System and method of providing a self-optimizing reservation in space of compute resources

• US 9,268,607 B2
• Filed: 03/11/2005
• Issued: 02/23/2016
• Est. Priority Date: 03/13/2004
• Status: Active Grant

• Current Assignee: III Holdings 12, LLC (Intellectual Ventures LLC)
• Original Assignee: Adaptive Computing Enterprises, Inc. (Ala Services LLC)
• Inventors: Jackson, David Brian
• Primary Examiner(s): Lee, Adam
• Application Number:
  US10/530,577
• Publication Number:
  US 20150220364A1
• Time in Patent Office: 4,001 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06F 15/161 Computing infrastructure, e.g. computer clusters, blade chassis or hardware partitioning casings, cabinets, racks or drawers for data centers H05K5/00
• CPC Class Codes: G06F 2209/5014 Reservation
• CPC Class Codes: G06F 2209/5015 Service provider selection
• CPC Class Codes: G06F 2209/5022 Workload threshold
• CPC Class Codes: G06F 2209/503 Resource availability
• CPC Class Codes: G06F 2209/506 Constraint
• CPC Class Codes: G06F 2209/508 Monitor
• CPC Class Codes: G06F 9/4881 Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues
• CPC Class Codes: G06F 9/5011 the resources being hardware resources other than CPUs, Servers and Terminals
• CPC Class Codes: G06F 9/5027 the resource being a machine, e.g. CPUs, Servers, Terminals
• CPC Class Codes: G06F 9/505 considering the load
• CPC Class Codes: G06F 9/5072 Grid computing

Abstract

A system and method of dynamically controlling a reservation of compute resources within a compute environment is disclosed. The method aspect of the invention comprises receiving a request from a requestor for a reservation of resources within the compute environment, reserving a first group of resources, evaluating resources within the compute environment to determine if a more efficient use of the compute environment is available and if a more efficient use of the compute environment is available, then canceling the reservation for the first group of resources and reserving a second group of resources of the compute environment according to the evaluation.

Claims

1. A method of dynamically controlling a reservation of compute resources within a compute environment, the method comprising:

receiving a request from a requestor for the reservation of compute resources within the compute environment for processing workload;

reserving a first group of compute resources to yield the reservation, the first group of compute resources having a first value, and wherein the first value is used to determine a threshold value;

evaluating compute resources within the compute environment to determine a second value for a second group of compute resources;

determining whether the reservation can be improved by comparing the second value of the second group of resources with the threshold value to yield a determination; and

if the determination indicates that the second value of the second group of compute resources is an improvement over the threshold value, then modifying the reservation from the first group of compute resources to the second group of compute resources of the compute environment.

2. The method of claim 1, wherein the compute environment is one of an enterprise compute farm, a cluster and a grid.

3. The method of claim 1, wherein the request further comprises a preferred criteria and a required criteria.

4. The method of claim 3, wherein the evaluating further comprises identifying compute resources that are available and meet explicit or implicit preferred criteria of the reservation.

5. The method of claim 4, wherein available compute resources comprise compute resources in an up state with no partial or complete failure.

6. The method of claim 3, wherein the reservation of the first group of compute resources meets the required criteria.

7. The method of claim 6, wherein evaluating compute resources within the cluster environment further comprises determining if at least one of the preferred criteria can be met by modifying the first group of compute resources allocated for the reservation.

8. The method of claim 7, wherein the evaluating includes a comparison of a cost of migrating the reservation from the first group of compute resources to the second group of compute resources with an improved resource allocation gained from meeting at least one of the preferred criteria.

9. The method of claim 8, wherein the reservation is not modified if the cost of modifying the reservation of first group of compute resources to the second group of compute resources is equal to or higher than either the threshold value or the improved resource allocation of the compute environment gained by meeting at least one of the preferred criteria.

10. The method of claim 8, wherein the evaluating further uses a per-reservation policy.

11. The method of claim 10, wherein the per-reservation policy is at least one of an administrator policy, a user-based policy, a policy of never taking an action, a policy of always taking an action and a cost-based policy.

12. The method of claim 1, wherein modifying the reservation further comprises at least one of canceling the reservation and creating a new reservation, dynamically modifying attributes of an existing reservation, and dynamically modifying attributes of the first group of compute resources to better satisfy the reservation.

13. The method of claim 1, wherein the reservation for the first group of compute resources and the reservation for the second group of compute resources overlap.

14. The method of claim 1, wherein the request is identified as a self-optimizing request.

15. The method of claim 14, wherein the requestor identifies the request as self-optimizing.

16. The method of claim 14, wherein the request is identified as self-optimizing in space.

17. The method of claim 14, wherein the requestor is charged more for a self-optimizing request relative to a charge for a non-self-optimizing request.

18. A system for dynamically controlling a reservation of compute resources within a compute environment, the system comprising:

a processor; and

a computer-readable storage medium storing instructions which, when executed by the processor, cause the processor to perform operations comprising;

receiving a request from a requestor for the reservation of compute resources within the compute environment for processing workload;

reserving a first group of compute resources to yield the reservation, the first group of compute resources having a first value, and wherein the first value is used to determine a threshold value;

evaluating compute resources within the compute environment to determine a second value for a second group of compute resources;

determining whether the reservation can be improved by comparing the second value of the second group of resources with the threshold value to yield a determination; and

if the determination indicates that the second value of the second group of compute resources is an improvement over the threshold value, then modifying the reservation from the first group of compute resources to the second group of compute resources of the compute environment.

19. A non-transitory computer-readable medium storing instructions which, when executed by a computing device to dynamically manage a reservation of compute resources within a compute environment, cause the computing device to perform operations comprising:

receiving a request from a requestor for the reservation of compute resources within the compute environment for processing workload;

reserving a first group of compute resources to yield the reservation, the first group of compute resources having a first value, and wherein the first value is used to determine a threshold value;

evaluating compute resources within the compute environment to determine a second value for a second group of compute resources;

determining whether the reservation can be improved by comparing the second value of the second group of resources with the threshold value to yield a determination; and

if the determination indicates that the second value of the second group of compute resources is an improvement over the threshold value, then modifying the reservation from the first group of compute resources to the second group of compute resources of the compute environment.