On-demand compute environment

• US 7,698,430 B2
• Filed: 03/16/2006
• Issued: 04/13/2010
• 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): Backer, Firmin
• Assistant Examiner(s): DESAI, MARGISHI V
• Application Number:
  US11/276,856
• Publication Number:
  US 20060212334A1
• Time in Patent Office: 1,489 Days
• Field of Search: 370/86, 725/94
• 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

An on-demand compute environment comprises a plurality of nodes within an on-demand compute environment available for provisioning and a slave management module operating on a dedicated node within the on-demand compute environment, wherein upon instructions from a master management module at a local compute environment, the slave management module modifies at least one node of the plurality of nodes.

Claims

1. An on-demand compute environment comprising:

a plurality of compute nodes within an on-demand compute environment available for provisioning and for processing workload; and

a slave management module operating on a compute node within the on-demand compute environment, wherein upon instructions from a master management module at a local compute environment separate from the on-demand compute environment, the local compute environment configured to process end user submitted workload according to a service level agreement, the slave management module creates a new management module on a first of the plurality of compute nodes such that at least one other compute node is provisioned to be managed by the new management module, wherein at least part of the workload submitted to the local compute environment is received by the on-demand compute environment and consumes resources of the provisioned at least one other compute node and wherein the new management module operates in addition to the slave management module and only for a duration of workload that consumes the provisioned at least one other compute node.

2. The on-demand compute environment of claim 1, wherein the the at least one other compute node is provisioned according to an analysis of the at least part of workload that will consume the resources.

3. The on-demand compute environment of claim 1, wherein provisioning the of at least one other compute node further comprises modifying an existing reservation associated with the at least one compute node.

4. The on-demand compute environment of claim 1, wherein the new management module communicates with the master management module in the local compute environment.

5. The on-demand compute environment of claim 1, wherein the modification of provisioning the at least one other compute node further comprises provisioning a resource manager with the new management module to manage the provisioned at least one other compute node.

6. A method of managing an on-demand compute environment using a computing device, the method causing the computing device to perform steps comprising:

receiving an instruction at a slave management module operating on a compute node and associated with an on-demand computing environment from a master management module associated with a local computing environment separate from the on-demand compute environment and configured to process workload;

based on the instruction, creating a new management module on a compute node in the on-demand computing environment and provisioning at least one compute node in the on-demand computing environment, wherein the new management module manages the at least one compute node and communicates with the master management module, wherein at least part of the workload submitted to the local environment is received by the on-demand compute environment and consumes resources of the provisioned at least one compute node of the plurality of compute nodes; and

removing the new management module after the received workload consumes the resources.

7. The method of claim 6, wherein the new management module exists for a duration of workload that consumes the resources of the provisioned at least one compute node.

8. A tangible computer readable medium storing instructions for controlling a computing device to manage an on-demand compute environment, the instructions causing the compute device to perform steps comprising:

receiving an instruction at a slave management module operating on the compute device and associated with an on-demand compute environment from a master management module associated with a local compute environment separate from the on-demand compute environment and configured to process workload;

based on the instruction, creating a new management module on a compute node in the on-demand compute environment and provisioning at least one compute node in the on-demand compute environment, wherein the new management module manages the at least one compute node and communicates with the master management module to receive and process user workload submitted to the local compute environment according to a service level agreement and transferred to the on-demand compute environment; and

removing the new management module after the workload processes the workload in the on-demand compute environment.

9. A method of managing access to an on-demand compute environment using a computing device, the method causing the computing device to perform steps comprising:

receiving workload into a local compute environment having a master management module that manages compute resources in the local compute environment, wherein the compute resources are available to be consumed by user workload submitted to the local compute environment;

identifying a triggering event in the local compute environment;

communicating via the master management module with a management module in a separately administered on-demand compute environment;

transmitting instructions from the master management module to the management module that cause the management module to establish a new management module and provision compute nodes in the on-demand compute environment to be managed by the new management module; and

communicating at least part of the user workload to the provisioned nodes in the on-demand compute environment, wherein at least part of the user workload consumes resources of the provisioned nodes, and wherein the new management module is removed when the workload finishes consuming resources.

10. The method of claim 9, wherein the management module is configured to reserve, allocate, and dynamically provision compute resources in the on-demand compute environment.

11. The method of claim 9, the method further causing the computing device to transmit from the master management module to the management module second instructions for unprovisioning compute nodes in the on-demand compute environment.

12. The method of claim 11, wherein after all provisioned compute nodes are unprovisioned, the management module is removed from the on-demand compute environment.

13. The method of claim 9, wherein the management module dynamically determines when to unprovision compute nodes.

14. The method of claim 12, the method further causing the computing device to perform steps comprising:

receiving updated instructions based on an update event associated with the workload from the master management module to the management module for provisioning additional compute nodes in the on-demand compute environment; and

communicating part of the received workload associated with the update event from the local compute environment to at least one of the provisioned compute nodes and the additional provisioned compute nodes in the on-demand compute environment.

15. The method of claim 9, the method further causing the computing device to transmit configuration instructions from the master management module to the management module for configuring the provisioned compute nodes in the on-demand compute environment.

16. A method of managing an on-demand compute environment using a computing device, the method causing the computing device to perform steps comprising:

receiving a communication from a master node in a separately administered local computer environment;

establishing a slave node in the on-demand compute environment;

receiving instructions from the master node that cause the slave node to provision compute nodes in the on-demand compute environment;

receiving workload submitted into the local compute environment and transferred to the on-demand environment;

communicating the received workload to the provisioned nodes in the on-demand compute environment, wherein the workload consumes resources of the provisioned nodes and de-establishing the slave node after the workload consumes the resources.

