VMware HCX Replication-assisted vMotion is a transformative solution for VM mobility. For the first time VMware HCX combines the benefits of vSphere replication and vMotion in a single VM migration option. It immensely simplifies the planning, execution, and operationalization of large-scale mobility to public or private clouds.
I am excited to announce VMware HCX’s newest VM mobility service—Replication-assisted vMotion. Customers have been using VMware HCX vMotion to move workloads live, i.e. without any VM downtime. But VMware HCX vMotion is limited in the sense that it can only move one VM at a time, and the actual time it would take for the VM to move is not predictable. The time for VMware vMotion to complete depends on the size of the VM, the change rate, etc.
Customers want to move large number of VMs live simultaneously and they need the operation to be predictable. Now, HCX Replication-assisted vMotion lets them do that. HCX Replication-assisted vMotion combines VMware vSphere replication and vMotion to create a large-scale, predictable model to move workloads live to the destination.
Customers can now create a migration schedule during which a large set of VMs can move live (without any downtime) to the destination at the scheduled migration window. With Replication-assisted vMotion, vMotion is now predictable and scalable!
How does it work?
Replication-assisted vMotion is a new service in the HCX service list. Make sure that service is enabled. You will need the HCX Enterprise license to be able to use that service.
Once selected, a new migration option will be visible in the migration screen called “Replication-assisted vMotion.” Choose the set of VMs you want to migrate and select that option.
Once the VMs you wish to move using Replication-assisted vMotion are selected, go to the migration window scheduler and schedule a migration time. As soon as the batch of VMs are scheduled to migrate, VMware HCX kickstarts the vSphere replication to replicate the virtual disks to the destination. After the initial seeding is done, HCX continues to keep the replicated disks in sync via continuous delta syncs.
When the migration scheduled window is reached, VM disks are initially committed and the control transfers over to the VMware HCX vMotion service to do the final data sync and VM state transfer (i.e. live vMotion).
Key features
- Large scale live mobility: A large set of VMs can be submitted for live migration and the migration window can be specified with Replication-assisted vMotion
- Continuous replication: Once a set of VMs are selected for RAV, HCX will not only do the initial seeding, it will continue to replicate the delta changes until the migration window is reached
- Concurrency: With Replication-assisted vMotion, multiple VMs can be replicated simultaneously. When the switchover window is reached, the control will switch over to delta vMotion and one VM at a time will be moved live. Since all the replication and synchronization is done beforehand, the delta vMotion timer per VM is very small and is within the switch over schedule
- Legacy version support: Replication-assisted vMotion is supported with the source vSphere version starting from vSphere 5.5 and the destination vSphere version starting with 6.5. This lets customers use RAV to migrate from legacy sites to existing vSphere environments. They don’t have to upgrade their vSphere environment or wait for the next vSphere release.
VMware HCX Replication-assisted vMotion is a transformative solution for VM mobility. For the first time VMware HCX combines the benefits of vSphere replication and vMotion in a single VM migration option. It immensely simplifies the planning, execution, and operationalization of large-scale mobility to public or private clouds.
To learn more, visit https://cloud.vmware.com/vmware-hcx or join us at a VMworld Session. Join the conversation @VMwareHCX.
