Brokering optimized resource supply costs in host cloud-based network using predictive workloads

• US 9,563,479 B2
• Filed: 11/30/2010
• Issued: 02/07/2017
• Est. Priority Date: 11/30/2010
• Status: Active Grant

• Current Assignee: Red Hat, Inc. (International Business Machines Corporation)
• Original Assignee: Red Hat, Inc. (International Business Machines Corporation)
• Inventors: Ferris, James Michael, Riveros, Gerry Edward
• Primary Examiner(s): Kassim, Khaled
• Assistant Examiner(s): Yang, Zhaohui
• Application Number:
  US12/957,274
• Publication Number:
  US 20120137002A1
• Time in Patent Office: 2,261 Days
• Field of Search: 718102-104, 709/224, 709/226
• US Class Current: 1/1
• CPC Class Codes: G06F 2209/5019 Workload prediction
• CPC Class Codes: G06F 2209/508 Monitor
• CPC Class Codes: G06F 9/5072 Grid computing

Abstract

Embodiments relate to brokering optimized resource supply costs in a host cloud-based network using predictive workloads. In aspects, a set of usage histories can store patterns for users in a host cloud-based network recording the consumption of processor, memory, storage, operating system, application, or other resources subscribed to by the user. The operator of the cloud management system hosting the workloads of one or more users can track the aggregate user histories reflecting those workloads, and analyze them in terms of different time periods over which demands for resources may intensity or relax. The cloud management system can examine those workload patterns, and generate a set of predictively re-assigned workloads to be supported by different sets of resource servers during offpeak and/or other periods. The workload realignment to different resource servers can permit the operator of the cloud management system and/or other cloud provider to realize cost savings or other benefits by procuring resources on a predictive basis.

Claims

1. A method, comprising:

identifying a set of operating workloads for a set of client machines in a set of host clouds in a first time period, the set of client machines being instantiated via a first set of resource servers, the set of host clouds being communicably coupled to a centralized cloud management system via a network,

determining, by a processing device of the cloud management system, resource consumption limits for the set of client machines in view of a subscription period in which subscription users are able to request resources from the set of host clouds to host the set of operating workloads;

accessing, by the processing device, a resource usage history of the set of client machines in the set of host clouds, the resource usage history comprising a record of consumption of resources consumed from the first set of resource servers;

generating, by the processing device, a set of predictively re-assigned workloads from the set of operating workloads in view of the resource usage history, a set of resource procurement parameters associated with the first set of resource servers and the resource consumption limits during the subscription period; and

responsive to a resource shortfall for the set of operating workloads in view of the resource consumption limits, migrating, by the processing device, the set of predictively re-assigned workloads to resource servers of a second set of resource servers, wherein the resource servers are scheduled to execute the set of predictively re-assigned workloads in a second time period, each of the resource servers of the first set of resource servers being different from each of the resource servers of the second set of resource servers,

wherein the set of predictively re-assigned workloads comprises a set of reduced estimated workloads to be inserted for execution on the resource servers of the second set of resource servers.

2. The method of claim 1, wherein the record of consumption of resources indicates a greater consumption of resources from the first set of resource servers in the first time period than in the second time period.

3. The method of claim 2, wherein the set of predictively re-assigned workloads comprises a set of reduced estimated workloads to be inserted for execution on the second set of resource servers.

4. The method of claim 3, wherein the second set of resource servers comprises a reduced workload as compared to the first set of resource servers.

5. The method of claim 4, wherein the set of resource procurement parameters for the second set of resource servers in the second time period comprises a lower net cost to a cloud provider for the provisioning of the set of consumed resources than the first set of resource servers.

6. The method of claim 1, further comprising updating the resource usage history to reflect resource consumption from the second set of resource servers in the second time period.

7. The method of claim 1, further comprising receiving user input to select the set of predictively re-assigned workloads.

8. The method of claim 1, wherein the resources consumed from the first set of resource servers comprises at least one of processor resources, memory resources, storage resources, bandwidth resources, communications port resources, operating system resources, application resources, or appliance resources.

9. The method of claim 1, wherein the set of resource procurement parameters comprises at least one of a subscription cost, a minimum number of software images to be deployed in the set of host clouds, a maximum number of software images to be deployed in the set of host clouds, an operating system specification, a software application specification, a minimum processor throughput specification, a maximum processor throughput specification, a minimum memory amount specification, a maximum memory amount specification, a minimum storage capacity specification, a maximum storage capacity specification, a minimum bandwidth specification, a maximum bandwidth specification, a minimum set of communications ports specification, a maximum set of communications ports specification, an appliance specification, a service level agreement specification, clustering services, identity management services, backup services, security services, or billing and administrative services.

11. The method of claim 1, further comprising adjusting the resource consumption limits in view of a subscription cost.

12. The method of claim 1, wherein the set of predictively re-assigned workloads reflects an accumulated usage consumption associated with a subscription user account over the first time period.

13. The method of claim 1, wherein the mitigating comprises determining a minimum cost reduction associated with a subscription user account to satisfying the set of predictively re-assigned workloads in view of the shortfall.

21. The method of claim 1, wherein the second set of resource servers offer a specified quantity of resources for the second time period.

10. A system, comprising:

an interface to a data store, the data store to store a resource usage history of a set of client machines in a set of host clouds, the resource usage history comprising a record of consumption of resources consumed from a set of resource servers; and

a processor of a centralized cloud management system, to communicate with the data store via the interface, the processor to;

identify a set of operating workloads for the set of client machines in the set of host clouds in a first time period, the set of client machines being instantiated via a first set of resource servers, the set of host clouds being communicably coupled to the cloud management system via a network,

determine resource consumption limits for the set of client machines in view of a subscription period in which subscription users are able to request resources from the set of host clouds to host the set of operating workloads,

access the resource usage history of the set of client machines in the set of host clouds,

generate a set of predictively re-assigned workloads from the set of operating workloads in view of the resource usage history, a set of resource procurement parameters associated with the with the first set of resource servers and the resource consumption limits during the subscription period, and

responsive to a resource shortfall for the set of operating workloads in view of the resource consumption limits, migrate the set of predictively re-assigned workloads resource servers of a second set of resource servers, wherein the resource servers are scheduled to execute the set of predictively re-assigned workloads in a second time period, each of the resource servers of the first set of resource servers being different from each of the resource servers of the second set of resource servers,

wherein the set of predictively re-assigned workloads comprises a set of reduced estimated workloads to be inserted for execution on the resource servers of the second set of resource servers.

14. The system of claim 10, wherein the set of resource procurement parameters comprises at least one of a subscription cost, a minimum number of software images to be deployed in the set of host clouds, a maximum number of software images to be deployed in the set of host clouds, an operating system specification, a software application specification, a minimum processor throughput specification, a maximum processor throughput specification, a minimum memory amount specification, a maximum memory amount specification, a minimum storage capacity specification, a maximum storage capacity specification, a minimum bandwidth specification, a maximum bandwidth specification, a minimum set of communications ports specification, a maximum set of communications ports specification, an appliance specification, a service level agreement specification, clustering services, identity management services, backup services, security services, or billing and administrative services.

15. The system of claim 10, wherein the record of consumption of resources indicates a greater consumption of resources from the first set of resource servers in the first time period than in the second time period.

16. The system of claim 10, wherein the second set of resource servers comprises a reduced workload as compared to the first set of resource servers.

17. The system of claim 16, wherein the set of resource procurement parameters for the second set of resource servers in the second time period comprises a lower net cost to a cloud provider for the provisioning of the set of consumed resources than the first set of resource servers.

18. The system of claim 10, wherein the processor further to update the resource usage history to reflect resource consumption from the second set of resource servers in the second time period.

19. The system of claim 10, wherein the processor further to receive user input to select the set of predictively re-assigned workloads.

20. The system of claim 10, wherein the resources consumed from the first set of resource servers comprises at least one of processor resources, memory resources, storage resources, bandwidth resources, communications port resources, operating system resources, application resources, or appliance resources.

22. The system of claim 10, wherein the second set of resource servers offer a specified quantity of resources for the second time period.

23. A non-transitory computer-readable medium comprising executable instructions that, when executed by a processing device, cause the processing device to:

identify a set of operating workloads for a set of client machines in a set of host clouds in a first time period, the set of client machines being instantiated via a first set of resource servers,

determine, by the processing device, resource consumption limits for the set of client machines in view of a subscription period in which subscription users are able to request resources from the set of host clouds to host the set of operating workloads,

access the resource usage history of the set of client machines in the set of host clouds,

generate a set of predictively re-assigned workloads from the set of operating workloads in view of the resource usage history, a set of resource procurement parameters associated with the with the first set of resource servers and the resource consumption limits during the subscription period, and

responsive to a resource shortfall for the set of operating workloads in view of the resource consumption limits, migrate the set of predictively re-assigned workloads to resource servers of a second set of resource servers, wherein the resource servers are scheduled to execute the set of predictively re-assigned workloads in a second time period, each of the resource servers of the first set of resource servers being different from each of the resource servers of the second set of resource servers,

wherein the set of predictively re-assigned workloads comprises a set of reduced estimated workloads to be inserted for execution on the resource servers of the second set of resource servers.