RESOURCE MANAGEMENT FOR CLOUD COMPUTING PLATFORMS

• US 20120331113A1
• Filed: 06/27/2011
• Published: 12/27/2012
• Est. Priority Date: 06/27/2011
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Jain, Navendu, Menache, Ishai
• Granted Patent: US 9,450,838 B2
• Time in Patent Office: Days
• US Class Current: 709/220
• CPC Class Codes: G06F 2209/508 Monitor
• CPC Class Codes: G06F 9/5072 Grid computing
• CPC Class Codes: H04L 41/5025 by proactively reacting to service quality change, e.g. by reconfiguration after service quality degradation or upgrade
• CPC Class Codes: H04L 47/803 Application aware
• CPC Class Codes: H04L 67/1008 based on parameters of servers, e.g. available memory or workload monitoring of computer activity G06F11/30

Abstract

A system for managing allocation of resources based on service level agreements between application owners and cloud operators. Under some service level agreements, the cloud operator may have responsibility for managing allocation of resources to the software application and may manage the allocation such that the software application executes within an agreed performance level. Operating a cloud computing platform according to such a service level agreement may alleviate for the application owners the complexities of managing allocation of resources and may provide greater flexibility to cloud operators in managing their cloud computing platforms.

Claims

1. A method for configuring a cloud computing platform to allocate resources of the cloud computing platform to a software application, the software application being associated with an application owner and the cloud computing platform being operated by a cloud operator, the method comprising:

via an interface of the cloud computing platform, receiving a specification of at least one term of a service level agreement (SLA) between the application owner and the cloud operator, the at least one term of the SLA comprising at least one performance level for execution of the software application by the cloud computing platform; and

configuring the cloud computing platform based on the at least one term such that the cloud computing platform allocates resources for use by the software application so as to meet the at least one performance level of the SLA.

2. The method of claim 1, further comprising:

during execution of the software application, comparing at least one current value of at least one performance metric to the at least one performance level of the SLA;

based on the comparing, adjusting allocation of resources to the software application such that the at least one current performance metric is compliant with the at least one performance level; and

adjusting billing of the application owner based on the comparing.

3. The method of claim 1, wherein receiving the specification of the at least one term of the SLA comprising the at least one performance level for execution of the software application comprises receiving a first specification of at least one limit on at least one performance metric.

4. The method of claim 3, wherein receiving the first specification of the at least one limit on the at least one performance metric comprises receiving an upper limit for acceptable execution times for an operation of the software application under the SLA.

5. The method of claim 4, wherein receiving the first specification of at least one limit further comprises receiving a lower limit for a success rate of the cloud computing platform in meeting the upper limit for acceptable execution times.

6. The method of claim 5, wherein:

receiving the lower limit for the success rate comprises receiving the lower limit and a unit of time over which to judge compliance with the lower limit for the success rate; and

configuring the cloud computing platform comprises configuring the cloud computing platform to allocate resources such that, within the unit of time, an actual success rate in maintaining execution times for the software application under the upper limit for acceptable execution times is above the lower limit for the success rate.

7. The method of claim 1, wherein receiving the specification of at least one term of the SLA further comprises receiving a second specification of at least one condition under which the at least one performance level will be met by the cloud computing platform.

8. The method of claim 7, wherein:

the software application is a user-interactive software application, and

receiving the second specification of the at least one condition comprises receiving a limit on an execution load to be imposed by the software application on the cloud computing platform, the limit on the execution load comprising a number of requests from users to be processed by the software application per unit of time.

9. The method of claim 1, wherein configuring the cloud computing platform such that the cloud computing platform allocates resources for use by the software application so as to meet the at least one performance level of the SLA comprises:

configuring the cloud computing platform to adjust an allocation of resources to the software application during execution of the software application such that values for at least one performance metric for execution of the software application are within the at least one performance level of the SLA.

