Distributed computing based on multiple nodes with determined capacity selectively joining resource groups having resource requirements

• US 20060179106A1
• Filed: 02/10/2005
• Published: 08/10/2006
• Est. Priority Date: 02/10/2005
• Status: Active Grant

• Current Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Original Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Inventors: Toebes, John, Walker, Douglas Jay, Turner, Bryan C.
• Granted Patent: US 8,051,170 B2
• Time in Patent Office: Days
• US Class Current: 709/201
• CPC Class Codes: G06F 2209/5011 Pool
• CPC Class Codes: G06F 2209/505 Clust
• CPC Class Codes: G06F 9/5072 Grid computing

Abstract

A network provides distributed computing services based on participation in respective resource groups by computing nodes, each resource group including a corresponding resource requirement for any computing node that joins the corresponding resource group for execution of the corresponding distributed computing service. Each computing node, in response to determining its corresponding available node capacity, is configured for selectively creating and joining at least one new resource group for execution of a corresponding distributed computing service having a corresponding resource requirement, and/or selectively joining at least one of the available resource groups, based on the corresponding available node capacity satisfying the corresponding resource requirement. Each computing node also is configured for selectively leaving any one of the joined resource groups based on determined conditions. Hence, each computing node may choose to participate in as many resource groups as needed for respective computing services, based on the corresponding available node capacity.

Claims

1. A method in a computing node, comprising:

determining an available node capacity for the computing node relative to any resource groups having been joined by the computing node, each resource group having a corresponding resource requirement for providing a corresponding distributed computing service within a network;

selectively creating and joining a first of the resource groups based at least on determining that the corresponding resource requirement is less than the available node capacity;

selectively joining, within the network, a second of the resource groups based on the corresponding resource requirement of the second resource group being less than the available node capacity; and

executing the distributed computing services for the resource groups having been joined by the computing node, according to the respective resource requirements.

2. The method of claim 1, wherein the determining includes determining the available node capacity relative to at least any one of available data storage space, available network bandwidth, available transaction capacity, and available CPU resources.

3. The method of claim 2, wherein one of the distributed computing services includes replicating data, the executing step including replicating a data element to each of other computing nodes having joined the corresponding resource group, at least one of the other computing nodes having a corresponding available node capacity that is at least an order of magnitude different than the available node capacity of the computing node.

4. The method of claim 3, wherein the order of magnitude provides a difference of at least ten (10) times between the available node capacity of the computing node and the available node capacity of the second computing node.

5. The method of claim 1, further comprising:

determining whether the first and second resource groups, having been joined by the computing node, should be merged into a merged resource group;

arbitrating among other computing nodes having joined at least one of the first and second resource groups, for merging the first and second resource groups into the merged resource group; and

merging the first and second resource groups into the merged resource group.

6. The method of claim 1, wherein the steps of selectively creating and joining the first resource group, and selectively joining the second resource group, each include reserving a corresponding reserved capacity for the corresponding resource group from the available node capacity, the method further comprising:

detecting a need for a change in the resource requirement for a corresponding one of the resource groups having been joined by the computing node, based on any one of an added distributed computing service relative to the corresponding distributed computing service, an added data type attribute relative to an existing data attribute of the one resource group, and determining that the resource requirement is insufficient;

arbitrating, between other computing nodes having joined the one resource group, for dividing the one resource group into first and second divided resource groups having respective first and second divided resource requirements generated based on the change; and

selectively joining at least one of the first and second divided resource groups based on the change and a determined attribute of the corresponding one divided resource group having been joined.

7. A method in a network, the method comprising:

providing an identification of resource groups, each resource group having a corresponding resource requirement for providing a corresponding distributed computing service within the network;

in each computing node of the network;

(1) determining a corresponding available node capacity relative to any of said resource groups having been joined by the computing node,

(2) selectively joining at least one of the resource groups based on the corresponding resource requirement being less than the available node capacity, and

(3) executing the distributed computing services for the resource groups having been joined by the computing node, according to the respective resource requirements;

identifying the computing nodes having joined the resource groups; and

providing connectivity for a user node, requesting one of the distributed computing services, to one of the computing nodes having joined the corresponding resource group.

