MANAGEMENT OF A VIRTUAL POWER INFRASTRUCTURE

• US 20110307110A1
• Filed: 06/10/2010
• Published: 12/15/2011
• Est. Priority Date: 06/10/2010
• Status: Active Grant

• Current Assignee: Hewlett Packard Enterprise Development LP (Hewlett-Packard Enterprise Company)
• Original Assignee: Hewlett Packard Enterprise Development LP (Hewlett-Packard Enterprise Company)
• Inventors: Sharma, Ratnesh Kumar, Shah, Amip J., Bash, Cullen E., Shih, Chih C., Patel, Chandrakant
• Granted Patent: US 9,063,715 B2
• Time in Patent Office: Days
• US Class Current: 700/291
• CPC Class Codes: G06F 1/26 Power supply means, e.g. regulation thereof for memories G11C

Abstract

A virtual power infrastructure for managing distribution of power to a plurality of power demanding units from a plurality of power generators includes a demand manager having logical descriptions of the power demanding units, in which the demand manager is configured to determine power load demands of the power demanding units using the logical descriptions of the power demanding units, and a capacity manager having logical descriptions of the power generators. The capacity manager is configured to determine allocation of power capacity supplied to the power generators from one or more of the power generators to meet the power load demands of the power demanding units, while satisfying one or more predetermined constraints associated with at least one of the power demanding units and the power generators.

Claims

1. A virtual power infrastructure for managing distribution of power to a plurality of power demanding units from a plurality of power generators, said infrastructure comprising:

a demand manager having logical descriptions of the power demanding units, wherein the demand manager is configured to determine power load demands of the power demanding units using the logical descriptions of the power demanding units; and

a capacity manager having logical descriptions of the power generators, said power generators being configured to supply power to the power demanding units, said capacity manager being configured to determine allocation of power capacity supplied to the power demanding units from one or more of the power generators to meet the power load demands of the power demanding units, while satisfying one or more predetermined constraints associated with at least one of the power demanding units and the power generators.

2. The virtual power infrastructure according to claim 1, wherein the demand manager is configured to determine the power load demands of the power demanding units based upon one or more aspects of workloads that are at least one of performed and scheduled to be performed by the power demanding units and one or more power demand constraints pertaining to performance of the workloads.

3. The virtual power infrastructure according to claim 1, wherein the demand manager is further configured to determine at least one characteristic of the power load demand, said additional characteristics comprising one or more of operational levels, redundancy, and cost of deployment to satisfy the power load demands of the power demanding units, and wherein the capacity manager is further configured to determine allocation of the power capacity supplied to the power generators based upon the at least one characteristic.

4. The virtual power infrastructure according to claim 1, wherein the logical descriptions of the power demanding units comprise one or more of power reliability requirements, power load demand descriptions, power load demand forecasts, power load demand recovery patterns, criticalities, and elasticity of demand based on pricing model, of the power demanding units.

5. The virtual power infrastructure according to claim 1, wherein the power load demands comprise one or more of power load demand estimates, locations of power load demands, durations of power load demands, costs of deploying the power load demands, power load demand zone designations, and sustainability limits of the power demanding units.

6. The virtual power infrastructure according to claim 1, wherein the logical descriptions of the power generators comprise one or more of descriptions, energy efficiencies, resource consumption levels, cost functions, emissions, environmental footprints, reliability levels, power output capacity limitations, and available capacities of the power generators.

7. The virtual power infrastructure according to claim 6, wherein the capacity manager is further configured to determine allocation of the power capacities of the power generators to substantially optimize at least one of efficiency, cost, and fault tolerance of the power generators in supplying power to the power demanding units based upon the logical descriptions of the power generators.

8. The virtual power infrastructure according to claim 1, wherein the capacity manager is configured to determine one or more of allocation of the power supply capacities, identification of zones for allocation of the power supply capacities, estimates of the power to be supplied, the total cost of ownership, the utilization levels, and the environmental impact associated with supplying power to meet the power load demands as determined by the demand manager.

9. The virtual power infrastructure according to claim 1, wherein the capacity manager is configured to follow one or more of techniques in determining the allocation of the power capacity, said one or more techniques comprising a hierarchical assignment technique, a parameterized match technique, and a phase assignment technique.

10. The virtual power infrastructure according to claim 1, wherein the one or more predetermined constraints include one or more of reliability levels of the electrical power required by the power demanding units, the quality of the electrical power required by the power demanding units, various regulations that the power demanding units are required to comply with, service level agreement (SLA) provisions, uptime requirements of the power demanding units, power generator capacity constraints, and power generator distribution constraints.

11. The virtual power infrastructure according to claim 1, wherein the demand manager is configured to interpret data pertaining to the power load demands of the power demanding units through one or more of stream mining of discrete data to identify recurring patterns in the data, pattern recognition techniques, and state-wise evaluation.

12. The virtual power infrastructure according to claim 1, further comprising:

a service operator configured to operate as a monitoring agent and a broker for distribution of power from the power generators to the power demanding units, wherein the service operator is further configured to monitor operations of the power demanding units and the power generators to substantially ensure that the power demanding units are receiving sufficient amounts of power while meeting the one or more predetermined constraints.

13. The virtual power infrastructure according to claim 1, wherein the demand manager, the capacity manager, and the service operator form part of a single overlay in an integrated power management system.

14. A method of virtualized management of power distribution among a plurality of power demanding units from a plurality of power generators, said method comprising:

determining power load demands of the power demanding units;

storing the power load demands as a first set of logical descriptions;

determining characteristics of the power generators;

storing the characteristics of the power generators as a second set of logical descriptions; and

determining allocation of power capacity from one or more of the power generators to meet the power load demands of the power demanding units based upon the first set of logical descriptions and the second set of logical descriptions while satisfying one or more predetermined constraints associated with at least one of the power demanding units and the power generators.

15. The method according to claim 14, wherein determining power load demands of the power demanding units further comprises determining the power load demands of the power demanding units based upon one or more aspects of workloads that are at least one of performed and scheduled to be performed by the power demanding units and one or more power demand constraints pertaining to performance of the workloads.

16. The method according to claim 14, wherein determining characteristics of the power generators further comprises determining one or more of descriptions, energy efficiencies, resource consumption levels, cost functions, emissions, environmental footprints, reliability levels, power output capacity limitations, and available capacities of the power generators.

17. The method according to claim 16, wherein determining allocation of the power capacities further comprises determining allocation of the power capacities to substantially optimize at least one of efficiency, cost, and fault tolerance of the power generators in supplying power to the power demanding units based upon the logical descriptions of the power generators.

18. The method according to claim 14, further comprising:

accessing the one or more predetermined constraints, wherein the one or more predetermined constraints comprise one or more constraints pertaining to the supply of electrical power to the power demanding units, reliability levels of the electrical power required by the power demanding units, the quality of the electrical power required by the power demanding units, various regulations that the power demanding units are required to comply with, service level agreement (SLA) provisions, uptime requirements of the power demanding units, power generator capacity constraints, and power generator distribution constraints.

19. A computer readable storage medium on which is embedded one or more computer programs, said one or more computer programs implementing a method of virtualized management of power distribution among a plurality of power demanding units from a plurality of power generators, said one or more computer programs comprising a set of instructions for:

determining power load demands of the power demanding units;

storing the power load demands as a first set of logical descriptions;

determining characteristics of the power generators;

storing the characteristics of the power generators as a second set of logical descriptions; and

determining allocation of power capacity from one or more of the power generators to meet the power load demands of the power demanding units based upon the first set of logical descriptions and the second set of logical descriptions while satisfying one or more predetermined constraints associated with at least one of the power demanding units and the power generators.

20. The computer readable storage medium according to claim 19, said one or more computer programs further including a set of instructions for:

determining the power load demands of the power demanding units based upon one or more aspects of workloads that are at least one of performed and scheduled to be performed by the power demanding units and one or more power demand constraints pertaining to performance of the workloads.