Remote desktop performance model for assigning resources

• US 7,870,256 B2
• Filed: 10/12/2005
• Issued: 01/11/2011
• Est. Priority Date: 03/25/2005
• Status: Active Grant

• Current Assignee: Valtrus Innovations Limited (f/k/a Dolya Holdco 9 Limited) (Key Patent Innovations Limited)
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: Talwar, Vanish, Basu, Sujoy, Kumar, Raj
• Primary Examiner(s): ISMAIL, SHAWKI SAIF
• Application Number:
  US11/248,932
• Publication Number:
  US 20060218285A1
• Time in Patent Office: 1,917 Days
• Field of Search: 709/223, 709/224, 709/226
• US Class Current: 709/226
• CPC Class Codes: G06F 2209/5016 Session
• CPC Class Codes: G06F 2209/5019 Workload prediction
• CPC Class Codes: G06F 9/5072 Grid computing
• CPC Class Codes: H04L 67/08 specially adapted for terminal emulation, e.g. Telnet
• CPC Class Codes: H04L 67/60 Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources admission control or resource allocation H04L47/70

Abstract

A request for a remote desktop session is received. A remote desktop session performance model is generated based on the applications to be used in the remote desktop session, and resources are assigned to the remote desktop session using the remote desktop session performance model.

Claims

1. A method comprising:

receiving a request for a remote desk-top session;

generating, by a processor a remote desktop session performance model based on applications to be used in the remote desktop session wherein the remote desktop session performance model is used to estimate the resource requirements for running the applications in the remote desktop session while maintaining a predetermined QoS; and

assigning at least one resource to run the remote desk-top session using the remote desktop session performance model.

2. The method of claim 1, wherein generating a remote desktop session performance model based on applications to be used in the remote desktop session further comprises:

retrieving an application performance model for each of the applications;

determining an order of execution for the applications; and

determining the remote desktop session performance model based on the application performance model for each application and the order of execution.

3. The method of claim 2, wherein determining an order of execution for the applications further comprises:

determining a model for estimating the order of execution for the applications.

4. The method of claim 3, wherein determining a model for estimating the order of execution for the applications further comprises:

selecting one of a plurality of models operable to estimate the order of execution.

5. The method of claim 3, wherein the model for estimating the order of execution models the applications as being executed simultaneously.

6. The method of claim 3, wherein the model for estimating the order of execution models the applications as being executed sequentially.

7. The method of claim 3, wherein the model for estimating the order of execution models some of the applications as being executed simultaneously and some of the applications as being executed sequentially.

8. The method of claim 3, wherein the model for estimating the order of execution models the applications as requiring a maximum utilization of resources for a compute node.

9. The method of claim 2, wherein determining the remote desktop session performance model further comprises:

determining a number of instances to be executed during the remote desktop session for each application; and

determining the remote desktop session performance model based on the number of instances to be executed for each application.

10. The method of claim 2, wherein determining an order of execution for the applications further comprises:

receiving the order of execution from a user.

11. The method of claim 1, further comprising:

determining static resource requirements for at least one resource for running the remote desktop session; and

determining whether at least one resource at a site satisfies the static resource requirements;

denying the request in response to no resources at the site satisfying the request; and

the assigning further comprises assigning the at least one resource determined to satisfy the static resource requirements to run the remote desktop session in response to determining the at least one resource satisfies the static resource requirements.

12. The method of claim 11, further comprising:

placing the request in a pending queue in response to the at least one resource being currently unavailable to run the remote desktop session.

13. The method of claim 1, further comprising:

simultaneously running the remote desktop session and a batch job on the at least one resource.

14. A resource management system, comprising:

a processor;

a remote desktop session performance model module operable to generate, by the processor, a remote desktop session performance model for estimating resources needed to run an interactive remote desktop session while maintaining a predetermined QoS;

a site admission control module operable to determine whether at least one resource satisfies resource requirements for the interactive remote desktop session based on the remote desktop session performance model; and

a resource assignment module operable to assign the at least one resource to the interactive remote desktop session, wherein the at least one resource satisfies the resource requirements for the interactive remote desktop session.

15. The resource management apparatus of claim 14, wherein the remote desktop session performance model module is operable to use application performance models for applications to be run in the interactive remote desktop session to generate the remote desktop session performance model.

16. The resource management apparatus of claim 15, wherein the application performance models are determined from trace data for the applications.

17. The resource management apparatus of claim 15, wherein the remote desktop session performance model module is operable to estimate an order of execution of the applications for generating the remote desktop session performance model.

18. The resource management apparatus of claim 14, further comprising:

a session admission control module for determining whether to start an application in the interactive remote desktop session based on the remote desktop session performance model.

19. The resource management apparatus of claim 18, further comprising:

a monitoring system operable to monitor the utilization of the allocated at least one resource.

20. The resource management apparatus of claim 14, wherein the site admission control module is operable to match resource models of resources at the site with static resource requirements from a user for running the interactive remote desktop session.

21. An apparatus comprising:

means for generating a remote desktop session performance model for each of a plurality of requests for remote desktop sessions;

means for determining whether at least one resource satisfies resource requirements for a requested remote desktop session;

means for assigning the at least one resource to the requested remote desktop session; and

means for determining whether to grant a request to start an application in the remote desktop session using the remote desktop performance model, wherein the remote desktop session performance model is used to estimate the resource requirements for running the application in the remote desktop session while maintaining a predetermined QoS.

22. A non-transitory computer readable storage medium upon which is stored a computer program including instructions which when executed by a processor cause the processor to perform the following comprising:

determining applications to be executed in a requested remote desktop session;

retrieving application performance models for the applications;

generating a remote desktop session performance model based on the application performance models;

wherein the remote desktop session performance model is used to estimate the resource requirements for running the applications in the remote desktop session while maintaining a predetermined QoS; and

selecting at least one resource to assign to the requested remote desktop session using the remote desktop session performance model.