Dynamic resource configuration based on context

• US 10,191,772 B2
• Filed: 01/29/2018
• Issued: 01/29/2019
• Est. Priority Date: 07/25/2012
• Status: Active Grant

• Current Assignee: Vmware LLC (Broadcom, Inc.)
• Original Assignee: VMware, Inc. (Broadcom, Inc.)
• Inventors: Beveridge, Daniel James, Agrawal, Banit
• Primary Examiner(s): Truong, Lechi
• Application Number:
  US15/883,000
• Publication Number:
  US 20180165130A1
• Time in Patent Office: 365 Days
• Field of Search: 719310, 718 1
• CPC Class Codes: G06F 2009/4557 Distribution of virtual machine instances; Migration and load balancing
• CPC Class Codes: G06F 2009/45591 Monitoring or debugging support
• CPC Class Codes: G06F 9/45558 Hypervisor-specific management and integration aspects
• CPC Class Codes: G06F 9/5005 to service a request
• CPC Class Codes: G06F 9/5011 the resources being hardware resources other than CPUs, Servers and Terminals
• CPC Class Codes: G06F 9/542 Event management; Broadcasting; Multicasting; Notifications
• CPC Class Codes: H04L 43/08 Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
• CPC Class Codes: H04L 43/20 the monitoring system or the monitored elements being virtualised, abstracted or software-defined entities, e.g. SDN or NFV
• CPC Class Codes: H04L 47/783 Distributed allocation of resources, e.g. bandwidth brokers
• CPC Class Codes: H04L 67/535 Tracking the activity of the user network monitoring arrangements H04L43/00; recording of computer activity G06F11/34

Abstract

Aspects of the disclosure allocate shares of processing resources or other physical resources among virtual machines (VMs) operating as, for example, virtual desktops on a plurality of host computing devices. Allocations of resources are adjusted based on the user activity, VM activity, and/or application activity detected by an agent executing on each VM. Allocated shares may be boosted, unboosted, or normalized, depending on the type and duration of detected activity, by a resource allocation manager executing on a management server.

Claims

1. A system for resource allocation in a virtualized computing environment, comprising at least one physical computing system hosting a plurality of virtual machines (VMs), the system comprising:

a memory;

a processor;

a plurality of agents, each associated with at least one of the plurality of VMs, that, when executed by the processor, collect activity information about the associated VM;

a module that is stored in the memory and, when executed by the processor, dynamically allocates physical computing resources of the at least one physical computing system amongst the plurality of VMs, each of the plurality of VMs configured to host a virtual desktop, a virtual desktop management module facilitating remote access to the virtual desktop by a user, the dynamic allocation performed by;

monitoring, by the plurality of agents, a state of the virtual desktop of each associated VM to detect activities forming at least a portion of the activity information, the activity information including the user accessing one or more user interfaces on the virtual desktop, a running task on the associated VM, or both;

comparing the collected activity information to a policy;

in response to the comparing, automatically modifying, based on the policy, an allocation of the physical computing resources of the at least one physical computing system to at least one VM of the plurality of VMs, wherein modifying the allocation of the physical computing resources to the at least one VM comprises;

upon detecting, based on the monitoring, that the at least one VM is idle, suspending the at least one VM; and

upon detecting, based on the monitoring, any new activity of the at least one VM, resuming the at least one VM.

2. The system of claim 1, further comprising an appliance that is stored on the physical computing system and that is configured, when executed by the processor, to request a change to physical resource allocation for the VMs executing on the physical computing device based on the collected activity information.

3. The system of claim 1, wherein the activity information further comprises information about at least one of user activity, machine activity, or application activity.

4. The system of claim 1, wherein the physical computing resources further comprise at least one of storage priority, graphics protocols, or CPU resources.

5. The system of claim 1, wherein the policy is federated by a central console to a resource allocation manager executing on the physical computing system.

6. The system of claim 1, wherein the policy is established by an administrator, in accordance with at least one of a terms of service agreement, a service level agreement (SLA), a quality of service (QoS) agreement, or a contract.

