services such as CloudWatch and EventBridge, allowing you to achieve an efficient operational system even with a small number of people.

These benefits allow companies to reduce system maintenance costs and devote more resources to business expansion and new service development, which is the main reason why migration from on-premises to AWS is accelerating.

What you need to prepare for a successful migration

Migrating from VMware to AWS is not just a simple system migration; it is a project that will have a significant impact on your business. If you proceed without sufficient preparation, you may encounter problems such as extended downtime and unexpectedly high operational costs. This article explains the preparations you need to make to ensure a smooth and safe migration.

Inventory and visualization of the current environment

The first step is to accurately understand and visualize the on-premise environment currently in operation.

Use management tools such as vCenter or vSphere to identify in detail the number of virtual machines and hardware resources such as CPU, memory, and storage capacity. In addition, organize the dependencies of applications and middleware.

It is especially important to understand which applications are running on which VMs and how they are connected to external systems. If you overlook dependencies, you may encounter problems after migration, such as "some systems not working" or "connections being lost."

Visualizing the inventory results in charts and tables so that all stakeholders can share the same information will make the transition plan formulation smoother.

Migration goals and requirements definition

It is important to clarify "why you are migrating to AWS" and "what you want to achieve after the migration." For example, define the following requirements in advance:

• Availability objectives: How much system downtime can you tolerate?

• Security requirements: access management, encryption, and audit log maintenance level

• Operational cost target: How much do you want to keep cloud usage fees?

• Downtime tolerance: How much business can be stopped when switching to production?

• Transition schedule: Phased transition or all-at-once switchover?

Defining these will not only enable you to select the appropriate migration method and AWS services, but will also ensure consistency in post-migration operational design.

Reviewing the network and connectivity design

AWS migration requires a secure and stable connection between on-premises and the cloud. Typical connection methods include VPN connection and AWS Direct Connect.

• VPN connection

  • It can be built in a short time and initial costs are kept low.

  • However, the bandwidth is limited and speeds may be insufficient for large-scale migrations.

• Direct Connect

  • High-bandwidth, low-latency connection via dedicated line

  • Suitable for production use and large-volume data migration

Many companies are adopting a hybrid configuration, starting with a VPN in the early stages of migration and then switching to Direct Connect when it is time to switch to production.

Additionally, you must prepare in advance the VPC design and security group rule settings on the AWS side. If you do not clarify the IP range and port requirements, there is a risk that communication will be cut off after migration. The network design should be reviewed by both on-premises and AWS personnel, and migration work should proceed with no misunderstandings.

VMware to AWS migration steps

We will explain a typical migration flow in six steps.

Step 1 | Minimize risk by conducting a proof of concept (PoC)

First, we conduct a small-scale proof of concept (PoC) to test the actual migration.

If you suddenly migrate the production system, there is a risk that unexpected problems will cause business operations to stop. In the PoC, we check the following on a test VM or a limited system.

• Check the operation of migration tools (AWS MGN and VM Import/Export)

• Network connection stability and transfer speed

• Estimated downtime

• System operation on AWS (compatibility and performance)

By identifying issues and considering countermeasures, you can minimize the risks of migrating to production.

Step 2 | Determine migration method and tools

Based on the results of the PoC, we will officially decide on the migration method and tools to be used.

There are two main methods for migrating from VMware to AWS:

• Lift and shift (re-host)

  • Migrate existing VMs to AWS almost as they are. Migration is possible in a short period of time, but optimization is limited.

  • → AWS MGN is the best choice

• Re-platform/partial modernization

  • A method of replacing part of your infrastructure with AWS managed services. This increases migration costs, but reduces operational burden.

  • → Use AWS MGN + RDS, EFS, and other services together

Choose the method that best suits your migration size, downtime tolerance, and cost constraints.

Step 3 | Prepare your AWS environment and connection settings

Build the AWS environment to which you will migrate and prepare to connect it to your on-premises environment.

• VPC design

  • Set up subnets and security groups with an eye on post-migration operations

• Connection method selection

  • Choose VPN or Direct Connect and set up your connection

• IAM design

  • Clearly defined permission management

• Monitoring Settings

  • Build a monitoring infrastructure using CloudWatch, etc.

If you do not organize IP ranges, communication ports, etc. in advance, you may encounter communication problems after the migration.

