Reserving resources in an on-demand compute environment from a local compute environment

• US 8,631,130 B2
• Filed: 03/16/2006
• Issued: 01/14/2014
• Est. Priority Date: 03/16/2005
• Status: Active Grant

• Current Assignee: III Holdings 12, LLC (Intellectual Ventures LLC)
• Original Assignee: Adaptive Computing Enterprises, Inc. (Ala Services LLC)
• Inventors: Jackson, David B.
• Primary Examiner(s): Choudhury, Azizul
• Assistant Examiner(s): DESAI, MARGISHI V
• Application Number:
  US11/276,855
• Publication Number:
  US 20060212333A1
• Time in Patent Office: 2,861 Days
• Field of Search: 709/227, 709/224, 705/52
• US Class Current: 709/226
• CPC Class Codes: G06F 11/2038 with a single idle spare processing component
• CPC Class Codes: G06F 15/173 using an interconnection network, e.g. matrix, shuffle, pyramid, star, snowflake
• CPC Class Codes: G06F 2209/5014 Reservation
• CPC Class Codes: G06F 2209/509 Offload
• CPC Class Codes: G06F 9/5027 the resource being a machine, e.g. CPUs, Servers, Terminals
• CPC Class Codes: G06F 9/5061 Partitioning or combining of resources
• CPC Class Codes: G06F 9/5072 Grid computing
• 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
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
• CPC Class Codes: G08C 17/00 Arrangements for transmitting signals characterised by the use of a wireless electrical link
• CPC Class Codes: H04L 47/70 Admission control; Resource allocation
• CPC Class Codes: H04L 47/80 Actions related to the user profile or the type of traffic
• CPC Class Codes: H04L 47/83 based on usage prediction
• CPC Class Codes: H04L 67/10 in which an application is distributed across nodes in the network software deployment G06F8/60; multiprogramming arrangements G06F9/46
• CPC Class Codes: H04L 67/1001 for accessing one among a plurality of replicated servers
• CPC Class Codes: H04L 67/125 involving control of end-device applications over a network

Abstract

Disclosed is an on-demand system and method for managing resources in an on-demand compute environment from a local compute environment. The method includes receiving information at a local resource broker that is associated with resources within an on-demand compute environment, based on the information, communicating instructions from the local resource broker to the on-demand compute environment and modifying resources associated with the on-demand compute environment based on the instructions.

Claims

1. A method comprising:

receiving a request for compute resources in an on-demand compute environment at a local resource broker, wherein a specification of resources associated with the request is one of fully explicit, partially explicit, fully implicit based on workload, and based on a virtual private cluster package concept where the virtual private cluster package can include aspects of provisioning a support environment and adjustments to resource request timeframes including pre-allocation, allocation duration, and post-allocation timeframe adjustments, and wherein the request being associated with the compute resources in the on-demand compute environment to be used to process workload submitted by a user to a local compute environment, wherein the each on-demand compute environment and the local compute environment are each separately managed environments and each comprises a multinode compute environment used for processing workload and wherein the local resource broker is separate from the local compute environment and from the on-demand compute environment;

based on the request, communicating instructions from the local resource broker to the on-demand compute environment reserving, at a first time, compute resources within the on-demand compute environment, for consumption by workload at a second time later than the first time, to yield reserved compute resources; and

providing instructions for transferring user workload received by the local compute environment to the on-demand compute environment, such that the reserved compute resources in the on-demand compute environment are consumed at the second time by the user workload.

2. The method of claim 1, wherein the request is one of a request for a reservation of compute resources in the on-demand compute environment and a modification of an existing reservation of compute resources in the on-demand compute environment.

3. The method of claim 2, wherein if the request is for a modification of an existing reservation of compute resources, then the step of reserving the compute resources within the chosen on-demand compute environment further comprises modifying the existing reservation of compute resources.

4. The method of claim 3, wherein modifying the existing reservation of compute resources comprises one of expanding, reducing or canceling the existing reservation of compute resources.

5. The method of claim 1, wherein reserving the compute resources further comprises adjusting a reservation duration to accommodate for provisioning overhead associated with a reservation in the on-demand compute environment.

6. The method of claim 1, wherein the on-demand compute environment has an associated slave management module and the local compute environment has an associated master management module.

7. The method of claim 1, wherein the local resource broker locally creates a reservation for at least one aspect of a job to consume the compute resources in the on-demand compute environment.