8. The method of claim 7, wherein the determining in each computing node includes determining the corresponding available node capacity relative to at least any one of available data storage space, available network bandwidth, available transaction capacity, and available CPU resources.

9. The method of claim 8, wherein one of the distributed computing services includes replicating data, the executing step for the computing nodes having joined the one resource group corresponding to the one distributed computing service including replicating a data element among the computing nodes having joined the one resource group.

10. The method of claim 9, wherein at least one of the computing nodes has a corresponding available node capacity that is at least an order of magnitude of at least ten (10) times different than the available node capacity of other of the computing nodes.

11. The method of claim 7, further comprising:

determining by one of the computing nodes whether first and second of the resource groups having been joined by the one computing node should be merged into a merged resource group;

arbitrating among other of the computing nodes having joined the first and second resource groups, for merging the first and second resource groups into the merged resource group; and

merging the first and second resource groups into the merged resource group.

12. The method of claim 7, further comprising dividing one of the resource groups by the computing nodes having joined the one resource group, based on:

detecting in at least one of the computing nodes having joined the one resource group a need for a change in the corresponding resource requirement, based on any one of an added distributed computing service, an added data type attribute, and determining that the resource requirement is insufficient;

arbitrating, between the computing nodes having joined the one resource group, for dividing the one resource group into first and second divided resource groups having respective first and second divided resource requirements generated based on the change; and

each of the computing nodes having joined the one resource group selectively joining at least one of the first and second divided resource groups based on the change and a determined attribute of the corresponding one divided resource group having been joined.

13. A computing node comprising:

a resource monitor configured for determining an available node capacity for the computing node relative to any resource groups having been joined by the computing node, each resource group having a corresponding resource requirement for providing a corresponding distributed computing service within a network;

a resource group arbitration module configured for;

(1) selectively creating and joining a first of the resource groups based at least on determining that the corresponding resource requirement is less than the available node capacity, and

(2) selectively joining, within the network, a second of the resource groups based on the corresponding resource requirement of the second resource group being less than the available node capacity; and

computing node resources configured for executing the distributed computing services for the resource groups having been joined by the computing node, according to the respective resource requirements, the resource monitor determining the available node capacity based on consumption of the computing node resources for the executed distributed computing services.

14. The computing node of claim 13, wherein the computing node resources includes data storage having available data storage space, a network interface having available network bandwidth, and a processor having available CPU resources and available transaction capacity, the resource monitor configured for determining the available node capacity relative to at least any one of the available data storage space, the available network bandwidth, the available transaction capacity, and the available CPU resources.

15. The computing node of claim 14, wherein one of the distributed computing services includes replicating data, the computing node resources configured for replicating a data element to each of other computing nodes having joined the corresponding resource group, at least one of the other computing nodes having a corresponding available node capacity that is at least an order of magnitude different than the available node capacity of the computing node.

16. The computing node of claim 15, wherein the order of magnitude provides a difference of at least ten (10) times between the available node capacity of the computing node and the available node capacity of the second computing node.

17. The computing node of claim 13, wherein the resource group arbitration module further is configured for:

determining whether the first and second resource groups, having been joined by the computing node, should be merged into a merged resource group;

arbitrating among other computing nodes having joined at least one of the first and second resource groups, for merging the first and second resource groups into the merged resource group; and

merging the first and second resource groups into the merged resource group.

18. The computing node of claim 13, wherein the resource group arbitration module is configured for:

reserving a corresponding reserved capacity from the available node capacity for each joined resource group;

detecting a need for a change in the resource requirement for a corresponding one of the resource groups having been joined by the computing node, based on any one of an added distributed computing service relative to the corresponding distributed computing service, an added data type attribute relative to an existing data attribute of the one resource group, and determining that the resource requirement is insufficient;

arbitrating, between other computing nodes having joined the one resource group, for dividing the one resource group into first and second divided resource groups having respective first and second divided resource requirements generated based on the change; and

selectively joining at least one of the first and second divided resource groups based on the change and a determined attribute of the corresponding one divided resource group having been joined.

