Intent-based orchestration using network parsimony trees

Patent metadata

Field	Value
Publication number	US-11343201-B2
Application number	US-202117176066-A
Country	US
Kind code	B2
Filing date	Feb 15, 2021
Priority date	Feb 25, 2020
Publication date	May 24, 2022
Grant date	May 24, 2022

Abstract

Novel tools and techniques are provided for implementing intent-based orchestration using network parsimony trees. In various embodiments, in response to receiving a request for network services that comprises desired characteristics and performance parameters for the requested network services without information regarding specific hardware, hardware type, location, or network, a computing system might generate a request-based parsimony tree based on the desired characteristics and performance parameters. The computing system might access, from a datastore, a plurality of network-based parsimony trees that are each generated based on measured network metrics, might compare the request-based parsimony tree with each of one or more network-based parsimony trees to determine a fitness score for each network-based parsimony tree, and might identify a best-fit network-based parsimony tree based on the fitness scores. The computing system might identify and might allocate network resources based on the identified best-fit network-based parsimony tree, for providing the requested network services.

First claim

What is claimed is:

1. A method, comprising:
receiving, with a computing system, a request for network services from a user device associated with a customer, the request for network services comprising desired characteristics and performance parameters for the requested network services, without information regarding any of specific hardware, specific hardware type, specific location, or specific network for providing the requested network services;
in response to receiving the request for network services, generating, with the computing system, a first request-based parsimony tree based at least in part on the desired characteristics and performance parameters contained in the request for network services, the first request-based parsimony tree being a graphical representation comprising an end-point of a first portion representing delivery location of the requested network services, an endpoint of each of one or more second portions that connect with the first portion representing a service provider site, each intersection between two or more second portions or between the first portion and one of the second portions representing a network resource node, and characteristics of the first and second portions representing the desired characteristics and performance parameters contained in the request for network services;
accessing, with the computing system and from a datastore, a plurality of first network-based parsimony trees, each of the plurality of first network-based parsimony trees being generated based on measured network metrics, each first network-based parsimony tree being a graphical representation comprising an end-point of a third portion representing the delivery location of the requested network services, an endpoint of each of one or more fourth portions that connect with the third portion representing a service provider site, each intersection between two or more fourth portions or between the third portion and one of the fourth portions representing a network resource node, and characteristics of the third and fourth portions representing measured characteristics and performance parameters based on the measured network metrics;
comparing, with the computing system, the first request-based parsimony tree with one or more first network-based parsimony trees among the plurality of first network-based parsimony trees to determine a fitness score for each first network-based parsimony tree, each fitness score being a value indicative of a level of heuristic matching between the first request-based parsimony tree with one of the one or more first network-based parsimony trees;
identifying, with the computing system, a best-fit network-based parsimony tree among the one or more first network-based parsimony trees based on the fitness scores of the one or more first network-based parsimony trees;
identifying, with the computing system, one or more first network resources among a first plurality of network resources for providing the requested network services, based at least in part on network resources represented within the identified best-fit network-based parsimony tree; and
allocating, with the computing system, at least one first network resource among the identified one or more first network resources for providing the requested network services.

2. The method of claim 1 , wherein the computing system comprises one of a server computer over a network, one or more graphics processing units (“GPUs”), a cloud-based computing system over a network, or a distributed computing system.

3. The method of claim 1 , wherein the desired performance parameters comprise at least one of a maximum latency, a maximum jitter, a maximum packet loss, a maximum cost, or a maximum number of hops.

4. The method of claim 1 , wherein the desired characteristics comprise at least one of requirement for network equipment to be geophysically proximate to the user device associated with the customer, requirement for network equipment to be located within a first geophysical location, requirement to avoid routing network traffic through a second geophysical location, requirement to route network traffic through a third geophysical location, requirement to exclude a first type of network resources from fulfillment of the requested network services, requirement to include a second type of network resources for fulfillment of the requested network services, requirement to fulfill the requested network services based on a single goal indicated by the customer, or requirement to fulfill the requested network services based on multiple goals indicated by the customer.

5. The method of claim 1 , wherein the first portion of the first request-based parsimony tree and the third portion of each first network-based parsimony tree are each represented by a trunk, wherein the one or more second portions of the first request-based parsimony tree and the one or more fourth portions of each first network-based parsimony tree are each represented by a branch, and wherein, in each parsimony tree, one or more branches connect with the trunk.

6. The method of claim 5 , wherein, in each of at least one parsimony tree, two or more branches connect with each other via one or more connector branches and via the trunk.

7. The method of claim 1 , wherein the characteristics of the first and second portions and the characteristics of the third and fourth portions comprise at least one of thickness of each portion, length of each portion from the delivery location of the requested network services, number of network resource nodes on each portion, color-code of each portion, number of second or fourth portions, angle of each second or fourth portion relative to the first or third portion, number of any connector portions between two or more second or fourth portions, relative location of any connector portions between two or more second or fourth portions, length of any connector portions between two or more second or fourth portions, or angle of any connector portions between two or more second or fourth portions.

8. The method of claim 7 , wherein the characteristics of the first and second portions and the characteristics of the third and fourth portions represent one or more of latency, jitter, packet loss, number of hops, bandwidth, utilization, capacity, or proximity.

9. The method of claim 7 , further comprising:
applying, with the computing system, a first filter to at least one first network-based parsimony tree among the one or more first network-based parsimony trees to filter out one or more characteristics or one or more sub-characteristics, prior to comparing the first request-based parsimony tree with the one or more first network-based parsimony trees.

10. The method of claim 7 , wherein the characteristics of the third and fourth portions comprise color-codes embodied as a color temperature or range of colors for each portion or for each parsimony tree that is indicative of characteristics or performance parameters including one or more of latency, jitter, packet loss, number of hops, bandwidth, utilization, capacity, or proximity, wherein the method further comprises:
applying, with the computing system, a second filter to at least one first network-based parsimony tree among the one or more first network-based parsimony trees to change the color temperature based on changes in measured network metrics.

11. The method of claim 1 , wherein generating the first request-based parsimony tree comprises generating a plurality of first request-based parsimony trees, each representing a desired characteristic or performance parameter; wherein the plurality of first network-based parsimony trees comprises a plurality of first network-ba

Assignees

• Level 3 Communications Llc

Inventors

• Mcbride Kevin M
• Dwyer Brett E
• Sutherland James E
• Lin Jamie
• Smith Brent
• Balanoff Glenn B
• Moss Frank
• Dreyer Bryan

Classifications

• G06T11/26

  Drawing of charts or graphs · CPC title

• G06T11/10

  Texturing; Colouring; Generation of textures or colours (retouching, inpainting or scratch removal G06T5/77) · CPC title

• H04L67/1021

  based on client or server locations · CPC title

• H04L67/1012

  based on compliance of requirements or conditions with available server resources · CPC title

• H04L43/20

  the monitoring system or the monitored elements being virtualised, abstracted or software-defined entities, e.g. SDN or NFV · CPC title