Step 4 | Replicate/Import the VM to be migrated

Once preparations are complete, copy the VMs to be migrated to AWS. In large-scale environments, AWS MGN is used to replicate on-premises VMs to AWS in real time. This allows you to synchronize the on-premises and cloud states until just before migration, minimizing downtime.

For small-scale migrations or testing environments, you can also use VM Import/Export to import VM images directly into AWS.

Step 5 | Test migration and validation

After replication, a test migration will be performed on AWS.

We will verify the following points and resolve any issues that arise before moving to production.

• Does the system run properly on AWS?

• Are there any problems with dependencies between applications?

• Are your network and external connections working properly?

• Does the performance meet your requirements?

The test results are documented and agreed upon by all parties before moving to production.

Step 6 | Production migration and operation switchover

Finally, switch your production system to AWS. To minimize the impact, we recommend switching over gradually using blue/green deployments and canary releases.

After the switchover, monitor the systems running on AWS and, if there are no problems, gradually shut down the on-premises environment. Also, immediately after the migration, use tools such as AWS Cost Explorer to analyze resource usage and optimize unnecessary costs.

Major migration tools and how to choose one

We will explain the positioning of older services and third-party tools, focusing on genuine AWS tools.

AWS Application Migration Service (AWS MGN)

AWS MGN is the standard migration tool currently recommended by AWS. Its greatest strength is that it can replicate on-premises VMs to AWS in real time, minimizing downtime during switchover.

Features

• Continuous data synchronization keeps the on-premise and cloud environments in the same state until just before migration

• Reduce downtime and enjoy peace of mind even with mission-critical systems

• As it is an official AWS service, there is a comprehensive support system

Applicable Scenes

• Cases where you want to minimize the impact on business operations, such as business systems currently in production

• Large-scale projects involving the migration of dozens of VMs

VM Import/Export

VM Import/Export is a service that allows you to import/export VMware virtual machine images directly to AWS. While it does not have the continuous replication function of AWS MGN, it makes it easy to bring VMs onto AWS.

Features

• VM images can be uploaded directly to AWS and used as AMIs

• Conversely, exporting from AWS to a VMware environment is also possible.

• No configuration changes are required, making it suitable for small-scale migrations

Applicable Scenes

• Migrating small-scale testing and development environments

• Systems that can tolerate some downtime

• Cases where you want to move VMs between AWS and on-premises

CloudEndure Migration positioning and integration into AWS MGN

CloudEndure Migration was previously offered as an official AWS tool, but it has now been integrated into AWS MGN and is no longer recommended for new deployments. Even if you are using it in your existing environment, the basic policy going forward will be to migrate to AWS MGN.

AWS MGN inherits CloudEndure technology and has improved UI and functionality, so companies planning to migrate should choose AWS MGN.

Comparison points when using third-party tools

If you have special requirements that cannot be met by official AWS tools, you may consider using third-party tools. Compare the following points.

• Supported virtualization platforms: Support for platforms other than VMware (Hyper-V, KVM, etc.)

• Automation capabilities: To what extent can migration planning, testing, and switchover procedures be automated?

• Cost structure: Total cost including license fee, pay-per-use fee, and support fee

• Support system: Japanese support, 24-hour support, collaboration with AWS

• Security and compliance: encryption, log management, regulatory compliance

However, AWS MGN and VM Import/Export, which are officially provided by AWS, can cover most migration needs. Third parties should be considered only if you have a complex multi-cloud environment or special compliance requirements.

Example of AWS configuration after migration

The configuration after migrating from VMware to AWS will vary greatly depending on the purpose of the migration and growth strategy. Here we will introduce three typical patterns.

Lift & Shift (transfer as is)

The simplest configuration is the "lift and shift" model, which simply recreates an existing VMware environment on AWS.

Features

• Run the existing system on AWS EC2 with almost no changes to the system configuration

• The biggest advantage is that it requires less design work and can be migrated in a short period of time.

Benefit

• You can quickly avoid the risk of maintenance expiring or failure of on-premise equipment

• Initial costs are low because configuration changes and application modifications are minimal.

Warnings

• AWS's features cannot be fully utilized, limiting cost optimization

• Operational load and security issues may remain the same as in the on-premise era

This is suitable for those who prioritize speed and want to migrate to the cloud for the time being, and then adopt a strategy of gradually improving things.

Modernize (step-by-step improvement)

After the migration, AWS managed services will be utilized to gradually optimize operations.