10. At least one computer-readable storage medium encoded with computer-executable instructions that, when executed by a computer, cause the computer to carry out a method of managing allocation of resources of a cloud computing platform to a software application, the software application being associated with an application owner and the cloud computing platform being operated by a cloud operator, the method comprising:

configuring the cloud computing platform based on at least one term of a service level agreement (SLA) between the application owner and the cloud operator, the at least one term of the SLA identifying at least one performance level for execution of the software application by the cloud computing platform and at least one condition, the configuring comprising configuring the cloud computing platform to allocate the resources of the cloud computing platform to the software application so as to meet the at least one performance level specified in the SLA when the at least one condition is met;

during execution of the software application, comparing at least one current performance metric for the software application to the at least one performance level of the SLA and at least one current performance metric to the at least one condition; and

based on the comparing, managing allocation of the resources to the software application such that the at least one current performance metric has a value satisfying the at least one performance level when the at least one condition is met.

11. The at least one computer-readable storage medium of claim 10, wherein:

the cloud computing platform comprises a resource allocator to allocate the resources of the cloud computing platform to one or more software applications, and

the comparing and the managing are performed by the resource allocator.

12. The at least one computer-readable storage medium of claim 10, wherein managing allocation of the resources comprises adjusting allocation of hardware and/or software resources of the cloud computing platform to the software application.

13. The at least one computer-readable storage medium of claim 10, wherein:

comparing the at least one current performance metric to the at least one performance level comprises comparing a value for a current execution time for an operation of the software application to an upper limit for acceptable execution times for the operation, and

managing the allocation based on the comparing comprises, when the current execution time is above the upper limit for acceptable execution times and the at least one condition is met, allocating more resources to the software application.

14. The at least one computer-readable storage medium of claim 10, wherein:

comparing the at least one current performance metric to the at least one condition further comprises comparing a current execution load imposed by the software application on the cloud computing platform to a limit on an execution load to be imposed by the software application on the cloud computing platform, and

managing the allocation based on the comparing comprises, when the current execution load is below the limit on the execution load, allocating more resources to the software application when the at least one performance level is not met and, when the current execution load is above the limit on the execution load, refraining from allocating more resources to the software application when the at least one performance level is not met.

15. The at least one computer-readable storage medium of claim 10, wherein:

the resources comprise a plurality of virtual machines; and

adjusting the allocation of the resources comprises, when additional resources are to be allocated to the software application;

instantiating at least one virtual machine; and

configuring the at least one virtual machine to carry out at least a portion of functionality of the software application.

16. An apparatus to manage allocation of resources of a cloud computing platform to a software application, the software application being associated with an application owner and the cloud computing platform being operated by a cloud operator, the apparatus comprising:

at least one processor programmed to;

compare at least one current performance metric for the software application to at least one agreed performance level for the software application and to at least one agreed condition, the at least one current performance metric comprising an execution time for the software application and a current execution load imposed by the software application on the cloud computing platform, the at least one agreed performance level comprising an upper limit on acceptable execution times for an operation of the software application and the at least one agreed condition comprising a limit on an execution load to be imposed by the software application on the cloud computing platform; and

manage allocation of the resources to the software application such that, when the current execution load is less than the limit on the execution load to be handled by the software application, the at least one current performance metric is within the at least one agreed performance level.

17. The apparatus of claim 16, wherein the at least one processor is further programmed to:

monitor execution of the software application to determine the at least one current performance metric for the software application.

18. The apparatus of claim 16, wherein the at least one processor is programmed to manage allocation of the resources to the software application based at least in part on a determining regarding performance of resources previously allocated to the software application.

19. The apparatus of claim 16, wherein the at least one processor is programmed to manage allocation of the resources to the software application at least in part by, when the execution time is greater than the upper limit on acceptable execution times, allocating more resources to the software application.

20. The apparatus of claim 16, wherein the at least one processor is further programmed to:

configure the cloud computing platform based on at least one term of a service level agreement (SLA) between the application owner and the cloud operator, the at least one term of the SLA identifying the at least one agreed performance level for execution of the software application by the cloud computing platform, the configuring comprising configuring the cloud computing platform to allocate the resources of the cloud computing platform to the software application so as to meet the at least one agreed performance level specified in the SLA.