Reserving resources in a resource-on-demand system for user desktop utility demand

• US 8,209,695 B1
• Filed: 07/28/2006
• Issued: 06/26/2012
• Est. Priority Date: 07/28/2006
• Status: Active Grant

• Current Assignee: Hewlett Packard Enterprise Development LP (Hewlett-Packard Enterprise Company)
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: Pruyne, James C., Sahai, Akhil, Goswami, Kumar
• Primary Examiner(s): Wu, Qing
• Application Number:
  US11/495,357
• Time in Patent Office: 2,160 Days
• Field of Search: None
• US Class Current: 718/104
• CPC Class Codes: G06F 2209/5014 Reservation
• CPC Class Codes: G06F 9/5077 Logical partitioning of resources; Management or configuration of virtualized resources specific details on emulation or internal functioning of virtual machines G06F9/455

Abstract

Resources in a resource-on-demand system are reserved to support user desktop utility demand. A demand calendar specifies capacities of resources in the resource-on-demand system reserved to support user desktop utility demand for future time periods. A requested capacity is compared to capacities from the demand calendar for a requested future period of time to determine whether resources for the requested capacity are to be reserved.

Claims

1. A method for reserving resources in a resource-on-demand system to support user desktop utility demand, the method comprising:

storing a demand calendar specifying capacities of resources in the resource-on-demand system reserved to support user desktop utility demand for future time periods;

receiving a request for resources to support user desktop utility demand for a requested future time period;

determining a requested capacity for the user desktop utility demand; and

using the requested capacity and capacities from the demand calendar for the requested future period of time to determine whether resources for the requested capacity are to be reserved by

determining an available capacity for each of a plurality of intervals in the requested future period of time; and

reserving the requested capacity for the requested future period of time if the available capacity for each interval is greater than the requested capacity.

2. The method of claim 1, further comprising:

sizing a virtual machine for each interval, wherein each virtual machine is allocated to the requested capacity to support user desktop utility demand for the interval.

3. The method of claim 1, further comprising:

rejecting the request if the available capacity is less than the requested capacity for any of the intervals.

4. The method of claim 1, further comprising:

determining a resource allocation objective; and

selecting resources to be allocated to the requested capacity using a fitting function to meet the resource allocation objective.

5. The method of claim 4, further comprising:

calculating a fitness value to determine how well the resource allocation objective is being met.

6. The method of claim 1, further comprising:

storing a plurality of user classes and a capacity requirement for each user class;

determining the requested capacity based on a number of users in each user class specified in the request and the stored capacity requirement for each user class.

7. The method of claim 6, further comprising:

determining a priority for each of a plurality of requested capacities scheduled on the demand calendar; and

rejecting a scheduled requested capacity if a priority for a newly requested capacity of higher priority is received.

8. The method of claim 7, wherein the priority of each of the requested capacities is determined based on the highest priority user class associated with each requested capacity.

9. The method of claim 1, further comprising:

calculating a single capacity unit value to represent the requested capacity; and

comparing the single capacity unit value to capacity unit values for available resources for the requested future time period to determine whether to reserve the resources for the requested capacity for the requested future period of time.

10. The method of claim 9, wherein the request includes a request for resources characterized by multiple attributes and the method comprises:

identifying a quantum for each resource attribute;

calculating candidate capacity unit values for the resource attributes using the quantums; and

selecting one of the candidate capacity unit values as the single capacity unit value.

11. The method of claim 10, further comprising:

identifying the quantum for each resource attribute from a policy of a plurality of policies.

12. The method of claim 1, further comprising:

converting multiple resource attributes for the request to a plurality of capacity unit values; and

selecting one capacity unit value from the plurality of the capacity unit values as the requested capacity based on whether the selected capacity unit value is to represent the requested capacity or a capacity of a physical machine to support the user desktop utility demand for the request.

13. A system comprising:

a database storing a demand calendar specifying capacities of resources in the resource-on-demand system reserved to support user desktop utility demand for future time periods; and

a capacity manager receiving a request for resources to support user desktop utility demand for a requested future period of time and determining a requested capacity for the user desktop utility demand, wherein the capacity manager includes a fitter using the requested capacity and capacities from the demand calendar for the requested future period of time to determine whether resources for the requested capacity are to be reserved by

determining an available capacity for each of a plurality of intervals in the requested future period of time; and

reserving the requested capacity for the requested future period of time if the available capacity for each interval is greater than the requested capacity.

14. The system of claim 13, wherein the capacity manager further comprises:

a capacity unit converter converting multiple resource attributes for the request to a plurality of capacity unit values and the capacity manager selects one capacity unit value from the plurality of the capacity unit values as the requested capacity based on whether the selected capacity unit value is to represent the requested capacity or a capacity of a physical machine to support the user desktop utility demand for the request.

15. The system of claim 14, wherein the fitter compares the selected capacity unit value representing the requested capacity to available capacity unit values in the demand calendar for the requested future time period to determine whether to reserve the resources.

16. The system of claim 13, wherein the database stores a plurality of user classes and a capacity requirement for each user class, and the requested capacity is determined based on the capacity requirement for each user class.

17. The system of claim 16, wherein reservations for requested capacities are prioritized based on priorities for user classes.

18. A non-transitory computer readable medium upon which is embedded programmed instructions which when executed by a processor will cause the processor to perform a method for reserving resources in a resource-on-demand system to support user desktop utility demand, the method comprising:

storing a demand calendar specifying capacities of resources in the resource-on-demand system reserved to support user desktop utility demand for future time periods;

receiving a request for resources to support user desktop utility demand for a requested future time period;

determining a requested capacity for the user desktop utility demand; and

using the requested capacity and capacities from the demand calendar for the requested future period of time to determine whether resources for the requested capacity are to be reserved by

determining an available capacity for each of a plurality of intervals in the requested future period of time; and

reserving the requested capacity for the requested future period of time if the available capacity for each interval is greater than the requested capacity.

19. The computer readable medium of claim 18, wherein the method further comprises:

sizing a virtual machine for each interval, wherein each virtual machine is allocated to the requested capacity to support user desktop utility demand for the interval.

20. The computer readable medium of claim 18, wherein the method further comprises:

converting multiple resource attributes for the request to a plurality of capacity unit values; and

selecting one capacity unit value from the plurality of the capacity unit values to represent the requested capacity based on whether the selected capacity unit value is to represent the requested capacity or a capacity of a physical machine to support the user desktop utility demand for the request.