Sangfor HCI and aSV provide a unified infrastructure combining compute, storage, networking, and built-in security to simplify deployment, operations, and services.
• For x86, core2duo, Haswell-noTSX, Cascadelake-Server-noTSX, and Icelake-Server-v4 are supported.
• For Arm, phenom, EPYC, and EPYC-Milan are supported.
• For C86, phenom, Dhyana, and EPYC are supported.
Precautions
When migrating, or performing DRS, HA, and disaster recovery for a VM, it is recommended to select a destination host that is compatible with its vCPU; otherwise, the migration or HA will fail, or the VM cannot be recovered at the secondary site.
Rolling active upgrades from HCI 6.10.0 to later versions require VMs with the new instruction set to be shut down before being migrated to hosts that do not support their new instruction set.
A host cannot be added to a cluster if its supported vCPU model is lower than the vCPU model of any VM in the cluster.
If a cluster is added with a host whose supported vCPU model is lower than its default vCPU model but higher than or equal to the vCPU models of all the VMs in it, its default vCPU model will be lowered to that of the added host and remain unchanged after removing the host.
Prerequisites
None.
Steps
Go to System > Advanced > Default vCPU Settings to select the default vCPU model for the cluster.
Select the vCPU model when creating and editing a VM.
A VM cannot be migrated to a destination cluster that does not support the vCPU model of the VM.
The vCPU model will be automatically set to Haswell-noTSX when creating, editing, cloning, and importing a VM and deploying a VM from a template.