Method and apparatus for deploying a minimal-cost CCN topology

Patent metadata

Field	Value
Publication number	US-9537719-B2
Application number	US-201414309686-A
Country	US
Kind code	B2
Filing date	Jun 19, 2014
Priority date	Jun 19, 2014
Publication date	Jan 3, 2017
Grant date	Jan 3, 2017

Abstract

A CCN-deployment system can design and deploy a content centric network (CCN) topology, either across a collection of CCN nodes or across an existing computer network. During operation, the system analyzes a computer network of N network nodes to determine a physical network topology. The system also determines a number, k, of network nodes of the physical network on which to overlay a content centric network (CCN). The system then determines an average degree of connectivity, and a degree-of-connectivity distribution, that achieves an optimal performance metric for the CCN overlay network. The system generates a network topology of k network nodes that satisfies the average degree of connectivity, and that satisfies the degree-of-connectivity distribution. The system can deploy the content centric network topology across k nodes of the underlying physical network.

First claim

What is claimed is:

1. A computer-implemented method, comprising:
determining, by a computing device, a number k, of network nodes for a content centric networking (CCN) network;
determining an average node degree as twice a number of edges divided by a number of nodes, a degree-of-connectivity distribution, and a joint degree or higher-order distribution, that achieves an optimal performance metric for the CCN network to transmit a CCN Interest, wherein the joint degree distribution indicates an occurrence for a respective pair of node degrees, and wherein the higher-order distribution indicates a probability distribution of loops with at least three nodes;
generating a network topology of k network nodes that satisfies the average node degree, and satisfies the degree-of-connectivity distribution and joint degree or higher-order distribution;
mapping the k nodes of the generated network topology to nodes of a physical computer network; and
transmitting the CCN Interest via the physical computer network according to the mapped network topology.

2. The method of claim 1 , wherein mapping the k nodes of the generated network topology to nodes of the physical computer network involves selecting the k network nodes of the physical computer network on which to overlay the CCN network, such that the selected nodes satisfy the generated network topology for the CCN network; and
wherein the method further comprises deploying the CCN network across the k nodes selected from the physical computer network.

3. The method of claim 1 , wherein the performance metrics include at least one of:
CCN Interest overhead;
a number or percentage of CCN Interest retransmissions; and
an Interest-to-Content-Object round-trip delay.

4. The method of claim 1 , wherein determining the average node degree involves:
iterating over one or more average node degrees, to generate a network topology of k network nodes for each average node degree;
computing a performance metric for each network topology; and
selecting an average node degree with a highest performance metric.

5. The method of claim 1 , wherein determining the degree-of-connectivity distribution involves determining an optimal distribution based on one or more of:
a power-law distribution; and
a Gaussian distribution.

6. The method of claim 1 , wherein determining the degree-of-connectivity distribution involves:
iterating over one or more distribution functions, to generate a network topology of k network nodes that satisfies the average node degree based on a corresponding distribution function;
computing a performance metric for each network topology; and
selecting a degree-of-connectivity distribution with a highest performance metric.

7. The method of claim 6 , wherein determining the degree-of-connectivity distribution further involves:
iterating over one or more parameters for the distribution function, to generate a network topology of the k network nodes that satisfies the average node degree based on each distribution function;
computing a performance metric for each network topology; and
selecting network parameters with a highest performance metric.

8. The method of claim 1 , wherein the joint degree or higher-order distribution includes a degree-distribution matrix.

9. The method of claim 8 , wherein determining the degree-of-connectivity distribution involves:
generating the degree-distribution matrix to indicate the occurrence value for each pair of node degrees, wherein the degree-distribution matrix is multi-dimensional, and achieves an optimal performance metric for a network topology of the k network nodes that satisfies the average node degree.

10. A non-transitory computer-readable storage medium storing instructions that when executed by a computer cause the computer to perform a method:
determining a number k, of network nodes for a content centric networking (CCN) network;
determining an average node degree as twice a number of edges divided by a number of nodes, a degree-of-connectivity distribution, and a joint degree or higher-order distribution, that achieves an optimal performance metric for the CCN network to transmit a CCN Interest, wherein the joint degree distribution indicates an occurrence for a respective pair of node degrees, and wherein the higher-order distribution indicates a probability distribution of loops with at least three nodes;
generating a network topology of k network nodes that satisfies the average node degree, and satisfies the degree-of-connectivity distribution and joint degree or higher-order distribution;
mapping the k nodes of the generated network topology to nodes of a physical computer network; and
transmitting the CCN Interest via the physical computer network according to the mapped network topology.

11. The non-transitory computer-readable storage medium of claim 10 , wherein mapping the k nodes of the generated network topology to nodes of the physical computer network involves selecting the k network nodes of the physical computer network on which to overlay the CCN network, such that the selected nodes satisfy the generated network topology for the CCN network; and
wherein the method further comprises deploying the CCN network across the k nodes selected from the physical computer network.

12. The non-transitory computer-readable storage medium of claim 10 , wherein determining the average node degree involves:
iterating over one or more average node degrees, to generate a network topology of k network nodes for each average node degree;
computing a performance metric for each network topology; and
selecting an average node degree with a highest performance metric.

13. The non-transitory computer-readable storage medium claim 10 , wherein determining the degree-of-connectivity distribution involves determining an optimal distribution based on one or more of:
a power-law distribution; and
a Gaussian distribution.

14. The non-transitory computer-readable storage medium of claim 10 , wherein determining the degree-of-connectivity distribution involves:
iterating over one or more distribution functions, to generate a network topology of k network nodes that satisfies the average node degree based on a corresponding distribution function;
computing a performance metric for each network topology; and
selecting a degree-of-connectivity distribution with a highest performance metric.

15. The non-transitory computer-readable storage medium of claim 14 , wherein determining the degree-of-connectivity distribution further involves:
iterating over one or more parameters for the distribution function, to generate a network topology of the k network nodes that satisfies the average node degree based on each distribution function;
computing a performance metric for each network topology; and
selecting network parameters with a highest performance metric.

16. The non-transitory computer-readable storage medium of claim 10 , wherein the joint degree or higher-order distribution includes a degree-distribution matrix.

17. The non-transitory computer-readable storage medium of claim 16 , wherein determining the degree-of-connectivity distribution involves:
generating the degree-distribution matrix to indicate the occurrence value for each pair of node degrees, wherein the degree-distribution matrix is multi-dimensional, and achieves an optimal performance metric for a network topology of the k network nodes that satisfies the average node degree.

18. An apparatus, comprising:
a processor; and
storage medium storing

Assignees

• Palo Alto Res Ct Inc

Inventors

• Mahadevan Priya
• Quereilhac Alina

Classifications

• H04L41/12 Primary

  Discovery or management of network topologies · CPC title

• H04L45/126

  minimising geographical or physical path length · CPC title

• H04L41/145

  involving simulating, designing, planning or modelling of a network · CPC title

• H04L41/142

  using statistical or mathematical methods · CPC title

• Y04S40/164

  Cross-Sectional Technologies · mapped topic