17. A method of claim 16, wherein the slave node is configured to reserve, allocate, and dynamically provision compute resources in the on-demand compute environment.

18. A method of claim 16, wherein the method causes the computing device to de-establish the slave node after all provisioned compute nodes are unprovisioned.

19. A method of claim 16, the method further causing the computing device to perform steps comprising:

receiving updated instructions based on an update event associated with the workload from the master node for provisioning additional compute nodes in the on-demand compute environment; and

receiving second workload associated with the update event from the local compute environment to at least one of the provisioned compute nodes and the additional provisioned compute nodes in the on-demand compute environment.

20. A method of claim 16, the method further causing the computing device to perform the step of receiving configuration instructions from the master node for configuring the provisioned compute nodes in the on-demand compute environment.

21. An on-demand compute environment, comprising:

a management node and a plurality of compute nodes managed by the management node;

a module configured to receive at the management node a communication from a master node in a separately administered local computer environment;

a module configured to establish a slave node and provision compute nodes in the on-demand compute environment based on the communication;

a module configured to receive workload submitted into the local compute environment and transferred to the on-demand compute environment;

a module configured to communicate the received workload to the provisioned compute nodes in the on-demand compute environment, wherein the workload consumes resources of the provisioned compute nodes and

a module configured to de-establish the slave node after the workload finishes consuming the resources.

22. The on-demand compute environment of claim 21, wherein the slave node is configured to reserve, allocate, and dynamically provision resources of the plurality of compute nodes in the on-demand compute environment.

23. The on-demand compute environment of claim 21, wherein the the slave node is de-established after the provisioned compute nodes are released.

24. The on-demand compute environment of claim 21, further comprising:

a module configured to receive an update event associated with the workload;

a module configured to receive update instructions based on the update event from the master node for provisioning additional compute nodes of the plurality of compute nodes; and

a module configured to receive update workload associated with the update event from the local compute environment to at least one of the provisioned compute nodes and the additional provisioned compute nodes of the plurality of compute nodes.

25. The on-demand compute environment of claim 21, further comprising:

a module configured to receive configurations and instructions from the master node to the slave node for configuring the provisioned computed nodes in the on-demand compute environment to match workload policies associated with the local compute environment.

26. A tangible computer-readable storage medium storing instructions for controlling a computing device to manage an on-demand compute environment, the instructions causing the compute device to perform steps comprising:

receiving a communication from a master node in a separately administered local compute environment an on-demand compute environment;

establishing a slave node in the on-demand compute environment based on the communication;

receiving instructions from the master node at the slave node to cause the slave node to provision compute nodes in the on-demand compute environment;

receiving workload submitted into the local compute environment and transferred to the on-demand compute environment;

communicating the received workload to the provisioned compute nodes in the on-demand compute environment, wherein the workload consumes resources of the provisioned compute nodes; and

de-establishing the slave node after the workload consumes the resources.

27. A method of managing an on-demand compute environment, the method comprising:

receiving a communication from a master management module in a local compute environment at a first management module in an on-demand compute environment, the on-demand compute environment separately administered from the local compute environment;

based on the communication, provisioning a second management module, in addition to the first management module, on a node in the on-demand compute environment;

provisioning compute nodes in the on-demand compute environment for workload to be managed by the second management module;

receiving workload submitted into the local compute environment and transferred to the on-demand compute environment;

communicating the received workload to the provisioned compute nodes in the on-demand compute environment, wherein the workload consumes resources of the provisioned compute nodes as managed by the second management module; and

un-provisioning the second management module after the workload consumes the resources of the provisioned compute nodes.

28. An on-demand compute environment comprising:

a processor;

a management module in the on-demand compute environment that controls the processor to receive a communication from a master management module in a local compute environment, the on-demand compute environment separately administered from the local compute environment;

a module that controls the processor, based on the communication, to spawn a new management module in the on-demand compute environment, in addition to the management module;

a module that provisions compute nodes in the on-demand compute environment for workload to be managed by the new management module;

a module that receives workload from the local compute environment into the on-demand compute environment for processing;

a module that communicates the received workload to the provisioned compute nodes, wherein the workload consumes resources of the provisioned compute nodes as managed by the new management module; and

a module that removes the new management module from the on-demand compute environment after the workload consumes the resources of the provisioned compute nodes.

29. A method of managing access to an on-demand environment from a local compute environment, the method causing the computing device in the local compute environment to perform steps comprising:

transmitting a communication from a master management module in an on-demand compute environment, the on-demand compute environment separately administered from the local compute environment, the communication causing the on-demand compute environment to provision a new management module, in addition to the on-demand management module, on a node in the on-demand compute environment, the communication further causing the on-demand compute environment to provision compute nodes in the on-demand compute environment for workload to be managed by the new management module; and

transmitting workload submitted into the local compute environment to the on-demand compute environment, wherein the transmitted workload consumes resources of the provisioned compute nodes as managed by the new management module, wherein the new management module is unprovisioned after the workload consumes the resources of the provisioned compute nodes.

30. A local compute environment that accesses an on-demand compute environment, the local compute environment comprising:

a processor;

a master management module in the local compute environment that controls the processor to transmit a communication from the master management module to a management module in the on-demand compute environment, the on-demand compute environment separately administered from the local compute environment, wherein the communication causes the management module to spawn a new management module in the on-demand compute environment in addition to the management module and provision compute nodes in the on-demand compute environment for workload to be managed by the new management module; and

a module that transmits workload submitted into the local compute environment to the on-demand compute environment for processing, wherein the workload consumes resources of the provisioned compute nodes as managed by the new management module, and wherein the new management module is removed from the on-demand compute environment after the workload consumes the resources of the provisioned compute nodes.