Virtual clustering for scalable network control and management

• US 20090089300A1
• Filed: 09/28/2007
• Published: 04/02/2009
• Est. Priority Date: 09/28/2007
• Status: Active Grant

• Current Assignee: Intel Corporation
• Original Assignee: Intel Corporation
• Inventors: Vicente, John
• Granted Patent: US 7,996,510 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: H04L 41/044 comprising hierarchical management structures
• CPC Class Codes: H04L 41/0893 Assignment of logical groups to network elements
• CPC Class Codes: H04L 41/0894 Policy-based network configuration management
• CPC Class Codes: H04L 41/40 using virtualisation of network functions or resources, e.g. SDN or NFV entities

Abstract

In some embodiments, the invention involves a system and method relating to a framework to build a new class of network control (policy) and (state) management services. The framework may be used to build a number of self-management services to support decentralized (mesh) networks. In at least one embodiment, the present invention is intended to hybridize centralized and decentralized network models to generate a more stable and robust network environment using clustering. Other embodiments are described and claimed.

Claims

1. A system for virtual clustering and network control, comprising:

a plurality of computing nodes in a hierarchical network, each node comprising;

logic for autonomous self-promotion and demotion in the network, wherein a promoted node rises to a cluster level above a previous cluster level and becomes part of a virtual cluster at the level above, while remaining in a virtual cluster of the previous level, and wherein a demoted node falls to a cluster level below the previous cluster level and no longer remains in the virtual cluster at the higher level;

logic to provide management and control service functions to cluster k peer nodes in a node that has been promoted to a cluster-head of the cluster k;

logic to provide management and control service functions to other nodes in cluster levels below a current cluster level of the node from which the node has been promoted.

2. The system as recited in claim 1, wherein self-promotion and demotion are based on a node'"'"'s behavior relating to (a) reachability, (b) stability and (c) performance efficiency.

3. The system as recited in claim 2, wherein the logic for autonomous self-promotion and demotion in the network executes in a secure embedded partition on the node.

4. The system as recited in claim 2, wherein the logic for autonomous self-promotion and demotion in the network executes in a virtual machine on the node.

5. The system as recited in claim 2, wherein the logic for autonomous self-promotion and demotion in the network executes in a second processor on the node, wherein the second processor is one of a baseboard management controller and an active management technology controller.

6. The system as recited in claim 2, wherein stability of a node is determined based on an entropy-based statistical model.

7. The system as recited in claim 1, wherein clusters are identified in the network by a cluster address identifying a cluster level and cluster domain such that a cluster at a physical network level, or a lowest level cluster level n, is labeled as, cluster C(n, 0), where 0<

=n<

y, and y is the number of levels in the emergent hierarchy, and a next level higher cluster is labeled C(n−

1, m) where 0<

=m<

y, and y equals a number-clusters at this hierarchy level.

8. The system as recited in claim 7, wherein nodes in a cluster k and cluster-heads associated with a parent cluster of cluster k are configured to use the cluster address for sinking and responding to cluster operational communications, and wherein a global cluster-head uses the cluster address for sourcing operational communications and clustering control and management.

9. The system as recited in claim 8, wherein a cluster-head node communicates to peer cluster-heads at a same cluster level.

10. The system as recited in claim 9, wherein the cluster-head node is to receive cluster state conditions and to perform aggregation and correlation within respective clusters associated with the cluster-head, and wherein the cluster-head is to perform cluster state analysis through statistical computing techniques, and to hold cluster knowledge for state-level assimilation with peer clusters.

11. The system as recited in claim 1, wherein the hierarchical network is a hybrid of both centralized and decentralized network frameworks, centralization to be effected by self-promotion of a node to cluster-head within a virtual cluster in a cluster level and decentralization exists at a lowest cluster level in the network, as well as within each cluster where nodes communicate as if peer-to-peer.

