Azure Migrate: how to evaluate your VMware environment to address workloads migration to Azure

The digital transformation process that affects companies often involves the migration of workloads hosted in their data centers to the cloud to obtain better results in terms of governance, security and cost efficiency. The innovation of migrating to the cloud frequently becomes a business priority to the point that this process is no longer a matter of "if" or "when", but the real question now is "how" to deal with this migration? In this regard, Microsoft has developed the Azure Migrate solution that allows you to deal with the most common migration scenarios reducing complexity and costs. This article describes in detail how it is possible to discover and assess virtual machines hosted in a VMware environment with Azure Migrate, so you can better address the migration path.

Azure Migrate Overview

Azure Migrate structure the process of migration in different phase: discovery, assessment, and migration. These three steps fall under the Microsoft Cloud Adoption Framework for Azure which, on the path to adopting cloud solutions, defines six main stages that apply to most organizations:

  • Strategy: defines the "business justification" and the expected results.
  • Plan: aligns the cloud adoption plan with business results.
  • Ready: Prepare your cloud environment for the changes you want.
  • Adopt: implements the changes you want in your IT and business processes.
    • Discovery, assessment, and migrating with Azure Migrate are part of this phase
  • Manage: implements basic operational guidelines and best practices.
  • Govern: evaluates and implements best practices in governance.

The Azure Migrate Hub provides all the tools to perform, monitor and analyze your migration path to Azure. This approach provides an integrated experience that provides continuity and provides an overall view of the migration process.

Figure 1 – Overview of Azure Migrate

Azure Migrate Hub includes Azure services and third-party solutions to address different migration scenarios:

  • Windows and Linux servers
  • SQL and non-SQL databases
  • Web apps
  • Virtual desktop infrastructure
  • Data

Discovery and assessment for VMware environments

The process of discoverying the VMware environment through Azure Migrate can generate a large amount of information, useful for assessing the status of workloads. The assessment process carries out an assessment of the environment and is able to answer critical questions such as:

  • Your virtual machine is ready to run in azure environment?

Azure Migrate reports if there are configurations that are not suitable for Azure, for example the operating system version or disk size are not supported. It also provides recommendations on how to correct these situations to avoid problems with the migration.

  • What size should the VM in Azure have?

Most organizations want to know the size of the virtual machine in Azure in advance to efficiently perform the workload, as sizing helps predict costs. Azure Migrate eliminates the need for calculations and translates CPUs, disk and memory of an on-premises system in an equivalent environment in Azure. The service recommends a specific virtual machine size and disk type based on the performance data collected.

  • How much will it cost to run in Azure?

Azure Migrate provides an estimate of the monthly costs of running servers in Azure.

  • Which applications are running on the system and their dependencies?

Thanks to Server Assessment it is possible to analyze cross-server application dependencies and consequently optimize strategies for interdependent movement of servers to Azure. Dependency identification can be performed with agents installed on each virtual machine or in VMware environments it can be performed without agents. When you use the agent-based solution, data is sent to Azure Log Analytics so you can analyze it in great detail to find hidden dependencies that might otherwise escape detection.

To start this VMware environment discovery process, you need to create a new Azure Migrate project, in the "servers" section:

Figure 2 – Creating an Azure Migrate project

After assigning it a name and defining in which geographical area of ​​Azure the metadata sent should reside, it is advisable to choose the tool to carry out the assessment. In this case we have chosen to adopt the Microsoft solution Server Assessment, but you can also adopt solutions from other vendors.

Figure 3 – Choice of the tool to carry out the assessment

In a similar way, it is also possible to choose the tool to be used to perform the migration.

At this point you can start the discovery process.

Figure 4 – Initiation of the discovery process

To identify the servers and workloads to be evaluated, you can import the Azure Migrate appliance into your local environment or use a manual method by importing a CSV file.

In the case of use of the appliance, the process can be summarized with the following steps:

Figure 5 – Discovery and assessment process of VMware environments

The activation of the Azure Migrate appliance for VMware environments is documented in this Microsoft article. After completing the deployment of the OVA template, you need to continue with the following steps.

Figure 6 – Configuring prerequisites

The appliance needs to be registered to the Azure Migrate project created in its subscription. To do this you need to enter a key generated directly from the Azure portal.

Figure 7 – Appliance registration

At the end of the registration it is necessary to provide the vCenter credentials, useful for discovering VMware virtual machines, and details to connect to the vCenter server. Furthermore, you can specify the credentials to use to detect installed applications and various dependencies, all in agentless mode. For further details please visit the Microsoft-specific documentation.

Figure 8 – Managing VMware credentials and sources

At the end of the discovery process it is possible to consult the data collected by the Azure portal.

Figure 9 – Server discovery

The next step involves the creation of the assessment process, going to define the properties according to your needs.

Figure 10 – Properties of the assessment process

Following, you must specify the systems that you intend to migrate, that will be the subject of the assessment.

Figure 11 – Selection of the machines on which to carry out the assessment

The assessment process, if based on performance data collected by the Azure Migrate virtual application, has a level of reliability expressed with a degree of confidence from 1 to 5.

Figure 12 – Assessments carried out and levels of reliability

The assumed sizing for Azure systems is calculated by examining the performance collected in the previous days, in particular:

  • RAM and CPU usage
  • IOPS and throughput for each disk connected to the virtual machine.
  • Network I/O to manage performance-based sizing for each network adapter connected to a virtual machine.

For more information on the assessment process, please consult this Microsoft's document.

The outcome of the assessment can be consulted directly from the Azure portal, where you can also download an Excel sheet with its details.

Figure 13 – Assessment details

For each system it is also possible to explore the various application dependencies:

Figure 14 – Application dependencies of a single server

All application dependencies discovered by Azure Migrate can also be exported from the Azure portal to an Excel sheet.

Conclusions

Easily move VMware workloads to Azure is an increasingly felt need in order to increase productivity thanks to greater elasticity and scalability offered by the public cloud. Using Azure Migrate, you can easily and accurately complete the discovery and assessment phases of your VMware environment. These phases are of fundamental importance in order to face the process of migrating VMware application workloads to Azure in the best possible way.

Please follow and like us: