Dynamically allocating multitier applications based upon application requirements and performance and reliability of resources

• US 8,489,939 B2
• Filed: 10/25/2010
• Issued: 07/16/2013
• Est. Priority Date: 10/25/2010
• Status: Active Grant

• Current Assignee: AT&T Intellectual Property I LP (AT&T, Inc.)
• Original Assignee: AT&T Intellectual Property I LP (AT&T, Inc.)
• Inventors: Hiltunen, Matti, Jung, Gueyoung, Pu, Calton, Schlichting, Richard Dale, Joshi, Kaustubh R.
• Primary Examiner(s): Iqbal, Nadeem
• Application Number:
  US12/911,201
• Publication Number:
  US 20120102369A1
• Time in Patent Office: 995 Days
• Field of Search: 714/1, 714/2, 714/4.1, 714/6.13, 714/6.3, 714/6.31, 714/6.32, 714/15, 714/16, 714/37, 714/38.1, 714/39, 714/41, 714/47.1, 714/48, 714/49
• US Class Current: 714/48
• CPC Class Codes: G06F 11/008 Reliability or availability analysis
• CPC Class Codes: G06F 11/1482 by means of middleware or OS functionality
• CPC Class Codes: G06F 11/3495 for systems
• CPC Class Codes: G06F 2209/508 Monitor
• CPC Class Codes: G06F 9/50 Allocation of resources, e.g. of the central processing unit [CPU]
• CPC Class Codes: G06F 9/5027 the resource being a machine, e.g. CPUs, Servers, Terminals

Abstract

The present disclosure relates to dynamically allocating multitier applications based upon performance and reliability of resources. A controller analyzes resources and applications hosted by the resources, and collects operational data relating to the applications and resources. The controller is configured to determine an allocation scheme for allocating or reallocating the applications upon failure of a resource and/or upon rollout or distribution of a new application. The controller generates configuration data that describes steps for implementing the allocation scheme. The resources are monitored, in some embodiments, by monitoring devices. The monitoring devices collect and report the operational information and generate alarms if resources fail.

Claims

1. A method comprising:

detecting, at a controller executing an allocation application, a failure associated with resources being associated with a multitier application;

obtaining, by the controller, operational data associated with the resources and the multitier application; and

determining, by the controller, an allocation scheme for the multitier application, the allocation scheme comprising data indicating allocation of the multitier application across the resources, wherein determining the allocation scheme comprises identifying a resource level associated with the resources that satisfies an application reliability requirement of the multitier application, determining if the components of the multitier application can be allocated within the resource level, and if the components can be allocated within the resource level, outputting configuration data corresponding to the allocation scheme, the configuration data comprising computer executable instructions for allocating the multitier application across the resources according to the allocation scheme.

2. The method of claim 1, wherein the failure comprises an impending failure.

3. The method of claim 1, wherein the operational data comprises multitier application data relating to the multitier application and resource data relating to the resources.

4. The method of claim 3, wherein the resource data comprises performance information associated with the resources and reliability information associated with the resources.

5. The method of claim 4, wherein the application data indicates an availability requirement of the multitier application and a performance requirement of the multitier application.

6. The method of claim 4, wherein:

the application data indicates a reliability requirement of the multitier application and a performance requirement of the multitier application;

the reliability information comprises a mean time between failures associated with the resources;

the reliability requirement comprises a mean time between failures requirement associated with the multitier application; and

the availability requirement comprises a probability that the application is operational and accessible to serve application service requests.

7. The method of claim 3, wherein

the resource data comprises capacity information, communication latency information, and reliability information;

and

the reliability information comprises a mean time between failures associated with the resources.

8. The method of claim 1, further comprising:

if the application cannot be allocated within the resource level, identifying a reconfiguration action for allocating the application components;

determining if a reconfigured application can be allocated within the resource level;

if the reconfigured application can be allocated, outputting the configuration information; and

if the reconfigured application cannot be allocated, identifying alternative reconfiguration actions and determining if the application can be allocated until the allocation succeeds and the configuration information is outputted, wherein the reconfiguration action comprises one of reducing a resource allocation of an application component, reducing the resource allocation of an application tier, reducing the number of replicas for one or more application components associated with the application, and moving the application tier to a higher resource level, and wherein the reconfiguration action is chosen such that the reconfiguration action minimizes an overall performance degradation and maximizes an amount of additional resource capacity made available.

9. The method of claim 1, wherein the resource level corresponds to a lowest resource tier of a plurality of resource tiers of the resources, the plurality of resource tiers comprising a data center tier, a server cluster tier, a rack tier, and a server tier.

10. The method of claim 8, wherein:

identifying the reconfiguration action comprises executing a greedy search algorithm to identify an optimal reconfiguration action; and

determining if the application can be allocated at a resource level by executing a fit algorithm to analyze a configuration identified via execution of the greedy search algorithm.

11. The method of claim 10, wherein the fit algorithm comprises a hierarchical bin-packing algorithm configured to begin packing a top resource level of a resource hierarchy and to proceed with packing remaining resource levels until application components are allocated to individual resources of physical machines, wherein:

applications targeted to a resource level below a level being packed by the algorithm are considered to be one application bundle with resource demand equivalent to the sum of the application component resource levels;

application bundles are divided into individual application components when the algorithm reaches a level to which the application bundles are assigned;

a fit algorithm returns an error if the applications cannot be allocated; and

if the applications can be allocated, the fit algorithm returns the placement of application components in the resources.

12. A computer-readable storage medium having computer executable instructions stored thereupon that, when executed by a controller, cause the controller to perform operations comprising:

detecting, at the controller, a failure of resources in communication with the controller, the resources hosting a multitier application;

obtaining, in response to detecting the failure, operational data associated with the resources and the multitier application, the operational data comprising application data relating to the multitier application and resource data relating to the resources; and

determining an allocation scheme for the multitier application, the allocation scheme comprising data indicating allocation of the multitier application across the resources, wherein determining the allocation scheme comprises identifying a resource level associated with the resources, the resource level satisfying an application reliability requirement of the multitier application, determining if the components of the multitier application can be allocated within the resource level, and if the components can be allocated within the resource level, outputting configuration data corresponding to the allocation scheme, the configuration data comprising computer executable instructions for allocating the multitier application across the resources according to the allocation scheme.

13. The computer-readable storage medium of claim 12, wherein detecting the failure comprises:

detecting a new multitier application that is not supported by a current configuration of the resources; and

receiving an alarm generated by a monitoring device operating in communication with the resources.

14. The computer-readable storage medium of claim 12, wherein:

the resource data comprises performance information associated with the resources and reliability information associated with the resources; and

the application data indicates a reliability requirement of the multitier application and a performance requirement of the multitier application, wherein

the reliability information comprises a mean time between failures associated with the resources, and wherein

the reliability requirement comprises a mean time between failures requirement associated with the multitier application.

15. The computer-readable storage medium of claim 12, wherein:

the resource data comprises performance information associated with the resources and reliability information associated with the resources; and

the application data indicates an availability requirement of the multitier application and a performance requirement of the multitier application, wherein

the reliability information comprises a mean time between failures associated with the resources, and wherein

the availability requirement comprises a probability that the application is able to serve application requests at any given time.

16. The computer-readable storage medium of claim 12, further comprising computer executable instructions that, when executed by the controller, cause the controller to perform operations comprising:

if the application cannot be allocated within the resource level, identifying a reconfiguration action for allocating application components, the reconfiguration action comprising moving the application to a further resource level associated with the resources;

determining if the application can be allocated within the further resource level; and

if the application can be allocated within the further resource level, outputting configuration information corresponding to the further resource level.

19. The computer-readable storage medium of claim 12, wherein:

the application data indicates an availability requirement of the multitier application and a performance requirement of the multitier application,

the application data indicates a reliability requirement of the multitier application and a performance requirement of the multitier application,

the reliability information comprises a mean time between failures associated with the resources,

the reliability requirement comprises a mean time between failures requirement associated with the multitier application, and

the availability requirement comprises a probability that the application is operational and accessible to serve application service requests.

17. A method comprising:

detecting, at a controller executing an allocation application, a failure associated with the resources, the resources being associated with a multitier application and comprising a plurality of resource tiers;

obtaining, by the controller, operational data associated with the resources and the multitier application, the operational data comprising application data relating to the multitier application and resource data relating to the resources; and

determining, by the controller, an allocation scheme for the multitier application, the allocation scheme comprising data indicating allocation of the multitier application across the resources to adjust for the failure detected, wherein determining the allocation scheme comprises identifying a resource level associated with the resources, the resource level satisfying an application reliability requirement of the multitier application, determining if the components of the multitier application can be allocated within the resource level, and if the components can be allocated within the resource level, outputting configuration data corresponding to the allocation scheme, the configuration data comprising computer executable instructions for allocating the multitier application across the resources according to the allocation scheme.

18. The method of claim 17, wherein:

the resource data comprises performance information associated with the resources and reliability information associated with the resources; and

the application data indicates a reliability requirement of the multitier application and a performance requirement of the multitier application, wherein

the reliability information comprises a mean time between failures associated with the resources, and wherein

the reliability requirement comprises a mean time between failures requirement associated with the multitier application.

20. The method of claim 17, wherein:

the application data indicates a reliability requirement of the multitier application and a performance requirement of the multitier application;

the reliability information comprises a mean time between failures associated with the resources;

the reliability requirement comprises a mean time between failures requirement associated with the multitier application; and

the availability requirement comprises a probability that the application is operational and accessible to serve application service requests.