System and method to achieve virtual machine backup load balancing using machine learning

Patent metadata

Field	Value
Publication number	US-2021064479-A1
Application number	US-201916560926-A
Country	US
Kind code	A1
Filing date	Sep 4, 2019
Priority date	Sep 4, 2019
Publication date	Mar 4, 2021
Grant date	—

Abstract

A method for performing a backup operation includes obtaining a backup request, wherein the backup request specifies a virtual machine (VM) set, and, in response to the backup request: identifying a first set of virtual machines (VMs), wherein the first set of VMs comprises a first portion of the VM set, performing a criticality analysis on the first set of VMs using a machine learning algorithm and a trained machine learning model to obtain an ordered list of critical VMs, performing a non-critical VM prioritization on a second set of VMs using the machine learning algorithm and the trained machine learning model to obtain an ordered list of non-critical VMs, consolidating the ordered list of critical VMs and the ordered list of non-critical VMs to obtain a final list, and initiating a backup of virtual machines using the final list.

First claim

What is claimed is:

1 . A method for performing backup operations, the method comprising:
obtaining, by a backup agent, a backup request, wherein the backup request specifies a virtual machine (VM) set; and in response to the backup request:
identifying a first set of virtual machines (VMs), wherein the first set of VMs comprises a first portion of the VM set;
performing a critical VM prioritization on the first set of VMs using a machine learning algorithm and a trained machine learning model to obtain an ordered list of critical VMs;
performing a non-critical VM prioritization on a second set of VMs using the machine learning algorithm and the trained machine learning model to obtain an ordered list of non-critical VMs, wherein the second set comprises a second portion of the VM set;
consolidating the ordered list of critical VMs and the ordered list of non-critical VMs to obtain a final list; and
initiating a backup of virtual machines using the final list.

2 . The method of claim 1 , further comprising:
identifying at least one virtual machine (VM) characteristic; obtaining VM data; obtaining training data using the at least one VM characteristic and the VM data; and generating the trained machine learning model using the machine learning algorithm and the training data.

3 . The method of claim 2 , wherein the machine learning algorithm is a multi-linear regression model.

4 . The method of claim 2 , wherein the training data is a subset of the VM data.

5 . The method of claim 1 , wherein performing the non-critical VM prioritization on the second set of VMs using the machine learning algorithm to obtain the ordered list of non-critical VMs comprises:
identifying a portion of the second set of VMs that specifies VMs in an ineligible state; removing the portion from the second set of VMs to obtain a third set of VMs; obtaining a criticality for each VM of the third set of VMs; and ordering, using the criticality, each VM of the third set of VMs to generate the ordered list of non-critical VMs.

6 . The method of claim 5 , wherein the ineligible state is at least one of: an orphaned state, a stale state, and an off state.

7 . The method of claim 1 , wherein initiating the backup of virtual machines using the final list comprises sending a second request to at least one production agent executing on a production host hosting a plurality of virtual machines, wherein the second request specifies an order of the plurality of virtual machines based on the final list.

8 . A system, comprising:
a processor; and memory comprising instructions which, when executed by the processor, perform a method, the method comprising:
obtaining, by a backup agent, a backup request, wherein the backup request specifies a virtual machine (VM) set; and
in response to the backup request:
identifying a first set of virtual machines (VMs), wherein the first set of VMs comprises a first portion of the VM set;
performing a critical VM prioritization on the first set of VMs using machine learning algorithm and a trained machine learning model to obtain an ordered list of critical VMs;
performing a non-critical VM prioritization on a second set of VMs using the machine learning algorithm and the trained machine learning model to obtain an ordered list of non-critical VMs, wherein the second set comprises a second portion of the VM set;
consolidating the ordered list of critical VMs and the ordered list of non-critical VMs to obtain a final list; and
initiating a backup of virtual machines using the final list.

9 . The system of claim 8 , the method further comprising:
identifying at least one virtual machine (VM) characteristic; obtaining VM data; obtaining training data using the at least one VM characteristic and the VM data; and generating the trained machine learning model using the machine learning algorithm and the training data.

10 . The system of claim 9 , wherein the machine learning algorithm is a multi-linear regression model.

11 . The system of claim 9 , wherein the training data is a subset of the VM data.

12 . The system of claim 8 , wherein performing the non-critical VM prioritization on the second set of VMs using the machine learning algorithm to obtain an ordered list of non-critical VMs comprises:
identifying a portion of the second set of VMs that specifies VMs in an ineligible state; removing the portion from the second set of VMs to obtain a third set of VMs; obtaining a criticality for each VM of the third set of VMs; and ordering, using the criticality, each VM of the third set of VMs to generate the ordered list of non-critical VMs.

13 . The system of claim 8 , wherein initiating the backup of virtual machines using the final list comprises sending a second request to at least one production agent executing on a production host hosting a plurality of virtual machines, wherein the second request specifies an order of the plurality of virtual machines based on the final list.

14 . A non-transitory computer readable medium comprising computer readable program code, which when executed by a computer processor enables the computer processor to perform a method for performing a backup operation, the method comprising:
obtaining, by a backup agent, a backup request, wherein the backup request specifies a virtual machine (VM) set; and in response to the backup request:
identifying a first set of virtual machines (VMs), wherein the first set of VMs comprises a first portion of the VM set;
performing a critical VM prioritization on the first set of VMs using a machine learning algorithm and a trained machine learning model to obtain an ordered list of critical VMs;
performing a non-critical VM prioritization on a second set of VMs using the machine learning algorithm and the trained machine learning model to obtain an ordered list of non-critical VMs, wherein the second set comprises a second portion of the VM set;
consolidating the ordered list of critical VMs and the ordered list of non-critical VMs to obtain a final list; and
initiating a backup of virtual machines using the final list.

15 . The non-transitory computer readable medium of claim 14 , the method further comprising:
identifying at least one virtual machine (VM) characteristic; obtaining VM data; obtaining training data using the at least one VM characteristic and the VM data; and training a machine learning algorithm using the training data to obtain the trained machine learning model.

16 . The non-transitory computer readable medium of claim 15 , wherein the machine learning algorithm is a multi-linear regression model.

17 . The non-transitory computer readable medium of claim 15 , wherein the training data is a subset of the VM data.

18 . The non-transitory computer readable medium of claim 14 , wherein performing the non-critical VM prioritization on the second set of VMs using the machine learning algorithm to obtain an ordered list of non-critical VMs comprises:
identifying a portion of the second set of VMs that specifies VMs in an ineligible state; removing the portion from the second set of VMs to obtain a third set of VMs; obtaining a criticality for each VM of the third set of VMs; and ordering, using the criticality, each VM of the third set of VMs to generate the ordered list of non-critical VMs.

19 . The non-transitory computer readable medium of claim 18 , wherein the ineligible state is at least one of: an orphaned state, a stale state, and an off state.

Assignees

• Emc Ip Holding Co Llc

Inventors

• Kumar Avinash
• Appireddygari Venkataramana Mahesh Reddy
• Chockalingam Elango

Classifications

• G06F18/214

  Generating training patterns; Bootstrap methods, e.g. bagging or boosting · CPC title

• G06N20/00

  Machine learning · CPC title

• G06F9/45533

  Hypervisors; Virtual machine monitors · CPC title

• G06F11/1461 Primary

  Backup scheduling policy · CPC title

• G06F2201/81

  Threshold · CPC title