Minimizing impact of migrating virtual services

Patent metadata

Field	Value
Publication number	US-10977068-B2
Application number	US-201816160756-A
Country	US
Kind code	B2
Filing date	Oct 15, 2018
Priority date	Oct 15, 2018
Publication date	Apr 13, 2021
Grant date	Apr 13, 2021

Abstract

The present disclosure relates to systems, methods, and computer readable media that utilize a low-impact live-migration system to reduce unfavorable impacts caused as a result of live-migrating computing containers between physical server devices of a cloud computing system. For example, systems disclosed herein evaluates characteristics of computing containers on server devices to determine a predicted unfavorable impact of live-migrating the computing containers between the server devices. Based on the predicted impact, the systems disclosed herein can selectively identify which computing containers to live-migrate as well as carry out live-migration of the select computing containers in such a way the significantly reduces unfavorable impacts to a customer or client device associated with the computing containers.

First claim

What is claimed is:

1. A method for live-migrating virtual services between server nodes, comprising: identifying a plurality of virtual machines on a plurality of server devices; evaluating virtual machine characteristics of the plurality of virtual machines to determine a plurality of impact scores for the plurality of virtual machines, wherein the plurality of impact scores are associated with a predicted impact of live-migrating the plurality of virtual machines; determining a threshold impact score based on a state of available computing resources of the plurality of server devices, the state of available computing resources including a number of empty server devices from the plurality of server devices; determining whether the number of empty server devices is greater than or equal to a minimum threshold number of empty server devices for the plurality of server devices, wherein the threshold impact score is based on whether the number of empty server devices is greater than or equal to the minimum threshold number of empty server devices for the plurality of server devices; identifying a subset of virtual machines from the plurality of virtual machines as candidate virtual machines for live-migration based on impact scores for the subset of virtual machines being less than the determined threshold impact score; and selectively initiating live-migration of the subset of virtual machines from the plurality of virtual machines to one or more destination server devices.

2. The method of claim 1 , wherein identifying the subset of virtual machines for live-migration comprises identifying candidate virtual machines based on a determination that the impact score corresponding to the candidate virtual machines is less than an additional impact score corresponding to additional virtual machines from the plurality of virtual machines not included in the subset of virtual machines.

3. The method of claim 1 , wherein evaluating the virtual machine characteristics of the plurality of virtual machines to determine the plurality of impact scores comprises applying a prediction engine to the plurality of virtual machines, wherein the prediction engine is trained to determine impact scores for virtual machines based on associated virtual machine characteristics.

4. The method of claim 3 , wherein applying the prediction engine to the plurality of virtual machines comprises:
applying a blackout prediction engine trained to predict an estimated blackout time for migrating a given virtual machine based on a set of virtual machine characteristics for the given virtual machine, the estimated blackout time comprising an estimated duration of time that the given virtual machine will not execute any codes or instructions or that a client device will be disconnected from accessing the given virtual machine; and
determining the plurality of impact scores based on predicted blackout times for the plurality of virtual machines.

5. The method of claim 3 , wherein applying the prediction engine to the plurality of virtual machines comprises:
applying a brownout prediction engine trained to predict an estimated brownout time for migrating a given data container based on a set of virtual machine characteristics for the given virtual machine, the estimated brownout time comprising an estimated duration of time that the given data container will provide limited performance; and
determining the plurality of impact scores based on predicted brownout times for the plurality of virtual machines.

6. The method of claim 3 , wherein applying the prediction engine to the plurality of virtual machines comprises:
determining impact sensitivity for the plurality of virtual machines, wherein impact sensitivity for a given virtual machine comprises an indication of sensitivity of the given virtual machine to an interruption of a connection between the client device and the given virtual machine; and
determining the plurality of impact scores based on the impact sensitivity for the plurality of virtual machines.

7. The method of claim 3 , wherein the prediction engine is trained to predict a combined impact score associated with live-migrating a given virtual machine by:
determining two or more impact sub-scores for the given virtual machine, the two or more impact sub-scores comprising two or more of:
a first predicted impact sub-score associated with a predicted blackout time, the predicted blackout time comprising an estimated duration of time that the given virtual machine will not execute any codes or instructions or that a client device will be disconnected from accessing the given virtual machine;
a second predicted impact sub-score associated with a predicted brownout time, the predicted brownout time comprising an estimated duration of time that the given data container will provide limited performance;
a third predicted impact sub-score associated with an impact sensitivity, the impact sensitivity comprising an indication of how sensitive the given virtual machine is to an interruption of a connection between the client device and the given virtual machine; and
a fourth predicted impact sub-score associated with a lifetime of the given virtual machine, the lifetime of the given virtual machine comprising an indication of when the given virtual machine is expected to expire; and
combining the two or more impact sub-scores to determine the combined impact score associated with live-migrating the given virtual machine.

8. A system, comprising: one or more processors; memory in electronic communication with the one or more processors; instructions stored in the memory, the instructions being executable by the one or more processors to: identify a plurality of server devices including a plurality of virtual machines; evaluate virtual machine characteristics of the plurality of virtual machines to determine a plurality of impact scores associated with live-migrating the plurality of virtual machines, wherein an impact score of a virtual machine is associated with a predicted impact of live-migrating the virtual machine; determine a threshold impact score based on a state of available computing resources of the plurality of server devices, the state of available computing resources including a number of empty server devices from the plurality of server devices; determine whether the number of empty server devices is greater than or equal to a minimum threshold number of empty server devices for the plurality of server devices, wherein the threshold impact score is based on whether the number of empty server devices is greater than or equal to the minimum threshold number of empty server devices for the plurality of server devices; identify a candidate server device from the plurality of server devices based on one or more impact scores for one or more virtual machines from the plurality of virtual machines on the candidate server being less than the determined threshold impact score; and initiate live-migration of the one or more virtual machines from the candidate server device to a destination server device.

9. The system of claim 8 , wherein identifying the candidate server device comprises:
determining a combined impact score for the candidate server device based on a combination of the one or more impact scores for the one or more virtual machines on the candidate server; and
wherein identifying the candidate server device is further based on a determination that the combined impact score for the candidate server device is less than one or more combined impact scores for one or more additional server devices from the plurality of server devices.

10. The system of claim 8 , wherein evaluating the virtual machine characteristics of the plurality of virt

Assignees

• Microsoft Technology Licensing Llc

Inventors

• Cortez Eli
• Mani Ajay
• Fontoura Marcus Felipe
• Sheth Nisarg Tarakkumar
• Moscibroda Thomas
• Constantin Ana-Maria

Classifications

• G06F2009/4557

  Distribution of virtual machine instances; Migration and load balancing · CPC title

• G06N5/02

  Knowledge representation; Symbolic representation · CPC title

• G06F9/5077

  Logical partitioning of resources; Management or configuration of virtualized resources (specific details on emulation or internal functioning of virtual machines G06F9/455) · CPC title

• G06F9/45558 Primary

  Hypervisor-specific management and integration aspects · CPC title

• G06F9/5088 Primary

  involving task migration · CPC title