Intelligent scheduling of backups

1. A method of reducing VM stunting during backup of a set of virtual machines, the method comprising:

running an analytic process to learn resource utilization patterns of a hypervisor system monitoring the set of virtual machines;

determining an opportunistic window of reduced resource utilization based on the resource utilization patterns using a machine learning model, the machine learning model self-validating predictions of reduced resource utilization;

implementing a predefined maintenance service schedule including a backup frequency of the set of virtual machines, the backup frequency defined independently of a resource utilization among the resource utilization patterns or the determined opportunistic window; and

monitoring and establishing compliance with the predefined maintenance service schedule while simultaneously scheduling a backup for the set of virtual machines during the opportunistic window, wherein scheduling backup for the set of virtual machines further comprises recovering and rescheduling backup jobs affected by a node failure for re-execution based on a resource utilization pattern of the hypervisor system.

2. The method of claim 1 , wherein the resource utilization patterns identify cyclical gaps in the resource utilization to identify and predict the opportunistic window of reduced resource utilization.

3. The method of claim 2 , wherein incremental snapshots of the set of virtual machines are scheduled to be taken during the predicted opportunistic window of reduced resource utilization.

4. The method of claim 1 , wherein the resource utilization patterns identify cyclical gaps in the resource utilization to identify and predict a blackout window of high resource utilization.

5. The method of claim 4 , further comprising deferring a scheduled backup past a predicted blackout window and meeting a maintenance service schedule.

6. The method of claim 1 , further comprising causing the machine learning model to self-validate its predictions.

7. A non-transitory machine-readable medium including instructions which, when read by a machine, cause the machine to perform operations including, at least:

running an analytic process to learn resource utilization patterns of a hypervisor system monitoring a set of virtual machines;

determining an opportunistic window of reduced resource utilization based on the resource utilization patterns using a machine learning model, the machine learning model self-validating predictions of reduced resource utilization;

implementing a predefined maintenance service schedule including a backup frequency of the set of virtual machines, the backup frequency defined independently of a resource utilization among the resource utilization patterns or the determined opportunistic window;

monitoring and establishing compliance with the predefined maintenance service schedule while simultaneously

scheduling a backup for the set of virtual machines during the opportunistic window, wherein scheduling backup for the set of virtual machines further comprises recovering and rescheduling backup jobs affected by a node failure for re-execution based on a resource utilization pattern of the hypervisor system.

8. The medium of claim 7 , wherein the resource utilization patterns identify cyclical gaps in the resource utilization to identify and predict the opportunistic window of reduced resource utilization.

9. The medium of claim 8 , wherein incremental snapshots of the set of virtual machines are scheduled to be taken during the predicted opportunistic window of reduced resource utilization.

10. The medium of claim 7 , wherein the resource utilization patterns identify cyclical gaps in the resource utilization to identify and predict a blackout window of high resource utilization.

11. The medium of claim 10 , further comprising deferring a scheduled backup past a predicted blackout window and meeting a maintenance service schedule.

12. The medium of claim 7 , wherein the operations further comprise causing the machine learning model to self-validate its predictions.

13. A system for reducing VM stunting during backup of a set of virtual machines, the system comprising:

processors; and

a memory storing instructions that, when executed by at least one processor among the processors, cause the system to perform operations comprising, at least:

running an analytic process to learn resource utilization patterns of a hypervisor system monitoring the set of virtual machines;

determining an opportunistic window of reduced resource utilization based on the resource utilization patterns using a machine learning model, the machine learning model self-validating predictions of reduced resource utilization;

implementing a predefined maintenance service schedule including a backup frequency of the set of virtual machines, the backup frequency defined independently of a resource utilization among the resource utilization patterns or the determined opportunistic window;

monitoring and establishing compliance with the predefined maintenance service schedule while simultaneously

scheduling a backup for the set of virtual machines during the opportunistic window, wherein scheduling backup for the set of virtual machines further comprises recovering and rescheduling backup jobs affected by a node failure for re-execution based on a resource utilization pattern of the hypervisor system.

14. The system of claim 13 , wherein the resource utilization patterns identify cyclical gaps in the resource utilization to identify and predict the opportunistic window of reduced resource utilization.

15. The system of claim 14 , wherein incremental snapshots of the set of virtual machines are scheduled to be taken during the predicted opportunistic window of reduced resource utilization.

16. The system of claim 13 , wherein the resource utilization patterns identify cyclical gaps in the resource utilization to identify and predict a blackout window of high resource utilization.

17. The system of claim 16 , further comprising deferring a scheduled backup past a predicted blackout window and meeting a maintenance service schedule.

18. The system of claim 13 , wherein the operations further comprise causing the machine learning model to self-validate its predictions.