Prioritizing tasks from virtual machines

• US 8,424,007 B1
• Filed: 09/30/2008
• Issued: 04/16/2013
• Est. Priority Date: 09/30/2008
• Status: Expired due to Fees

• Current Assignee: Veritas Technologies, LLC (Whitehouse Group Ltd.)
• Original Assignee: Symantec Corporation (NortonLifeLock Inc.)
• Inventors: Hernacki, Brian, Satish, Sourabh, Sobel, William E.
• Primary Examiner(s): Chan, Eddie P
• Assistant Examiner(s): LINDLOF, JOHN M
• Application Number:
  US12/242,685
• Time in Patent Office: 1,659 Days
• Field of Search: None
• US Class Current: 718/103
• CPC Class Codes: G06F 2209/503 Resource availability
• CPC Class Codes: G06F 9/485 Task life-cycle, e.g. stopping, restarting, resuming execution G06F9/4881 takes precedence
• CPC Class Codes: G06F 9/4881 Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues

Abstract

A computer-implemented method for prioritizing virtual machine tasks may include receiving a request to perform a first task from a virtual machine. The request may include information relevant to determining a priority of the task. The method may include determining the priority of the task based on the information. The method may further include scheduling the first task based on the priority of the task. The method may include selecting the first task for execution based on the scheduling. The method may include notifying the virtual machine that the first task has been selected for execution. Various related methods, computer-readable media, and systems are also disclosed.

Claims

1. A computer-implemented method for prioritizing tasks from virtual machines, the computer-implemented method comprising:

receiving, from a task scheduling agent within a first virtual machine, a request to perform a first task of an application, wherein;

the request comprises priority information,

the request is received at a scheduling module of a hypervisor that manages the first virtual machine,

the priority information comprises an identifier of a resource needed to execute the first task,

and the task scheduling agent sends the request to the scheduling module before the first task is sent to the hypervisor for execution;

determining a priority of the first task based on the priority information by predicting future availability of the resource, wherein predicting the future availability of the resource comprises tracking use of the resource to identify a performance metric of the resource;

scheduling the first task based on the priority of the first task by evaluating the performance metric of the resource;

selecting, based on the scheduling, the first task for execution;

notifying the task scheduling agent within the first virtual machine that the first task has been selected for execution.

2. The computer-implemented method of claim 1, further comprising:

determining that the first task is being executed;

determining that a second task from a second virtual machine has priority over the first task;

querying the task scheduling agent within the first virtual machine to determine whether the first task can be suspended;

determining, based on the response from the task scheduling agent within the first virtual machine, that the first task can be suspended;

requesting that the first virtual machine suspend the first task;

receiving, from the task scheduling agent, notification of suspension of the first task;

notifying the second virtual machine that the second task has been selected for execution.

3. The computer-implemented method of claim 1, further comprising:

requesting additional priority information about the first task;

receiving, from the task scheduling agent, information indicating a resource-access category of the first task, wherein the resource-access category of the first task distinguishes the first task from other categories of tasks based on one or more types of resources that are required to perform the first task.

4. The computer-implemented method of claim 1, wherein:

the task scheduling agent is programmed to wait for a response from the scheduling module before allowing the application to execute the first task;

the task scheduling agent is programmed to allow the first task to be sent to the hypervisor for execution after being notified that the first task has been selected for execution.

5. The computer-implemented method of claim 1, wherein the task scheduling agent is programmed to identify the first task by monitoring an operating system scheduler of an operating system of the first virtual machine.

6. The computer-implemented method of claim 1, wherein:

the resource comprises a storage device;

the performance metric of the resource comprises a read time for the storage device and/or a write time for the storage device.

7. The computer-implemented method of claim 1, wherein:

the resource comprises a network interface;

the performance metric of the resource comprises a bandwidth of the network interface.

8. The computer-implemented method of claim 1, wherein predicting the future availability of the resource is based on prior usage patterns of the resource.

9. The computer-implemented method of claim 1, wherein scheduling the first task comprises:

reevaluating priorities of one or more other tasks in a task queue to determine where the first task should be scheduled in the task queue.

10. The computer-implemented method of claim 1, wherein determining the priority of the first task further comprises:

determining that a security threat level is high;

determining, based on the priority information, that the first task is security related;

scheduling the first task comprises giving the first task priority over one or more other tasks based on the security threat level being high and the first task being security related.

11. A computer-implemented method for prioritizing virtual machine tasks, the computer-implemented method comprising:

identifying a first task within a virtual machine that is managed by a hypervisor, the identifying being performed by a task scheduling agent within the virtual machine;

sending, from the task scheduling agent to the hypervisor, a request to perform the first task, wherein the request comprises priority information and the request is sent by the task scheduling agent before the first task is sent to the hypervisor for execution;

determining, at a scheduling module in the hypervisor, that a security threat level is high, wherein the security threat level indicates a security threat relevant to a computing system that comprises the hypervisor and the virtual machine;

determining, at the scheduling module and based on the priority information, that the first task is security related;

scheduling, at the scheduling module, the first task comprises giving the first task priority over one or more other tasks based on the security threat level being high and the first task being security related;

receiving, at the task scheduling agent, a notification that the first task has been selected for execution.

12. The computer-implemented method of claim 11, wherein identifying the first task comprises:

hooking a process of an operating system of the virtual machine to intercept a request from an application to perform the first task.

13. The computer-implemented method of claim 11, wherein:

receiving the notification comprises receiving scheduling information about the first task from the scheduling module;

the computer-implemented method further comprises using, at the task scheduling agent, the scheduling information to determine when to allow the first task to be sent from the virtual machine to the hypervisor for execution.

14. The computer-implemented method of claim 11, further comprising:

waiting for the notification before allowing an application to execute the first task;

allowing the first task to be sent to the hypervisor for execution after being notified that the first task has been selected for execution, wherein the waiting and allowing are performed by the task scheduling agent.

15. The computer-implemented method of claim 11, further comprising:

sending, with the request to perform the first task, an instruction for executing the first task such that the first task may be executed before responding to the request from the task scheduling agent.

16. A computer-implemented method for prioritizing tasks from virtual machines, the computer-implemented method comprising:

receiving, from a task scheduling agent within a virtual machine, a request to perform a task, wherein the request comprises priority information, the request is received at a scheduling module of a hypervisor that manages the virtual machine, the priority information comprises an identifier of a resource needed to execute the task, and the task scheduling agent sends the request to the scheduling module before the task is sent to the hypervisor for execution;

determining a priority of the task based on the priority information by predicting future availability of the resource, wherein predicting the future availability of the resource comprises tracking use of the resource to identify a performance metric of the resource;

scheduling the task based on the priority of the task by evaluating the performance metric of the resource.

17. The computer-implemented method of claim 16, further comprising:

selecting, based on the scheduling, the task for execution;

executing an instruction programmed to perform the task, wherein the request comprises the instruction.

18. The computer-implemented method of claim 17, further comprising:

creating the request by including the priority information with the instruction.

19. The computer-implemented method of claim 16, further comprising:

sending scheduling information associated with scheduling the task to the task scheduling agent, wherein the scheduling information provides the task scheduling agent with system resource use information and the task scheduling agent uses the system resource use information to determine when to send the task to the hypervisor for execution.

20. The computer-implemented method of claim 16, wherein:

the identifier of the resource indicates that the task needs to use at least one of;

a CPU resource,

a network resource;

determining the priority of the task comprises determining availability of the at least one of;

the CPU resource,

the network resource.