Management of virtual machine resources in computing environments

Patent metadata

Field	Value
Publication number	US-10129106-B2
Application number	US-201414272968-A
Country	US
Kind code	B2
Filing date	May 8, 2014
Priority date	Apr 9, 2014
Publication date	Nov 13, 2018
Grant date	Nov 13, 2018

Abstract

Embodiments of the present disclosure call for a method, a system, and a computer program product for managing virtual resources for a virtual machine of interest on a physical server. The method includes monitoring a plurality of physical servers, on a computing system. Each physical server provides an initial level of virtual resources to a plurality of virtual machines. The method also includes receiving a requested level of virtual resources for a virtual machine of interest hosted by a first physical server from the plurality of physical servers. The method also includes determining whether the first physical server is able to host the requested level of virtual resources for the virtual machine of interest. The method also includes determining, in response to the first physical server not being able to host the requested level of resources for the virtual machine of interest, a path.

First claim

What is claimed is:

1. A system comprising:
a plurality of physical servers operating in a computing environment, a first physical server of the plurality of physical servers configured to provide virtual resources at an initial level to a plurality of virtual machines;
a cloud controller that manages virtual resources for the plurality of virtual machines, the cloud controller configured to:
monitor the plurality of physical servers, on a computing system, each of the plurality of physical servers provides an initial level of virtual resources to the plurality of virtual machines, wherein a first amount of virtual resources is allocated to a virtual machine of interest of the plurality of virtual machines, wherein the virtual resources include at least a first virtual resource;
receive a request for an increased level of virtual resources for the virtual machine of interest hosted by the first physical server from the plurality of physical servers;
determine that the first amount of virtual resources allocated to the virtual machine of interest exceeds the respective amounts of virtual resources allocated to each other virtual machine of the plurality of virtual machines;
determine whether the first physical server is unable to provide the requested level of virtual resources to the virtual machine of interest, in response to determining that the first amount of virtual resources exceeds the respective amounts of virtual resources allocated to each of the other virtual machines;
implement, in response to determining that the first physical server is not able to provide the requested level of resources for the virtual machine of interest and further in response to determining that a schedule of actions is within an action threshold, a path to accommodate the virtual machine of interest involving the increased level of virtual resources by executing the schedule of actions for a plurality of secondary virtual machines on the plurality of physical servers, the path implemented by:
predicting the schedule of actions that define the path;
determining whether the schedule of actions is within the action threshold, wherein the action threshold is a maximum number of scheduled actions that a path may include, wherein a scheduled action is one of a virtual machine migration and a virtual machine resizing;
resizing the plurality of secondary virtual machines to create a first set of unallocated virtual resources, wherein each of the resized plurality of secondary virtual machines is smaller than a corresponding version in the plurality of secondary virtual machines;
measuring unallocated virtual resources in the plurality of physical servers separate from the first physical server; and
migrating at least one of the resized plurality of secondary virtual machines from the first physical server to the plurality of physical servers with sufficient unallocated virtual resources to accommodate the at least one of the resized plurality of secondary virtual machines to create a second set of unallocated virtual resources on the first physical server to accommodate the virtual machine of interest;

implement the path for the secondary virtual machines on the plurality of physical servers; and
following the migration of the at least one resized plurality of secondary virtual machines from the first physical server to the plurality of physical servers, increasing the virtual resources to the requested level for the virtual machine of interest by allocating the first set of unallocated virtual resources, and at least a portion of the second set of unallocated virtual resources on the first physical server to the virtual machine of interest.

2. The system of claim 1 , wherein the cloud controller is further configured to:
increase, in response to the first physical server being able to host the requested level of resources for the virtual machine of interest, the virtual resources to the requested level for the virtual machine of interest on the first physical server, wherein migrating the plurality of virtual machines includes:
determining whether the virtual machine of interest is able to be migrated to a second physical server from the plurality of servers; and
migrating the virtual machine of interest to the second physical server in response to the virtual machine of interest being able to be migrated to the second physical server.

3. The system of claim 1 , wherein resizing the plurality of secondary virtual machines includes:
determining an operating threshold of the plurality of secondary virtual machines; and
resizing the plurality of secondary virtual machines to the operating threshold.

4. The system of claim 1 , wherein the at least one of the resized plurality of secondary virtual machines includes a second amount of virtual resources that is less than the first amount of virtual resources, the method further comprising:
selecting the at least one of the resized plurality of secondary virtual machines, wherein the selecting the at least one of the resized plurality of secondary virtual machines is based on a priority of the plurality of secondary virtual machines, wherein the selecting the at least one of the resized plurality of secondary virtual machines further comprises:
monitoring resource usages of each virtual machine of the plurality of secondary virtual machines on the first physical server, wherein the monitored resources include CPU processing, memory, network, system timing, and storage;
determining resource usages of each virtual machine of the plurality of secondary virtual machines relative to a total amount of computing resources of the first physical server; and
prioritizing each virtual machine of the plurality of secondary virtual machines based on resource usage, wherein virtual machines having lower resource usages are prioritized over virtual machines having higher resource usages.

5. The system of claim 1 , wherein migrating the at least one of the resized plurality of secondary virtual machines from the first physical server to the one of the plurality of physical servers is based in part on an availability score of the one of the plurality of physical servers that satisfies a predetermined threshold, wherein the migrating the at least one of the resized plurality of secondary virtual machines further comprises:
receiving one or more performance factors from the plurality of physical servers, wherein the performance factors include CPU utilization, memory utilization, disk bandwidth, and network utilization;
determining the availability score of each of the plurality of physical servers based on applying a weighting function to the one or more performance factors, wherein the weighting function depends in part on a system policy and properties of each physical server, wherein the availability score for a particular server is a numerical representation of whether the particular server can host the at least one of the resized plurality of secondary virtual machines, wherein the system policy includes determining a weight for each prioritizing resource based on resource utilization of the at least one of the resized plurality of secondary virtual machines;
prioritizing each physical server of the plurality of physical servers based on their respective availability scores; and
determining that the availability score for the one of the plurality of physical servers satisfies the predetermined threshold.

6. The system of claim 1 , wherein the cloud controller is further configured to:
determine an operating boundary of the at least one of the plurality of secondary virtual machines by:
determining a viable operating boundary for the at least one of the plurality of secondary virtual machines, the viable operating boundary being the amount of virtual resources

Assignees

• IBM

Inventors

• Cropper Joseph W
• Crowell Susan F
• Nikolai Jason A
• Thorstensen Andrew T

Classifications

• G06F2009/45591

  Monitoring or debugging support · CPC title

• H04L43/0817

  by checking functioning · CPC title

• H04L41/5054 Primary

  Automatic deployment of services triggered by the service manager, e.g. service implementation by automatic configuration of network components · CPC title

• H04L41/5022

  by giving priorities, e.g. assigning classes of service · CPC title

• G06F9/45558

  Hypervisor-specific management and integration aspects · CPC title