Bulk migration uses the host-based replication to move a virtual machine between HCX data centers.

To reduce the downtime, the source VM remains online during the replication and is bootstrapped on the destination ESX host after replication completes.

A Bulk Migration request triggers the following actions:

  1. Replication begins a full synchronization transfer to the remote site. The time it takes to replicate is a function of the size of the VM and available bandwidth.

  2. Replication bandwidth consumption varies depending on how the workload changes blocks on the disk.

  3. When full synchronization finishes, a delta synchronization occurs.

  4. When the delta synchronization finishes, a switchover is triggered. You can start immediately or delay the switchover until a specific time using the scheduled migration option. By using the scheduled migration option, the switchover can occur during a maintenance window.

  5. Following the switchover, the source VM is powered-off, and the migrated replica is powered-on. If for some reason the VM cannot power on, the new VM is powered off (or remains powered off) and the original is powered on. You must have sufficient resources to power on the VM.

  6. HCX Manager renames the original VM using a POSIX timestamp suffix to avoid a naming conflict with the migrated VM. If you have not enabled the Retain MAC option, the migrated VM obtains a new MAC address.

  7. The migration completes.

VMware HCX copies the original VM to the Migrated VMs folder in the vSphere Templates view. You can recover a saved VM.

Note:

There are two uses for these copies:

  1. The copy can act as seed, in the event the VM on Site B must be protected on Site A.

  2. Protect against any VM corruption (due to external factors) during migration.

Requirements for HCX Bulk Migration

  • The Hybrid Interconnect Service and Bulk Migration Service must be enabled and in a healthy state in the relevant service mesh.

  • The resources to create, power on and use the virtual machine must be available in the destination environment.

  • Virtual machines must be running Hardware Version 7 or higher.

  • Virtual machines must have VMware Tools installed.

  • Virtual machines must reside in a Service Cluster (defined in the Compute Profile).

  • Network Extension is required for low downtime migration operations.

  • Personalization Scripts and System Identity changes (Hostname, IP, SID) require the system to be rebooted one additional time during the switchover phase.

  • Bulk Migration potential throughput can vary depending on bandwidth available for migrations, latency, available CPU/MEM/IOPS, and disk read speed. For successful switchover phase, the bandwidth and network conditions must be sufficient to satisfy the operation considering the dataset and virtual machine data change rate. For more information about how to determine bandwidth requirements, see Bandwidth Requirements for vSphere Replication.

Restrictions for HCX Bulk Migration

  • Virtual machines with Raw Device Mappings (RDM) in Physical Compatibility mode cannot be bulk migrated.

  • Virtual machines with mounted ISO images cannot be migrated. The HCX bulk migration operation can be used for force unmount ISO images.

  • Virtual machine snapshots will not be migrated. The HCX bulk migration operation has an option to remove the snapshots.

  • Virtual machines with DirectPath I/O configurations cannot be migrated without first removing the DirectPath device.

  • Virtual machines with Multi-Writer or FT-enabled virtual machines can be migrated, but the multi-writer configuration will no longer be functional.

  • Virtual machines with SCSI bus sharing cannot be migrated.

  • Virtual machines that cannot be gracefully powered off cannot be migrated. HCX can override with the Force Power-off VM option.

  • NSX Security tags and configurations related to the virtual machine will not be migrated. vSphere tags will be migrated.

  • With the Bulk migration option, new disk UUIDs are generated at the destination environment. Use HCX vMotion when the application has disk UUID-related dependencies.

  • Virtual machines using virtual NVMe (vNVME) Controllers cannot be migrated.