System and method for efficient allocation of resources in virtualized desktop environments

• US 8,234,236 B2
• Filed: 08/26/2009
• Issued: 07/31/2012
• Est. Priority Date: 06/01/2009
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Beaty, Kirk A., Bhowmik, Rajdeep, Kochut, Andrzej
• Primary Examiner(s): Starks, Wilbert L
• Application Number:
  US12/547,726
• Publication Number:
  US 20100306163A1
• Time in Patent Office: 1,070 Days
• Field of Search: 706/52, 706/45
• US Class Current: 706/52
• CPC Class Codes: G06F 11/301 where the computing system is a virtual computing platform, e.g. logically partitioned systems virtual machines G06F9/45533; logical partitioning of resources G06F9/5077
• CPC Class Codes: G06F 11/3093 Configuration details thereof, e.g. installation, enabling, spatial arrangement of the probes
• CPC Class Codes: G06F 11/3414 Workload generation, e.g. scripts, playback
• CPC Class Codes: G06F 11/3423 where the assessed time is active or idle time
• CPC Class Codes: G06F 11/3452 Performance evaluation by statistical analysis
• CPC Class Codes: G06F 2201/81 Threshold
• CPC Class Codes: G06F 2201/86 Event-based monitoring
• CPC Class Codes: G06F 2209/508 Monitor
• CPC Class Codes: G06F 9/5077 Logical partitioning of resources; Management or configuration of virtualized resources specific details on emulation or internal functioning of virtual machines G06F9/455
• CPC Class Codes: G06F 9/5088 involving task migration
• CPC Class Codes: G06N 5/00 Computing arrangements using knowledge-based models
• CPC Class Codes: Y02B 70/3225 Demand response systems, e.g. load shedding, peak shaving
• CPC Class Codes: Y04S 20/222 Demand response systems, e.g. load shedding, peak shaving

Abstract

A system and method for managing responsiveness of virtual desktops using passive monitoring includes monitoring interactive desktop events for a computer device and measuring a time between the interactive desktop event and a result of the interactive desktop event. Based on the time, desktop responsiveness is evaluated. System resources are reallocated to improve the responsiveness.

Claims

1. A method for managing responsiveness of virtual desktops using passive monitoring, comprising:

monitoring interactive desktop events for a computer device;

measuring a time between an interactive desktop event and a result of the interactive desktop event;

based on the time, evaluating desktop responsiveness; and

reallocating system resources to improve the responsiveness.

2. The method as recited in claim 1, wherein monitoring interactive desktop events includes monitoring at least one of mouse events, keyboard events and pixel images on a display.

3. The method as recited in claim 1, wherein the result of the interactive desktop event includes a response to the interactive desktop event including at least a change in appearance of a screen icon.

4. The method as recited in claim 1, wherein measuring includes comparing a screen state or event with a library of events stored in memory and employing the screen state or event to determine a start time and an end time for the interactive desktop event.

5. The method as recited in claim 1, wherein measuring includes determining a probability that a virtual component will remain active or idle and migrating workload in accordance with the probability.

6. The method as recited in claim 5, wherein determining a probability includes detecting idleness and active periods by setting a threshold for computer device usage and marking all periods below this threshold as idle and above the threshold as active.

7. The method as recited in claim 5, wherein measuring includes monitoring keyboard events, mouse events and screen buffer update activities on a user device to determine active and idle periods.

8. The method as recited in claim 5, wherein determining a probability includes computing the probability by F(x)=1−

a*x<sup>−

b</sup> (1) where b is a parameter depending on a workstation and a is a normalizing factor which enforces that F(x) is a probability distribution.

9. The method as recited in claim 8, wherein reallocating system resources to improve the responsiveness includes determining a probability that a given workstation is likely to remain idle or active knowing that it has been idle or active for time y (P(Residual lifetime>

x|Age=y)) such that a distribution is matched by the Equation 1, and the probability can be evaluated as;

10. A computer readable medium comprising a computer readable program for managing responsiveness of virtual desktops using passive monitoring, wherein the computer readable program when executed on a computer causes the computer to:

monitoring interactive desktop events for a computer device;

measuring a time between an interactive desktop event and a result of the interactive desktop event;

based on the time, evaluating desktop responsiveness; and

reallocating system resources to improve the responsiveness.

11. A method for managing responsiveness of virtual desktops using passive monitoring, comprising:

monitoring keyboard events, mouse events and screen buffer update activities on a user device to determine active and idle periods of a computer device;

measuring a time between an interactive desktop event and a result of the interactive desktop event and employing the time to determine a probability that a virtual component will remain active or idle;

based on the probability, migrating workload in accordance with the probability that resources will remain active or idle to improve responsiveness.

12. The method as recited in claim 11, wherein measuring includes comparing a screen state or event with a library of events stored in memory and employing the screen state or event to determine a start time and an end time for an interactive desktop event.

13. The method as recited in claim 12, wherein the probability is determined by detecting idleness and active periods by setting a threshold for computer device usage and marking all periods below this threshold as idle and above the threshold as active.

14. The method as recited in claim 13, wherein the probability is determined by computing F(x)=1−

a*x<sup>−

b</sup> (1) where b is a parameter depending on a workstation and a is a normalizing factor which enforces that F(x) is a probability distribution.

15. The method as recited in claim 14, wherein migrating workload to improve responsiveness includes determining a probability that a given workstation is likely to remain idle or active knowing that it has been idle or active for time y (P(Residual lifetime>

x|Age=y)) such that a distribution is matched by the Equation 1, and the probability can be evaluated as;

16. A computer readable medium comprising a computer readable program, wherein the computer readable program when executed on a computer causes the computer to perform the steps as recited in claim 11.

17. A system for managing responsiveness of virtual desktops, comprising:

a control system configured to receive responsiveness feedback and to reallocate system resources in accordance with the feedback, the control system being implemented in a computer device with memory storage, and comprising;

an activity/idleness profiler configured to receive information about interactive desktop events from a virtual desktop;

an allocation controller configured to receive responsiveness measurements based upon the information about interactive desktop events and to receive resource utilization measurements to determine how to allocate resources to improve responsiveness in the system; and

an executor configured to allocate the resources as determined by the allocation controller.

18. The system as recited in claim 17, wherein the activity/idleness profiler monitors timing between at least one of mouse events, keyboard events and pixel changes on a display and a result of the at least one event.

19. The system as recited in claim 17, further comprising a library of events stored in memory wherein a display screen is monitored to determine and compare displayed information with the library to determine a start time and an end time for an interactive desktop event.

20. The system as recited in claim 17, wherein the allocation controller determines a probability that a virtual component will remain active or idle to determine an expected workload.

21. The system as recited in claim 20, wherein the probability includes detecting idleness and active periods by setting a threshold for computer device usage and marking all periods below this threshold as idle and above the threshold as active.

22. The system as recited in claim 20, wherein the probability is based upon timing of at least one of keyboard events, mouse events and screen buffer update activities on a user device to determine active and idle periods.

23. The system as recited in claim 20, wherein the probability includes F(x)=1−

a*x<sup>−

b</sup> (1) where b is a parameter depending on a workstation and a is a normalizing factor which enforces that F(x) is a probability distribution.

24. The system as recited in claim 23, wherein the executor reallocates system resources to improve the responsiveness by determining that a given workstation is likely to remain idle or active knowing that it has been idle or active for time y (P(Residual lifetime>

x|Age=y)) such that a distribution is matched by the Equation 1, and the probability is evaluated as;