In-situ oam (ioam) network risk flow-based "topo-gram" for predictive flow positioning

Patent metadata

Field	Value
Publication number	US-2019182103-A1
Application number	US-201715834284-A
Country	US
Kind code	A1
Filing date	Dec 7, 2017
Priority date	Dec 7, 2017
Publication date	Jun 13, 2019
Grant date	—

Abstract

A system and method predict risks of failure or performance issues in a network to predictively position traffic flows in the network. For a traffic flow through a network, first data accumulated in a header of packets for the traffic flow is obtained, which header is populated by network elements along a path of the traffic flow through the network. Second data is obtained about the network in general including other network elements not along the path of the traffic flow. Machine learning analysis is performed to derive rules that characterize failure or performance risk issues in the network. The rules and topology data describing a topology of the network are applied to a model to create a topological graphical representation indicating failure or performance issues in the network that affect the traffic flow. A path for the traffic flow is modified based on the topological graphical representation.

First claim

What is claimed is:

1 . A method comprising:
for a traffic flow through a network, obtaining first data accumulated in a header of packets for the traffic flow, which header is populated by network elements along a path of the traffic flow through the network; obtaining second data about the network in general including other network elements not along the path of the traffic flow; deriving from the first data and the second data information describing operational attributes of individual network elements and attributes of service levels associated with traffic flowing through the network; analyzing the information to derive rules that characterize failure or performance risk issues in the network; applying the rules and topology data describing a topology of the network to a model to create a topological graphical representation indicating failure or performance issues in the network that affect the traffic flow; and modifying a path for the traffic flow based on the topological graphical representation.

2 . The method of claim 1 , wherein processing the first data and the second data to derive the information includes deriving at least a first attribute set and a second attribute set, the first attribute set describing operational attributes of individual network elements and the second attribute set describing service level related attributes.

3 . The method of claim 2 , wherein analyzing includes performing machine learning analysis includes deriving rules in terms of conditions, percentile and composite-value, where conditions specify the network conditions that occur, percentile specifies a percentage of the conditions attributes that occur and composite-value is a cumulative value of the number of occurrences that the attributes occur according to the percentile threshold.

4 . The method of claim 3 , wherein the topological graphical representation includes a numerical value for each network element and for each link between network elements in the network, wherein a magnitude of the numeral value represents a level of failure or performance risk for the network element or link.

5 . The method of claim 4 , wherein modifying comprises selecting a new path for the traffic flow to avoid a network element or link whose numeral value indicates an unacceptable level of failure or performance risk.

6 . The method of claim 1 , wherein obtaining the first data comprises receiving the first data at a computing device from a last hop network element of the path that extracted the first data from the header of packets for the traffic flow, and wherein the processing, the performing machine learning analysis, the applying and the modifying are performed at the computing device.

7 . The method of claim 6 , further comprising a last hop network element performing machine learning analysis on the first data prior to sending the first data to the computing device.

8 . The method of claim 1 , wherein the first data includes identifiers of network elements, timestamps, interfaces visited, and queue depth.

9 . A system comprising:
a network including a plurality of network elements, wherein for a traffic flow through the network, first data is accumulated in a header of packets for the traffic flow, which header is populated by network elements along a path of the traffic flow; a computing device in communication with the plurality of network elements, wherein the computing device is configured to:
obtain the first data;
receive second data about the network in general and including other network elements not along the path of the traffic flow;
derive from the first data and the second data information describing operational attributes of individual network elements and attributes of service levels associated with traffic flowing through the network;
analyze the information to derive rules that characterize failure or performance risk issues in the network;
apply the rules and topology data describing a topology of the network to a model to create a topological graphical representation indicating failure or performance issues in the network that affect the traffic flow; and
modify a path for the traffic flow based on the topological graphical representation.

10 . The system of claim 9 , wherein the computing device is configured to process first data and the second data to derive the information includes deriving at least a first attribute set and a second attribute set, the first attribute set describing operational attributes of individual network elements and the second attribute set describing service level related attributes.

11 . The system of claim 10 , wherein the computing device is configured to perform machine learning analysis by deriving rules in terms of conditions, percentile and composite-value, where conditions specify the network conditions that occur, percentile specifies a percentage of the conditions attributes that occur and composite-value is a cumulative value of the number of occurrences that the attributes occur according to the percentile threshold.

12 . The system of claim 11 , wherein the topological graphical representation includes a numerical value for each network element and for each link between network elements in the network, wherein a magnitude of the numeral value represents a level of failure or performance risk for the network element or link.

13 . The system of claim 12 , wherein the computing device is configured to modify by selecting a new path for the traffic flow to avoid a network element or link whose numeral value indicates an unacceptable level of failure or performance risk.

14 . The system of claim 9 , wherein the computing device is configured to receive the first data from a last hop network element of the path that extracted the first data from the header of packets for the traffic flow.

15 . The system of claim 14 , wherein a lost hop network element in the path is configured to perform machine learning analysis on the first data prior to sending the first data to the computing device.

16 . An apparatus comprising:
a network interface configured to enable network communication; a processor coupled to the network interface, wherein the processor is configured to:
receive first data accumulated in a header of packets for a traffic flow, which header is populated by network elements in a network along a path of the traffic flow;
receive second data about the network in general and including other network elements not along the path of the traffic flow;
derive from the first data and the second data information describing operational attributes of individual network elements and attributes of service levels associated with traffic flowing through the network;
analyze the information to derive rules that characterize failure or performance risk issues in the network;
apply the rules and topology data describing a topology of the network to a model to create a topological graphical representation indicating failure or performance issues in the network that affect the traffic flow; and
modify a path for the traffic flow based on the topological graphical representation.

17 . The apparatus of claim 16 , wherein the processor is configured to process first data and the second data to derive the information includes deriving at least a first attribute set and a second attribute set, the first attribute set describing operational attributes of individual network elements and the second attribute set describing service level related attributes.

18 . The apparatus of claim

Assignees

• Cisco Tech Inc

Inventors

• Pignataro Carlos M
• Nainar Nagendra Kumar
• Asati Rajiv

Classifications

• H04L47/24

  Traffic characterised by specific attributes, e.g. priority or QoS · CPC title

• H04L41/0668 Primary

  by dynamic selection of recovery network elements, e.g. replacement by the most appropriate element after failure · CPC title

• H04L41/5025

  by proactively reacting to service quality change, e.g. by reconfiguration after service quality degradation or upgrade · CPC title

• H04L41/16

  using machine learning or artificial intelligence · CPC title

• H04L41/147

  for predicting network behaviour · CPC title