12. A method for virtual clustering and network control in a hierarchical network having a plurality of computing nodes, comprising:

autonomously determining, by a node, whether the node should be promoted, demoted or stay at a same level in the network hierarchy, wherein a promoted node rises to a cluster level above a previous cluster level and becomes pail of a virtual cluster at the level above, while remaining in a virtual cluster of the previous level, and wherein a demoted node falls to a cluster level below the previous cluster level and no longer remains in the virtual cluster at the higher level;

providing management and control service functions to cluster k peer nodes when the node has been promoted to a cluster-head of the cluster k; and

providing management and control service functions to other nodes in cluster levels below a current cluster level of the node from which the node has been promoted.

13. The method as recited in claim 12, wherein self-promotion and demotion are based on the node'"'"'s behavior relating to (a) reachability, (b) stability and (c) performance efficiency.

14. The method as recited in claim 13, wherein the autonomously determining, by a node, executes in one of a secure embedded partition on the node, a virtual machine on the node, or in a second processor on the node, wherein the second processor is one of a baseboard management controller and an active management technology controller.

15. The method as recited in claim 13, wherein stability of a node is determined based on an entropy-based statistical model.

16. The method as recited in claim 12, wherein clusters are identified in the network by a cluster address identifying a cluster level and cluster domain such that a cluster at a physical network level, or a lowest level cluster level n, is labeled as, cluster C(n, 0), where 0<

=n<

y, and y is the number of levels in the emergent hierarchy, and a next level higher cluster is labeled C(n−

1, m) where 0<

=m<

y, and y equals a number clusters at this hierarchy level.

17. The method as recited in claim 16, further comprising:

configuring nodes in a cluster k and cluster-heads associated with a parent cluster of cluster k using the cluster address for sinking and responding to cluster operational communications, and wherein a global cluster-head uses the cluster address for sourcing operational communications and clustering control and management.

18. The system as recited in claim 17, further comprising:

communicating by the cluster-head node to peer cluster-heads at a same cluster level.

19. The method as recited in claim 18, further comprising:

receiving by a cluster-hear node cluster state conditions;

performing aggregation and correlation within respective clusters associated with the cluster-head,

performing, by the cluster-head, analysis through statistical computing techniques; and

holding cluster knowledge for state-level assimilation with peer clusters.

20. The method as recited in claim 12, wherein the hierarchical network is a hybrid of both centralized and decentralized network frameworks, centralization to be effected by self-promotion of a node to cluster-head within a virtual cluster in a cluster level and decentralization exists at a lowest cluster level in the network, as well as within each cluster where nodes communicate as if peer-to-peer.

21. A machine readable storage medium having instructions stored therein, the instructions for virtual clustering and network control in a hierarchical network having a plurality of computing nodes, that when the instructions are executed in a machine, cause the machine to:

autonomously determine, by a node, whether the node should be promoted, demoted or stay at a same level in the network hierarchy, wherein a promoted node rises to a cluster level above a previous cluster level and becomes part of a virtual cluster at the level above, while remaining in a virtual cluster of the previous level, and wherein a demoted node falls to a cluster level below the previous cluster level and no longer remains in the virtual cluster at the higher level;

provide management and control service functions to cluster k peer nodes when the node has been promoted to a cluster-head of the cluster k; and

provide management and control service functions to other nodes in cluster levels below a current cluster level of the node from which the node has been promoted.

22. The medium as recited in claim 21, wherein self-promotion and demotion are based on the node'"'"'s behavior relating to (a) reachability, (b) stability and (c) performance efficiency.

23. The medium as recited in claim 22, wherein the autonomously determining, by a node, executes in one of a secure embedded partition on the node, a virtual machine on the node, or in a second processor on the node, wherein the second processor is one of a baseboard management controller and an active management technology controller.

24. The medium as recited in claim 22, wherein stability of a node is determined based on an entropy-based statistical model.

25. The medium as recited in claim 21, wherein clusters are identified in the network by a cluster address identifying a cluster level and cluster domain such that a cluster at a physical network level, or a lowest level cluster level n, is labeled as, cluster C(n, 0), where 0<

=n<

y, and y is the number of levels in the emergent hierarchy, and a next level higher cluster is labeled C(n−

1, m) where 0<

=m<

y, and y equals a number clusters at this hierarchy level.