Automated dynamic resource configuration for hard provisioned virtual machines

Patent metadata

Field	Value
Publication number	US-12327145-B2
Application number	US-202217725183-A
Country	US
Kind code	B2
Filing date	Apr 20, 2022
Priority date	Apr 20, 2022
Publication date	Jun 10, 2025
Grant date	Jun 10, 2025

Abstract

Methods and systems for managing provisioning of virtual machines. Virtual machines may host applications that may provide computer implemented services. Various hardware resources may be allocated to the virtual machines via a hypervisor. As the workloads of the applications change, the virtual machines may become over or under provisioned. To manage provisioning of virtual machines, various types of resource consumption estimates may be obtained. The resource consumption estimates may be used to ascertain how to provision various virtual machines to reduce or eliminate inefficient allocations of hardware resources for use by the virtual machines.

First claim

What is claimed is:

1. A computer-implemented method for managing resources allocated to virtual machines, the method comprising:
obtaining a resource consumption estimate for a hard provisioned virtual machine of the virtual machines, the resource consumption estimate being calculated based on types of workloads being executed by the hard provisioned virtual machine, and at least one type of the types of the workloads is ignored during a calculation of the resource consumption estimate;
identifying:
a resource inefficiency for the hard provisioned virtual machine based on the resource consumption estimate, and
a new resource allocation based on the resource inefficiency; and
provisioning, to remediate the identified resource inefficiency of the hard provisioned virtual machine, the hard provisioned virtual machine with the new resource allocation, the new resource allocation comprising computer hardware resources of a data processing system on which the hard provisioned virtual machine is hosted,
wherein the virtual machines are hosted by one or more host data processing systems as a first type of virtual machines and a second type of virtual machines, the data processing system being one of the one or more host data processing systems, and the first type of virtual machines comprise one or more soft provisioned virtual machines that utilize, at least in part, shared resources of the one or more host data processing systems among one another in a time-slicing manner where each of the one or more soft provisioned virtual machines is granted use of the shared resources at different times within a predetermined period of time, and the second type of virtual machines comprise the hard provisioned virtual machine that exclusively uses a portion of resources of the one or more host data processing systems without having to share the portion of the resource in the time-slicing manner with the one or more soft provisioned virtual machines.

2. The computer-implemented method of claim 1 , wherein identifying the resource inefficiency comprises:
comparing the resource consumption estimate to an existing resource allocation for the hard provision virtual machine to identify a resource delta;
making a determination that the resource delta exceeds a threshold; and
identifying the new resource allocation based on the determination.

3. The computer-implemented method of claim 2 , wherein the resource consumption estimate comprises a processing estimate and a memory estimate, and the existing resource allocation include a quantity of processing resources and a quantity of memory resources.

4. The computer-implemented method of claim 3 , wherein comparing the resource consumption estimate to the existing resource allocation comprises:
making a first sub-comparison between the processing estimate and the quantity of processing resources to identify a processing resource delta;
making a second sub-comparison between the memory estimate and the quantity of memory resources to identify a memory resource delta; and
aggregating the processing resource delta and the memory resource delta to obtain the resource delta.

5. The computer-implemented method of claim 4 , wherein identifying the new resource allocation comprises:
incrementing one of the quantity of processing resources and the quantity of the memory resources to obtain a provisional resource allocation;
identifying a quantity of allocable resources based, at least in part, on a reserved quantity of resources for dynamic allocation;
making a second determination, based on the allocable resources, that the provisional resource allocation will not over provision a host data processing system of the one or more host data processing systems for the virtual machines; and
promoting the provisional resource allocation to the new resource allocation.

6. The computer-implemented method of claim 1 , where obtaining the resource consumption estimate for the hard provisioned virtual machine of the virtual machines comprises:
obtaining resource consumption samples for the hard provisioned virtual machines until a threshold number of samples are obtained, the threshold number of samples being based on a duration of time and a sampling rate of the resource consumption samples; and
calculating an average of the resource consumption samples to obtain the resource consumption estimate.

7. The method of claim 1 , wherein provisioning the hard provisioned virtual machine with the new resource allocation comprises:
provisioning the computer hardware resources of the data processing system for exclusive use by the hard provisioned virtual machine based on the new resource allocation.

8. The method of claim 1 , wherein provisioning the hard provisioned virtual machine with the new resource allocation comprises:
scheduling provisioning of the computer hardware resources of the data processing system for exclusive use by the hard provisioned virtual machine based on the new resource allocation.

9. The method of claim 1 , wherein provisioning the hard provisioned virtual machine with the new resource allocation comprises:
modifying the computer hardware resources of the data processing system for exclusive use by the hard provisioned virtual machine while the hard provisioned virtual machine continues execution.

10. The computer-implemented method of claim 1 , wherein the types of the workloads comprise at least a productive use workload and an unproductive use workload, wherein each of the workloads is classified as the productive use workload or the unproductive use workload based on a behavior of each of the workloads and a predetermined set of classification criteria.

11. A non-transitory machine-readable medium having instructions stored therein, which when executed by a processor, cause the processor to perform operations for managing resources allocated to virtual machines, the operations comprising:
obtaining a resource consumption estimate for a hard provisioned virtual machine of the virtual machines, the resource consumption estimate being calculated based on types of workloads being executed by the hard provisioned virtual machine, and at least one type of the types of workloads is ignored during a calculation of the resource consumption estimate;
identifying:
a resource inefficiency for the hard provisioned virtual machine based on the resource consumption estimate, and
a new resource allocation based on the resource inefficiency; and
provisioning, to remediate the identified resource inefficiency of the hard provisioned virtual machine, the hard provisioned virtual machine with the new resource allocation, the new resource allocation comprising computer hardware resources of a data processing system on which the hard provisioned virtual machine is hosted,
wherein the virtual machines are hosted by one or more host data processing systems as a first type of virtual machines and a second type of virtual machines, the data processing system being one of the one or more host data processing systems, and the first type of virtual machines comprise one or more soft provisioned virtual machines that utilize, at least in part, shared resources of the one or more host data processing systems among one another in a time-slicing manner where each of the one or more soft provisioned virtual machines is granted use of the shared resources at different times within a predetermined period of time, and the second type of virtual machines comprise the hard provisioned virtual machine that exclusively uses a portion of resources of the one or more host data processing systems without having to share the portion of the resource in the time-slicing manner with the on

Assignees

• Dell Products Lp

Inventors

• Caisse Joseph B
• Goodman Bradley K
• Tummala Himabindu

Classifications

• G06F9/45558 Primary

  Hypervisor-specific management and integration aspects · CPC title

• G06F9/5077 Primary

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

• G06F9/5022

  Mechanisms to release resources · CPC title

• G06F9/5005

  to service a request · CPC title

• G06F2009/45579

  I/O management, e.g. providing access to device drivers or storage · CPC title