19. A network comprising:

a plurality of computing nodes configured for selectively joining resource groups, each resource group having a corresponding resource requirement for providing a corresponding distributed computing service within the network;

each computing node configured for;

(1) determining a corresponding available node capacity relative to any of said resource groups having been joined by the computing node,

(2) selectively joining at least one of the resource groups based on the corresponding resource requirement of the second resource group being less than the available node capacity, and

(3) executing the distributed computing services for the resource groups having been joined by the computing node, according to the respective resource requirements;

the computing nodes providing distributed computing services to client nodes based on the respective joined resource groups.

20. The network of claim 19, wherein each computing node is configured for determining the corresponding available node capacity relative to at least any one of available data storage space, available network bandwidth, available transaction capacity, and available CPU resources.

21. The network of claim 20, wherein one of the distributed computing services includes replicating data, the computing nodes having joined the one resource group corresponding to the one distributed computing service replicating a data element among the computing nodes having joined the one resource group.

22. The network of claim 21, wherein at least one of the computing nodes has a corresponding available node capacity that is at least an order of magnitude of at least ten (10) times different than the available node capacity of other of the computing nodes.

23. The network of claim 21, each computing node further is configured for:

determining whether a corresponding first and second of the resource groups having been joined should be merged into a merged resource group;

arbitrating among other of the computing nodes having joined the first and second resource groups, for merging the first and second resource groups into the merged resource group; and

merging the first and second resource groups into the merged resource group.

24. The network of claim 19, wherein the computing nodes having joined one of the resource groups are configured for dividing the one resource group, based on:

detecting in at least one of the computing nodes having joined the one resource group a need for a change in the corresponding resource requirement, based on any one of an added distributed computing service, an added data type attribute, and determining that the resource requirement is insufficient;

arbitrating, between the computing nodes having joined the one resource group, for dividing the one resource group into first and second divided resource groups having respective first and second divided resource requirements generated based on the change; and

each of the computing nodes having joined the one resource group selectively joining at least one of the first and second divided resource groups based on the change and a determined attribute of the corresponding one divided resource group having been joined.

25. A computing node comprising:

means for determining an available node capacity for the computing node relative to any resource groups having been joined by the computing node, each resource group having a corresponding resource requirement for providing a corresponding distributed computing service within a network;

means for selectively creating and joining a first of the resource groups based at least on determining that the corresponding resource requirement is less than the available node capacity, the means for selectively creating and joining further configured for selectively joining, within the network, a second of the resource groups based on the corresponding resource requirement of the second resource group being less than the available node capacity; and

means for executing the distributed computing services for the resource groups having been joined by the computing node, according to the respective resource requirements.

26. The computing node of claim 25, wherein the means for determining is configured for determining the available node capacity, relative to resources utilized by the means for executing, based on at least any one of available data storage space, available network bandwidth, available transaction capacity, and available CPU resources.

27. The computing node of claim 26, wherein one of the distributed computing services includes replicating data, the means for executing configured for replicating a data element to each of other computing nodes having joined the corresponding resource group, at least one of the other computing nodes having a corresponding available node capacity that is at least an order of magnitude different than the available node capacity of the computing node.

28. The computing node of claim 27, wherein the order of magnitude provides a difference of at least ten (10) times between the available node capacity of the computing node and the available node capacity of the second computing node.

29. The computing node of claim 25, wherein the means for selectively creating and joining further is configured for:

determining whether the first and second resource groups, having been joined by the computing node, should be merged into a merged resource group;

arbitrating among other computing nodes having joined at least one of the first and second resource groups, for merging the first and second resource groups into the merged resource group; and

merging the first and second resource groups into the merged resource group.

30. The computing node of claim 25, wherein the means for selectively creating and joining is configured for reserving from the available node capacity a corresponding reserved capacity for each corresponding resource group having been joined, the means for selectively creating and joining further configured for:

detecting a need for a change in the resource requirement for a corresponding one of the resource groups having been joined by the computing node, based on any one of an added distributed computing service relative to the corresponding distributed computing service, an added data type attribute relative to an existing data attribute of the one resource group, and determining that the resource requirement is insufficient;

arbitrating, between other computing nodes having joined the one resource group, for dividing the one resource group into first and second divided resource groups having respective first and second divided resource requirements generated based on the change; and

selectively joining at least one of the first and second divided resource groups based on the change and a determined attribute of the corresponding one divided resource group having been joined.