7. The system of claim 1, wherein modifying the allocation of the physical computing resources to the at least one VM further comprises:

on resuming the at least one VM, determining if the new activity comprises a whitelist operation or a blacklist operation;

on determining that the new activity comprises the whitelist operation, elevating a resource share of the at least one VM; and

on determining that the new activity comprises the blacklist operation, reducing the resource share of the at least one VM.

8. The system of claim 1, wherein upon determining that the monitored activities have ended, normalizing the resources allocated to the at least one VM.

9. The system of claim 1, wherein modifying the allocation of the physical computing resources to the at least one VM further comprises:

upon suspending the at least one VM, allocating physical computing resources freed up by the suspension of the at least on VM to other VMs of the plurality of VMs.

10. A method for resource allocation in a virtualized computing environment comprising at least one physical computing system hosting a plurality of VMs, the method comprising:

dynamically allocating physical computing resources of the at least one physical computing system amongst the plurality of VMs, each of the plurality of VMs configured to host a virtual desktop, a virtual desktop management module facilitating remote access by a user, the dynamic allocation including;

monitoring a state of the virtual desktop of each associated VM to detect activities forming at least a portion of activity information of the each associated VM, the activity information including the user accessing one ore more user interfaces on the virtual desktop, a running task on the associated VM, or both;

comparing the collected activity information to a policy;

in response to the comparing, automatically modifying, based on the policy, an allocation of the physical computing resources of the at least one physical computing system to at least one VM of the plurality of VMs, wherein modifying the allocation of the physical computing resources to the at least one VM comprises;

upon detecting, based on the monitoring, that the at least one VM is idle, suspending the at least one VM; and

upon detecting, based on the monitoring, any new activity of the at least one VM, resuming the at least one VM.

11. The method of claim 10, wherein modifying the allocation of the physical computing resources to the at least one VM further comprises:

upon suspending the at least one VM, allocating physical computing resources freed up by the suspension of the at least one VM to other VMs of the plurality of VMs.

12. The method of claim 10, wherein the activity information further comprises information about at least one of user activity, machine activity, or application activity.

13. The method of claim 10, wherein the physical computing resources further comprise at least one of storage priority, graphics protocols, or CPU resources.

14. The method of claim 10, wherein the policy is federated by a central console to a resource allocation manager executing on the physical computing system.

15. The method of claim 10, wherein the policy is established by an administrator, in accordance with at least one of a terms of service agreement, a service level agreement (SLA), a quality of service (QoS) agreement, or a contract.

16. The method of claim 10, further comprising:

collecting performance metrics of applications executing on the plurality of VMs; and

dynamically adjusting the policy based on the collected performance metrics.

17. A non-transitory computer-readable medium including computer-executable instructions that, when executed by a processor, allocate resources in a virtualized computing environment comprising at least one physical computing system hosting a plurality of VMs, the computer-executable instructions comprising:

dynamically allocating physical computing resources of the at least one physical computing system amongst the plurality of VMs, each of the plurality of VMs configured to host a virtual desktop, a virtual desktop management module facilitating remote access by a user, the dynamic allocation including;

receiving reports of detected activities of the user and a VM of the user, the activities including the user accessing one or more user interfaces on the virtual desktop, a running task on the VM of the user, or both;

comparing the detected activities to a policy;

in response to the comparing, automatically modifying, based on the policy, an allocation of the physical computing resources of the at least one physical computing system to the plurality of VMs, wherein modifying the allocation of the physical computing resources further comprises;

upon determining, based on the received reports, that the user or the VM of the user is idle, suspending the VM of the user; and

upon determining, based on the received reports, any new activity of the user or the VM of the user, resuming the VM of the user.

18. The non-transitory computer-readable medium of claim 17, wherein modifying the allocation of the physical computing resources further comprises:

upon suspending the VM of the user, allocating physical computing resources freed up by the suspension of the VM of the user to other VMs of the plurality of VMs.

19. The non-transitory computer-readable medium of claim 17, wherein the physical computing resources further comprise at least one of storage priority, graphics protocols, or CPU resources.

20. The non-transitory computer-readable medium of claim 17, wherein the policy is federated by a central console to a resource allocation manager executing on the physical computing system.