CPC G06F 11/203 (2013.01) [G06F 11/0712 (2013.01); G06F 11/2041 (2013.01); G06F 11/3419 (2013.01)] | 17 Claims |
1. A method of migrating virtual machine (VM) comprising:
monitoring a status of at least one compute node;
determining whether the at least one compute node meets a trigger condition; wherein the trigger condition comprising a time period of lost connection of the at least one compute node reaching a predetermined time period, or an unstable status of the at least one compute node; and
if the at least one compute node meets the trigger condition, transmitting a message to a control node to migrate the VM;
wherein if there are multiple compute nodes determined to meet the trigger condition, the method further comprises:
when starting to migrate the VM of one of the multiple compute nodes, switching to an insurance status to stop migrating the VM of other compute nodes of the multiple compute nodes; and
after finishing the migration of the VM of the compute node, switching to a working status to continuously monitoring the status of the at least one compute node.
|