Features

• Replace some of the systems migrated using Lift & Shift with AWS managed services

• Common practices include migrating databases to Amazon RDS, integrating logs into CloudWatch Logs, and improving availability with Auto Scaling.

Benefit

• Reduces operational burden and reduces troubleshooting and maintenance work

• Gradually shifting to cloud native allows risk dispersion

• Cost efficiency can be optimized in the medium to long term

Usage examples

• Auto Scaling automatically increases or decreases EC2 instances according to load

• Centralize backup policies with AWS Backup

• Automate monitoring and notifications with CloudWatch and EventBridge

This configuration is suitable for companies that want to avoid major changes during the initial migration and want to optimize over time.

Future-proof refactoring

This is a full-scale cloud-native approach in which the system is redesigned from scratch with the cloud in mind.

Features

• Breaking down traditional monolithic structures and adopting microservices and serverless architectures

• Representative examples include Kubernetes operation using EKS and a completely serverless configuration using AWS Lambda.

Benefit

• Improving development speed and ensuring scalability

• Operational automation has progressed, allowing a small number of people to manage large-scale systems.

• Build a flexible foundation that can quickly respond to changing business requirements

Warnings

• Design and development efforts are significant, making it unsuitable for short-term migration

• Advanced AWS skills and systems required

This is the ideal configuration for companies looking to grow their business over the long term or to eliminate technical debt in their systems.

Common migration mistakes and solutions

There are many cases where problems arise after migration due to insufficient planning or design errors. Here we will introduce some common mistakes seen in practice and how to deal with them.

Licensing issues (Windows Server, etc.)

Software licenses used on-premises may not be usable after migration to AWS. In particular, the licensing format for Windows Server and SQL Server may differ when used in a cloud environment.

Common mistakes

• Failure to meet BYOL (Bring Your Own License) requirements and pointed out during an audit

• Migrating without checking the license terms resulted in additional costs

• Older OS versions are not supported by AWS

(I.e.

• Verify licensing terms with software vendors during migration planning

• Use AWS-specific licenses (License Included)

• If necessary, upgrade your OS to the latest version before migrating

Security misconfiguration

AWS has a wide range of security features, but incorrect configuration can lead to serious risks. In some cases, configuration errors are discovered after migration.

Common mistakes

• Keep the security group open at "0.0.0.0/0"

• IAM policies are vague and over-permitted

• Starting operations without implementing log monitoring or encryption settings

(I.e.

• IAM is designed based on the principle of least privilege, and operational rules are clearly documented.

• Use AWS Config and Security Hub to automatically check configurations

• Obtain and monitor audit logs with CloudTrail and CloudWatch Logs

• Involve security personnel in projects from the early stages

Migration delays due to insufficient network bandwidth

When migrating to AWS, network bandwidth tends to become a bottleneck because large amounts of VMs and data are transferred to the cloud.

Common mistakes

• As a result of starting the migration using only a VPN connection, the transfer speed was too slow, causing delays to the schedule.

• Direct Connect was used, but bandwidth design was insufficient, causing communication congestion during migration.

• Communication interruptions occur due to insufficient network device settings

(I.e.

• Measure actual transfer speeds at the initial PoC stage and estimate required bandwidth

• Leverage AWS Direct Connect for large-scale migrations

• For data migrations of several hundred TB or more, AWS Snowball is used to reduce bandwidth load.

• Network design is reviewed by both on-premise and AWS teams

Operation started without adequate monitoring and operational design

Even if the migration work itself is completed, if production operations begin without an operational system in place, response will be delayed when problems occur.

Common mistakes

• No alerts are sent when a problem occurs on AWS

• Although a backup was taken, the restore procedure has not been verified and recovery is not possible.

• Lack of AWS skills among operations staff leads to configuration changes

(I.e.

• Establish alerts using CloudWatch and EventBridge

• Conduct regular recovery tests and check RTO/RPO

• Provide AWS training to the operations team

• Create an operational design document and share it with stakeholders

My Feelings, Then and Now

The key to a successful migration from VMware to AWS is preparation, tool selection, and a phased migration plan. First, understand the current situation and identify risks through a PoC, then use tools such as AWS MGN to ensure a safe migration.

After migration, it is practical to first prioritize speed with "Lift & Shift," and then optimize through modernization and refactoring. Take proactive measures to address typical failure factors, such as licenses, security, networks, and operational design, and build a cloud infrastructure that can be used for the long term.