Method and system for health rank based virtual machine restoration using a conformal framework

Patent metadata

Field	Value
Publication number	US-11755433-B2
Application number	US-202017131450-A
Country	US
Kind code	B2
Filing date	Dec 22, 2020
Priority date	Dec 22, 2020
Publication date	Sep 12, 2023
Grant date	Sep 12, 2023

Abstract

A method and system for health rank based virtual machine restoration using a conformal framework. With respect to enterprise data protection, solutions need to address two primary responsibilities: at the onset of failure or disaster, restore any running applications, as well as any lost or damaged data; and minimize (if not eliminate) the future occurrence of such failures or disasters. In tackling the latter responsibility, the disclosed method and system leverage health-score assessments to ensure the restoration of virtual machines onto healthy infrastructure. The aforementioned health-score assessments employ clustering to identify, and a conformal framework to rank, healthy hosts onto which virtual machines may be restored.

First claim

What is claimed is:

1. A method for virtual machine restoration, comprising:
detecting a failure of a source virtual machine;
in response to detecting the failure:
identifying a set of available virtual machines;
collecting performance metrics for each available virtual machine in the set of available virtual machines;
assigning, based on the performance metrics, each available virtual machine in the set of available virtual machines to one selected from a group consisting of a healthy class and an unhealthy class;
ranking, in descending order and to obtain a ranked subset of available virtual machines, a subset of the set of available virtual machines based on a health score calculated for each available virtual machine in the subset of the set of available virtual machines, wherein each available virtual machine in the ranked subset of available virtual machines is a member of the healthy class;
selecting a target virtual machine from the ranked subset of available virtual machines; and
restoring, onto the target virtual machine, at least a defined process once hosted on the source virtual machine,
wherein the health score calculated for each available virtual machine in the subset of the set of available virtual machines is provided using a conformal framework, and
wherein assignment of each available virtual machine in the set of available virtual machines, to one selected from the group consisting of the healthy class and the unhealthy class, resulted from machine learning classification performed through cluster analysis.

2. The method of claim 1 , wherein the conformal framework associates a confidence value with each assignment mapping an available virtual machine in the subset of the set of available virtual machines to the healthy class.

3. The method of claim 2 , wherein the health score, calculated for each available virtual machine in the subset of the set of available virtual machines, comprises the confidence value associated with the assignment mapping the available virtual machine to the healthy class.

4. The method of claim 1 , further comprising:
prior to selecting the target virtual machine:
identifying a data criticality associated with the at least defined process once hosted on the source virtual machine,
wherein selection of the target virtual machine is based on the data criticality.

5. The method of claim 4 , further comprising:
prior to identifying the data criticality associated with the at least defined process once hosted on the source virtual machine:
partitioning, based on a health score threshold, the healthy class into a premium sub-class and a non-premium sub-class,
wherein the health score, calculated for each available virtual machine that is a member of the premium sub-class, at least matches the health score threshold,
wherein the health score, calculated for each available virtual machine that is a member of the non-premium sub-class, fails to at least match the health score threshold.

6. The method of claim 5 , wherein the target virtual machine is selected from available virtual machines of the premium sub-class when the data criticality reflects that the at least defined process is of high-importance.

7. The method of claim 5 , wherein the target virtual machine is selected from available virtual machines of the non-premium sub-class when the data criticality reflects that the at least defined process is of low-importance.

8. The method of claim 1 , further comprising:
prior to restoring the at least defined process, once hosted on the source virtual machine, onto the target virtual machine:
making a determination that a worker node comprises sufficient available storage to accommodate defined process data pertinent to the at least defined process,
wherein the target virtual machine resides on the worker node.

9. A non-transitory computer readable medium (CRM) comprising computer readable program code, which when executed by a computer processor, enables the computer processor to:
detect a failure of a source virtual machine;
in response to detecting the failure:
identify a set of available virtual machines;
collect performance metrics for each available virtual machine in the set of available virtual machines;
assign, based on the performance metrics, each available virtual machine in the set of available virtual machines to one selected from a group consisting of a healthy class and an unhealthy class;
rank, in descending order and to obtain a ranked subset of available virtual machines, a subset of the set of available virtual machines based on a health score calculated for each available virtual machine in the subset of the set of available virtual machines, wherein each available virtual machine in the ranked subset of available virtual machines is a member of the healthy class;
select a target virtual machine from the ranked subset of available virtual machines; and
restore, onto the target virtual machine, at least a defined process once hosted on the source virtual machine,
wherein the health score calculated for each available virtual machine in the subset of the set of available virtual machines is provided using a conformal framework, and
wherein assignment of each available virtual machine in the set of available virtual machines, to one selected from the group consisting of the healthy class and the unhealthy class, resulted from machine learning classification performed through cluster analysis.

10. The non-transitory CRM of claim 9 , wherein the conformal framework associates a confidence value with each assignment mapping an available virtual machine in the subset of the set of available virtual machines to the healthy class.

11. The non-transitory CRM of claim 10 , wherein the health score, calculated for each available virtual machine in the subset of the set of available virtual machines, comprises the confidence value associated with the assignment mapping the available virtual machine to the healthy class.

12. The non-transitory CRM of claim 9 , further comprising computer readable program code, which when executed by the computer processor, further enables the computer processor to:
prior to selecting the target virtual machine:
identify a data criticality associated with the at least defined process once hosted on the source virtual machine,
wherein selection of the target virtual machine is based on the data criticality.

13. The non-transitory CRM of claim 12 , further comprising computer readable program code, which when executed by the computer processor, further enables the computer processor to:
prior to identifying the data criticality associated with the at least defined process once hosted on the source virtual machine:
partition, based on a health score threshold, the healthy class into a premium sub-class and a non-premium sub-class,
wherein the health score, calculated for each available virtual machine that is a member of the premium sub-class, at least matches the health score threshold,
wherein the health score, calculated for each available virtual machine that is a member of the non-premium sub-class, fails to at least match the health score threshold.

14. The non-transitory CRM of claim 13 , wherein the target virtual machine is selected from available virtual machines of the premium sub-class when the data criticality reflects that the at least defined process is of high-importance.

15. The non-transitory CRM of claim 13 , wherein the target virtual machine is selected from available virtual machines of the non-premium sub-class when the data criticality reflects that the at leas

Assignees

• Emc Ip Holding Co Llc

Inventors

• Chopra Shelesh
• Ambaljeri Mahantesh M
• Vishwakarma Rahul Deo
• Singh Gopal
• Prasad Parmeshwr

Classifications

• G06F11/3055 Primary

  Monitoring arrangements for monitoring the status of the computing system or of the computing system component, e.g. monitoring if the computing system is on, off, available, not available (error or fault processing without redundancy G06F11/0703; error detection or correction by redundancy in data representation G06F11/08; error detection or correction of the data by redundancy in operations G06F11/14; error detection or correction by redundancy in hardware G06F11/16) · CPC title

• G06F11/203 Primary

  using migration · CPC title

• G06F2201/815

  Virtual · CPC title

• G06F11/301

  where the computing system is a virtual computing platform, e.g. logically partitioned systems (virtual machines G06F9/45533; logical partitioning of resources G06F9/5077) · CPC title

• G06F11/1484

  involving virtual machines · CPC title