System and method of using transaction IDS for managing reservations of compute resources within a compute environment

• US 8,321,871 B1
• Filed: 06/17/2005
• Issued: 11/27/2012
• Est. Priority Date: 06/18/2004
• 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): An, Meng
• Assistant Examiner(s): LEE, TAMMY EUNHYE
• Application Number:
  US11/155,090
• Time in Patent Office: 2,720 Days
• Field of Search: 718/100, 718/102, 718/104, 709/226, 705/8
• US Class Current: 718/104
• CPC Class Codes: G06F 2209/5014 Reservation
• CPC Class Codes: G06F 2209/504 Resource capping
• CPC Class Codes: G06F 9/4881 Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues
• CPC Class Codes: G06F 9/50 Allocation of resources, e.g. of the central processing unit [CPU]
• CPC Class Codes: G06F 9/5005 to service a request
• CPC Class Codes: G06F 9/5011 the resources being hardware resources other than CPUs, Servers and Terminals
• CPC Class Codes: G06F 9/5022 Mechanisms to release resources
• CPC Class Codes: G06F 9/5027 the resource being a machine, e.g. CPUs, Servers, Terminals
• CPC Class Codes: G06F 9/5072 Grid computing
• CPC Class Codes: H04L 41/00 Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
• CPC Class Codes: H04L 41/50 Network service management, e.g. ensuring proper service fulfilment according to agreements
• CPC Class Codes: Y02D 10/00 Energy efficient computing, e.g. low power processors, power management or thermal management

Abstract

A system and method for reserving resources within a compute environment such as a cluster or grid are disclosed. The method aspect of the disclosure includes receiving a request for resource availability in a compute environment from a requestor, associating a transaction identification with the request and resources within the compute environment that can meet the request and presenting the transaction identification to the requestor. The transaction ID can also be associated with a time frame in which resources are available and can also be associated with modifications to the resources and supersets of resources that could be drawn upon to meet the request. The transaction ID can also be associated with metrics that identify how well the resource fit with the request and modifications that can make the resources better match the workload which would be submitted under the request.

Claims

1. A method of managing resources within a multi-node compute environment, the method comprising:

receiving a request for compute resource availability in the multi-node compute environment managed by a common workload manager, wherein the common workload manager reserves compute resources at a first time for jobs to consume the compute resources at a second time that is later than the first time;

associating, via a processor, a transaction identification with available compute resources that can meet the request via a reservation of the available compute resources;

presenting the transaction identification to a requestor of the request, wherein associating the transaction identification and presenting the transaction identification occur prior to reserving the available compute resources; and

reserving the available compute resources associated with the transaction identification to yield reserved compute resources and provisioning software associated with the request on the reserved compute resources by automatically loading the software on nodes within the multi-node compute environment.

2. The method of claim 1, further comprising presenting information associated with the available compute resources that can meet the request further comprises presenting information associated with probabilities of availability associated with the available compute resources.

3. The method of claim 2, further comprising, after presenting the information and presenting associated transaction identification:

receiving instructions from the requestor to commit the request for the available compute resources.

4. The method of claim 3, wherein receiving instructions to commit the request for available compute resources is accomplished with reference to the transaction identification.

5. The method of claim 2, wherein the presented information further comprises at least one time period in which the available compute resources can meet the request.

6. The method of claim 5, wherein the transaction identification is associated with both the available compute resources and at least one time period in which the available compute resources can meet the request.

7. The method of claim 6, wherein receiving instructions to commit the request for the available complete resources further comprises receiving instructions to reserve the available compute resources for the at least one time period at which the available compute resources can meet the request.

8. The method of claim 1, further comprising presenting an option to modify the compute resources.

9. The method of claim 8, wherein if the requestor selects the option to modify the available compute resources, the method further comprises:

receiving instructions to modify at least one compute resource of the available compute resources.

10. The method of claim 9, wherein receiving instructions to modify at least one compute resource further comprises receiving instructions to modify at least one of the following parameters associated with a reservation of the available compute resources associated with the transaction identification:

disk space, processor speed, number of nodes reserved, memory allocation, bandwidth and licenses.

11. The method of claim 9, wherein receiving instructions to modify at least one compute resource further comprises receiving instructions to perform provisioning of compute resources.

12. The method of claim 11, wherein the instruction to perform provisioning of compute resources further comprises changing an operating system on nodes within the multi-node compute environment.

13. The method of claim 2, wherein presenting information associated with the available compute resources that can meet the request further comprises presenting information associated with modified compute resources that, if so modified, can meet the request in a given time frame.

14. The method of claim 13, wherein the transaction identification reflects a potential configuration of the available compute resources.

15. The method of claim 1, further comprising:

analyzing the available compute resources in the multi-node compute environment and identifying at least one time frame when the available compute resources are available to meet the request, wherein;

associating a transaction identification with the request and the available compute resources within the multi-node compute environment that can meet the request further comprises associating a transaction identification with each at least one time frame when the available compute resources are available to meet the request.