8. The method of claim 1, wherein the received information at a local resource broker relates to resource requirement information based on user specification, a current workload and a predicted workload.

9. The method of claim 1, wherein reserving the compute resources further comprises modifying compute resources for provisioning by customizing a delivered compute environment within the on-demand compute environment.

10. The method of claim 1, wherein reserving the compute resources further comprises co-allocating resources including at least two of a compute, network, storage, license, or service resources as part of a reservation.

11. The method of claim 1, wherein reserving the compute resources further comprises co-allocating resources over disjoint timeframes to improve availability and utilization of resources.

12. A comprising:

a processor; and

a non-transitory computer-readable storage medium storing instructions for controlling the processor to perform steps comprising;

receiving a request for compute resources in an on-demand compute environment at a local resource broker, wherein a specification of resources associated with the request is one of fully explicit, partially explicit, fully implicit based on workload, and based on a virtual private cluster package concept where the virtual private cluster package can include aspects of provisioning a support environment and adjustments to resource request timeframes including pre-allocation, allocation duration, and post-allocation timeframe adjustments, and wherein the request being associated with the compute resources in the on-demand compute environment to be used to process workload submitted by a user to a local compute environment, wherein the on-demand compute environment and the local compute environment are each separately managed environments and each comprises a multinode compute environment used for processing workload and wherein the local resource broker is separate from the local compute environment and from the on-demand compute environment;

based on the request, communicating instructions from the local resource broker to the on-demand compute environment reserving, at a first time, compute resources within the on-demand compute environment, for consumption by workload at a second time later than the first time, to yield reserved compute resources; and

providing instructions for transferring user workload received by the local compute environment to the on-demand compute environment, such that the reserved compute resources in the on-demand compute environment are consumed at the second time by the user workload.

13. A non-transitory computer-readable storage medium storing instructions which, when executed by a computing device, control the computing device to manage resources in an on-demand compute environment, the instructions comprising:

receiving a request for compute resources in an on-demand compute environment at a local resource broker, wherein a specification of resources associated with the request is one of fully explicit, partially explicit, fully implicit based on workload, and based on a virtual private cluster package concept where the virtual private cluster package can include aspects of provisioning a support environment and adjustments to resource request timeframes including pre-allocation, allocation duration, and post-allocation timeframe adjustments, and wherein the request being associated with the compute resources in the on-demand compute environment to be used to process workload submitted by a user to a local compute environment, wherein the on-demand compute environment and the local compute environment are each separately managed environments and each comprises a multinode compute environment used for processing workload and wherein the local resource broker is separate from the local compute environment and from the on-demand compute environment;

based on the request, communicating instructions from the local resource broker to the chosen on-demand compute environment reserving, at a first time, compute resources within the on-demand compute environment, for consumption by workload at a second time later than the first time, to yield reserved compute resources; and

providing instructions for transferring user workload received by the local compute environment to the on-demand compute environment, such that the reserved compute resources in the on-demand compute environment are consumed at the second time by the user workload.

14. The non-transitory computer-readable storage medium of claim 13, wherein the request for a reservation of compute resources in the on-demand compute environment or a modification of an existing reservation of compute resources in the on-demand compute environment.

15. The computer-readable storage medium of claim 14, wherein if the request is for a modification of an existing reservation of compute resources, then the step of modifying the compute resources associated with the on-demand compute environment further comprises modifying the existing reservation of compute resources.

16. The computer-readable storage medium of claim 15, wherein modifying the existing reservation of compute resources comprises one of expanding, reducing or canceling the existing reservation of compute resources.

17. A method of managing resources, the method comprising:

receiving instructions from a local resource broker, wherein the instructions are associated with compute resources to be used to process user workload, wherein the local broker, an on-demand compute environment and a local compute environment are each separate environments, wherein a specification of resources associated with the instructions are one of fully explicit, partially explicit, fully implicit based on workload, and based on a virtual private cluster package concept where the virtual private cluster package can include aspects of provisioning a support environment and adjustments to resource request timeframes including pre-allocation, allocation duration, and post-allocation timeframe adjustments, and wherein;

reserving, at a first time, compute resources in the on-demand compute environment based on the instructions to yield reserved compute resources for use at a second time later than the first time; and

receiving a transfer of the user workload from the local compute environment to the on-demand compute environment such that the user workload consumes the reserved compute resources in the on-demand compute environment at the second time, the user workload having been submitted to the local compute environment.