Data center capability summarization

Patent metadata

Field	Value
Publication number	US-9747362-B2
Application number	US-201514678244-A
Country	US
Kind code	B2
Filing date	Apr 3, 2015
Priority date	Sep 16, 2011
Publication date	Aug 29, 2017
Grant date	Aug 29, 2017

Abstract

A method for summarizing capabilities in a hierarchically arranged data center includes receiving capabilities information, wherein the capabilities information is representative of capabilities of respective nodes at a first hierarchical level in the hierarchically arranged data center, clustering nodes based on groups of capabilities information, generating a histogram that represents individual node clusters, and sending the histogram to a next higher level in the hierarchically arranged data center. Relative rankings of capabilities may be used to order a sequence of clustering operations.

First claim

What is claimed is:

1. A method comprising:
receiving, via an electronic network, at a capabilities database, capabilities information, the capabilities information being representative of capabilities of respective nodes at a first hierarchical level in a hierarchically arranged data center, each of the respective nodes comprising a plurality of devices;
clustering the nodes by summarizing the capabilities information, resulting in node clusters, wherein each node cluster is represented by:
a plurality of capabilities, one capability of the plurality of capabilities being indicated by a pair of values that includes a minimum numeric value and a maximum numeric value, and
a count of devices in the cluster; and

sending, via the electronic network, information representative of the node clusters to a next higher level in the hierarchically arranged data center, wherein
the information representative of the node clusters is expanded by uniformly assigning, to a number of devices equal to the count of devices, values between the minimum numeric value and the maximum numeric value.

2. The method of claim 1 , wherein the capabilities information comprises attributes of the capabilities and wherein clustering comprises clustering based on a relative ranking of the attributes.

3. The method of claim 2 , further comprising clustering based on a capability having a first ranking, and then clustering based on a capability having a second ranking higher than the first ranking.

4. The method of claim 1 , wherein the nodes include at least one of a load balancer, firewall or computing device.

5. The method of claim 1 , wherein the nodes comprise a plurality of collections of devices, including at least load balancers, firewalls or computing devices.

6. The method of claim 1 , wherein the capabilities information comprises non-numeric data, and clustering further comprises determining a hamming distance based on the non-numeric data and clustering also comprises clustering based on the hamming distance.

7. The method of claim 1 , wherein clustering comprises clustering the nodes based on capabilities represented by numeric data and capabilities represented by non-numeric data.

8. The method of claim 1 , further comprising enabling a user to control at least one of a value that represents the number of clusters to be formed or relative rankings of capabilities in the capabilities information.

9. The method of claim 1 , further comprising clustering by using k-means, including clustering a given node in a first cluster when the given node is deemed to be closer to a centroid of the first cluster than to a centroid of a second cluster.

10. An apparatus comprising:
a processor;
a memory in communication with the processor; and
a network interface unit in communication with the processor and memory,
wherein the processor is configured, according to instructions stored in the memory, to:
receive, via the network interface unit, capabilities information, the capabilities information being representative of capabilities of respective nodes at a first hierarchical level in a hierarchically arranged data center, each of the respective nodes comprising a plurality of devices;
cluster the nodes by summarizing the capabilities information, resulting in node clusters, wherein each node cluster is represented by:
a plurality of capabilities, one capability of the plurality of capabilities being indicated by a pair of values that includes a minimum numeric value and a maximum numeric value, and
a count of devices in the cluster; and

send, via the network interface unit, information representative of the node clusters to a next higher level in the hierarchically arranged data center, wherein
the information representative of the node clusters is expanded by uniformly assigning, to a number of devices equal to the count of devices, values between the minimum numeric value and the maximum numeric value.

11. The apparatus of claim 10 , wherein the capabilities information comprises attributes of the capabilities and the processor is further configured to cluster based on a relative ranking of the attributes.

12. The apparatus of claim 11 , wherein the processor is further configured to cluster based on a capability having a first ranking, and then cluster based on a capability have a second ranking higher than the first ranking.

13. The apparatus of claim 10 , wherein the processor is configured to cluster nodes based on capabilities information received from at least one of a load balancer, firewall or computing device.

14. The apparatus of claim 10 , wherein the processor is configured to cluster nodes based on capabilities information received from collections of devices, including at least load balancers, firewalls or computing devices.

15. The apparatus of claim 10 , wherein the capabilities information comprises capabilities represented by non-numeric data, and the processor is further configured to determine a hamming distance among the capabilities represented by non-numeric data and to cluster based on the hamming distance.

16. The apparatus of claim 10 , wherein the processor is configured to cluster the nodes based on capabilities represented by numeric data and capabilities represented by non-numeric data.

17. A non-transitory computer-readable memory medium storing instructions that, when executed by a processor, cause the processor to:
receive capabilities information, the capabilities information being representative of capabilities of respective nodes at a first hierarchical level in a hierarchically arranged data center, each of the respective nodes comprising a plurality of devices;
cluster the nodes by summarizing the capabilities information, resulting in node clusters, wherein each node cluster is represented by:
a plurality of capabilities, one capability of the plurality of capabilities being indicated by a pair of values that includes a minimum numeric value and a maximum numeric value, and
a count of devices in the cluster; and

send information representative of the node clusters to a next higher level in the hierarchically arranged data center, wherein
the information representative of the node clusters is expanded by uniformly assigning, to a number of devices equal to the count of devices, values between the minimum numeric value and the maximum numeric value.

18. The non-transitory computer-readable memory medium of claim 17 , wherein the capabilities information comprises attributes of the capabilities, and wherein the instructions that cause the processor to cluster groups of nodes cause the processor to cluster based on a relative ranking of the attributes.

19. The non-transitory computer-readable memory medium of claim 18 , wherein the instructions that cause the processor to cluster groups of nodes cause the processor to cluster based on a capability having a first ranking, and then to cluster based on a capability having a second ranking higher than the first ranking.

20. The non-transitory computer-readable memory medium of claim 17 , wherein the capabilities information comprises non-numeric data, and wherein the instructions that cause the processor to cluster groups of nodes cause the processor to determine a hamming distance among the capabilities and to cluster based on the hamming distance.

Assignees

• Cisco Tech Inc

Inventors

• Rangwala Sumit
• Dutta Debojyoti
• Tadimeti Raja Rao
• Banerjee Subrata
• Zhu Yuanbo

Classifications

• G06F17/30489

  Physics · mapped topic

• G06F17/30598 Primary

  Physics · mapped topic

• H04L67/10

  in which an application is distributed across nodes in the network (software deployment G06F8/60; multiprogramming arrangements G06F9/46) · CPC title

• G06F2209/503

  Resource availability · CPC title

• G06F17/30876

  Physics · mapped topic