16. The method of claim 15, further comprising:

comparing potential modifications of the available compute resources in the multi-node compute environment to a modification threshold; and

if no modifications of the available compute resources are necessary according to the comparison with the modification threshold, then associating the transaction identification with the available compute resources at the at least one time frame; and

if modifications to the available compute resources are necessary according to the comparison with the modification threshold, then associating the transaction identification with the available compute resources as modified and the at least one time frame.

17. The method of claim 16, wherein the transaction identification is associated with a metric which relates to how closely the available compute resources match the request.

18. The method of claim 17, wherein the metric relates to at least one of:

response time, overhead required to modify resources, and ownership of resources.

19. The method of claim 13, further comprising:

presenting the requestor with an option to commit a reservation of the available compute resources using the at least one transaction identification.

20. The method of claim 19, further comprising:

presenting an option to reference resources by the at least one transaction identification when the requestor commits a reservation.

21. The method of claim 19, further comprising:

receiving from the requestor an indication to commit the reservation based on the transaction identification.

22. The method of claim 15, wherein a plurality of transaction identifications is generated and wherein the requestor can commit multiple reservations referencing multiple transaction identifications of the plurality of transaction identifications.

23. The method of claim 15, wherein the at least one transaction identification further comprises information about modifications to the available compute resources in the multi-node compute environment to be made in order to process workload according to the request.

24. The method of claim 23, wherein if the requestor commits the reservation based on the at least one transaction identification the method comprising reserving the available compute resources to yield reserved computer resources and modifying the reserved compute resources for processing the workload.

25. The method of claim 13, where presenting the requestor with an option to commit the reservation further comprises presenting the requestor with at least one option to commit the reservation without modifications to the multi-node compute environment and presenting the requestor with at least one option to commit reservations with modifications to the multi-node compute environment.

26. A method of using transaction identifications, the method comprising:

receiving a request from a requestor for compute resource availability in a multi-node compute environment managed by a common workload manager, wherein the common workload manager reserves compute resources at a current time for jobs to consume the compute resources at a future time relative to the current time;

analyzing, via a processor, compute resources in the multi-node compute environment to determine, for the request, what compute resources are available to yield available compute resources and at least one time frame at which the available compute resources are available;

analyzing at least one superset of the available compute resources for the request and at least one time frame associated with each at least one superset of available compute resources;

generating a transaction identification associated with each of (1) the available compute resources and each at least one time frame and (2) each at least one superset of available compute resources with its associated at least one time frame, wherein generating the transaction identification occurs prior to a reservation of the available compute resources; and

reserving the available compute resources associated with the transaction identification to yield reserved compute resources and provisioning software associated with the request on the reserved compute resources by automatically loading the software on nodes within the multi-node compute environment.

27. The method of claim 26, further comprising, presenting each generated transaction identification to the requestor prior to reserving the available compute resources.

28. A system for managing resources within a multi-node compute environment, the system comprising:

a processor; and

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

receiving a request for compute resource availability in the multi-node compute environment managed by a common workload manager, wherein the common workload manager reserves compute resources at a first time for lobs to consume the compute resources at a second time that is later than the first time;

associating, via a processor, a transaction identification with available compute resources that can meet the request via a reservation of the available compute resources;

presenting the transaction identification to a requestor of the request, wherein associating the transaction identification and presenting the transaction identification occur prior to reserving the available compute resources; and

reserving the available compute resources associated with the transaction identification to yield reserved compute resources and provisioning software associated with the request on the reserved compute resources by automatically loading the software on nodes within the multi-node compute environment.

29. The system of claim 28, wherein the computer readable medium stores further instructions which result in the method further comprising:

analyzing the compute resources in the multi-node compute environment and identify at least one time frame when the compute resources are available to meet the request, and wherein the second module is configured to control the processor to further associate the transaction identification with each at least one time frame when the compute resources are available to meet the request.

30. A non-transitory computer-readable medium storing instructions for controlling a computing device to managing resources within a multi-node compute environment, the instructions causing the computing device to perform a method comprising:

receiving a request for compute resource availability in the multi-node compute environment managed by a common workload manager, wherein the common workload manager reserves compute resources at a first time for jobs to consume the compute resources at a second time that is later than the first time;

associating, via a processor, a transaction identification with the available compute resources within the multi-node compute environment that can meet the request via a reservation of the available compute resources;

presenting the transaction identification to a requestor of the request, wherein associating the transaction identification and presenting the transaction identification occur prior to reserving the available compute resources; and

reserving the available compute resources associated with the transaction identification to yield reserved compute resources and provisioning software associated with the request on the reserved compute resources by automatically loading the software on nodes within the multi-node compute environment.

31. The non-transitory computer-readable medium of claim 30, wherein the non-transitory computer-readable medium stores further instructions which result in the method further comprising:

analyzing the compute resources in the multi-node compute environment and identifying at least one time frame when the available compute resources are available to meet the request, wherein associating the transaction identification with the available compute resources further comprises associating the transaction identification with each at least one time frame when the available compute resources are available to meet the request.