Providing virtual machine high-availability and fault tolerance via solid-state backup drives
The present invention is directed to a virtualization system using a solid-state drive for disaster recovery.
1. A process, comprising:
providing first, second, and third memory devices, wherein the first memory device comprises at least a first virtual machine and the at least a first virtual machine executes on a processor in communication with the first memory device, the second memory device, and the third memory device, and wherein the second memory device is long term storage discrete from the first memory device and the third memory device is a solid-state drive discrete from the first and second memory devices;
mirroring at least one of filesystem, first memory, and network states of the at least a first virtual machine in the second memory device;
detecting a disk failure impacting the second memory device; and
in response to detecting the disk failure of the second memory device:
switching references and pointers for the at least a first virtual machine from the first and second memory devices to volatile memory in the third memory device;
running an instance of the at least a first virtual machine on the third memory device and terminating running the at least a first virtual machine on the first memory device;
detecting a power outage for the third memory device;
in response to detecting the power outage for the third memory device and to conserve backup power, determining that the instance of the first virtual machine, but not a second virtual machine, is to continue operation; and
terminating execution of the second virtual machine by the processor.
2. The process of claim 1 , wherein the first memory device is in a server, wherein the second memory device is dislocated from the first memory device, wherein the first and second memory devices are in communication via a network, and wherein the second memory device is disk storage and further comprising of a limited period uninterruptible Power Supply (“UPS”) with a power-outage notification/detection capability.
3. The process of claim 1 , wherein the switching comprises:
parsing the at least a first virtual machine to provide physical pages of the at least a first virtual machine in the first memory device;
remapping the physical pages in the first memory device to physical pages in the third memory device; and
thereafter restarting, on the third memory device, operation of the at least a first virtual machine.
4. A tangible, non-transitory computer readable medium comprising processor executable instructions that, when executed by the processor, perform the steps of claim 1 .
5. A machine, comprising:
first, second, and third memory devices, wherein the first memory device comprises at least a first virtual machine and wherein the second memory device is long term storage discrete from the first memory device and the third memory device is a solid-state drive discrete from the first and second memory devices; and
a processor, operable to:
execute the at least a first virtual machine,
mirror at least one of filesystem, first memory, and network states of the at least a first virtual machine in the second memory device;
detect a disk failure impacting the second memory device; and
switching references and pointers for the at least a first virtual machine from the first memory device and the second memory device to volatile memory in the third memory device, in response to detecting the disk failure of the second memory device;
using data storage pointers to volatile memory in the solid-state drive for machine state information of the at least a first virtual machine formerly mirrored by the second memory device to continue running the at least a first virtual machine on the first memory device;
wherein the processor is further operable to detect a power outage impacting at least the third memory device and in response to detecting the power outage and to conserve backup power, determine that the first virtual machine, but not a second virtual machine, is to continue operation; and
in response to determining that the second virtual machine is not to continue operation, terminate execution of the second virtual machine by the processor.
6. The machine of claim 5 , wherein the first memory device is in a server, wherein the second memory device is dislocated from the first memory device, wherein the first and second memory devices are in communication via a network, and wherein the second memory device is disk storage.
7. The machine of claim 5 , wherein in response to the disk failure:
parsing the at least a first virtual machine to provide physical pages of the at least a first virtual machine in the first memory device;
remapping the physical pages in the first memory device to physical pages in the third memory device; and
thereafter restarting, on the third memory device, operation of the at least a first virtual machine.
8. The machine of claim 7 , wherein, while the at least a first virtual machine runs on the third memory device, the processor mirrors, on the first memory device, at least one of filesystem, first memory, and network states of the at least a first virtual machine.
9. The machine of claim 7 , wherein the third memory device comprises volatile and nonvolatile memory and wherein, while the at least a first virtual machine runs on volatile memory of the third memory device, the processor mirrors, on nonvolatile memory of the third memory device, at least one of filesystem, first memory, and network states of the at least a first virtual machine.