Planning is crucial in any migration effort and Azure Migrate helps you plan your datacenter migration to Azure. It helps you discover your on-premises environment, create move-groups and assess the move groups for migration to Azure.
How does assessing the move-group help in migration?
Assessing your move-groups helps you identify:
- Azure readiness: Whether the on-premises VMs are ready for running on Azure or not?
- Azure sizing: What should be the right VM size or disk type for the VM in Azure?
- Azure cost estimation: What would be the cost of running the VMs in Azure?
How does Azure Migrate assess if a VM is ready for Azure?
Azure Migrate analyses the configuration details of each VM to identify if the configuration is supported in Azure. For example, it analyses the operating system of the VM to identify if the OS is supported by Azure. Similarly, it analyses properties like the number of cores, memory size, disk size of each attached to the VM to identify if the required configuration is supported in Azure. Based on the analysis, it categorizes the VMs into Ready, Conditionally Ready and Not Ready categories. For the VMs that are in the Conditionally Ready or Not Ready categories, Azure Migrate provides remediation guidance regarding how the readiness issues can be fixed.
How does Azure Migrate perform sizing?
Azure Migrate helps you do intelligent rightsizing for Azure by looking at the performance data of the on-premises VMs and disks. It looks at the utilization data for CPU and memory for each VM and recommends a VM size in Azure accordingly. Similarly, it looks at the disk utilization data for each disk and recommends a disk type in Azure. Performance-based sizing helps you ensure that you are not over-allocating your VMs in Azure and saving cost.
Do I need to change my vCenter statistics level for performance data collection?
You don’t need to that anymore! With the continuous discovery option in Azure Migrate, you can now profile your on-premises environment to collect real-time utilization data for VMs and disks. Earlier, Azure Migrate required you to set the statistics settings in vCenter Server to level three and wait for at least a day before you could start the discovery. The continuous discovery model does not depend on the vCenter Server statistics settings and you do not have to change the statistics settings in vCenter Server. Additionally, since it collects real-time utilization data, the risk of under-sizing the VMs for Azure due to collection of average counters from vCenter Server is no longer there. Learn more about continuous discovery in Azure Migrate.
How do I create move-groups for assessments?
Azure Migrate helps you visualize dependencies of on-premises VMs and create move groups based on the dependencies. If you have legacy applications in your on-premises environment and are struggling to identify the set of VMs that should be moved together to Azure, the dependency visualization functionality is ideal for you. To visualize dependencies, Azure Migrate leverages the Service Map solution in Azure and displays the network connections going in and out of the on-premises VMs. The dependency map helps you identify VMs that are related to each other. Once you identify the related VMs you can create groups of VMs that are self-sufficient and ensure that you are not leaving anything behind when you are migrating to Azure.
I am already using Service Map and have my on-premises agents configured to a Log Analytics workspace, can I attach my existing workspace with Azure Migrate?
Absolutely, in addition to allowing the creation of new Log Analytics workspaces, Azure Migrate now allows you to attach an existing workspace to the migration project. So, if you have an existing Log Analytics workspace, you can attach the same to Azure Migrate to leverage dependency visualization. If you have already installed the MMA and dependency agents on your on-premises VMs and configured them to talk to the workspace, you do not have to do anything else, once you have attached the workspace to the Azure Migrate project, the dependencies will start showing up in Azure Migrate.
Do I have to pay anything for dependency visualization, if I use an existing workspace?
Even if you use an existing workspace, the dependency visualization functionality would be free in Azure Migrate, for the first 180 days, from the day of associating a workspace to a project. However, if you are using any other solution in the workspace, other than Service Map, standard Log Analytics charges would apply. Learn more about Azure Migrate pricing.
This is great! But, where is the data discovered from the on-premises environment stored?
Azure Migrate only collects metadata about your on-premises VMs and allows you to select an Azure geography where the discovered metadata would be stored. Based on the geography selected, Azure Migrate stores the discovered metadata in one of the Azure regions in the geography. Azure Migrate currently only supports the United States as the Azure geography for metadata storage. Support for Azure Government is planned for October 2018, followed by Europe and Asia in December 2018, support for other Azure geographies will be enabled in future. Note that, even though the metadata would be stored in the United States currently, you can still plan your migration to any Azure region by specifying the appropriate Azure region in the assessment properties.
With the new enhancements, we believe that Azure Migrate will make it even more easy for you to bring your on-premises VMs to Azure. So, if you haven’t already, get started now with Azure Migrate!
Visit the Azure migration center to learn how to apply Azure Migrate in your cloud migration journey.