Back to results
US 12,587,438 B2Granted
Methods and apparatus for network analysis
AssigneePromptlink Communications, Inc.
InventorsFoad Towfiq, Alexander Podarevsky
Life of this patent
Estimating the term…
Abstract
Various techniques for analyzing data networks with impaired network components, recording network telemetry data and labeling the information for use in training one or more machine learning algorithms is described. The machine learning algorithms trained in this way are used to detect impaired network components and/or noise on operational data networks.
Claims20
A method, comprising:
- creating a first network condition in a first data network, wherein the first network condition impairs operation of one or more network components of the first data network;
- while the first network condition impairs the operation of one or more network components of the first data network, recording first telemetry data produced by a plurality of network components of the first data network;
- recording topology information about the first data network;
- associating the first telemetry data with a characterization of the first network condition and the topology information about the first data network;
- and
- training one or more machine learning models using the first telemetry data in association with the characterization of the first network condition and the topology information about the first data network.
3 more elements
A method, comprising:
- receiving first telemetry data produced by a plurality of network components of a first data network that includes a first network condition that impairs operation of one or more network components of the first data network, wherein the first telemetry data was produced while the first network condition impairs the operation of one or more network components of the first data network;
- receiving topology information about the first data network, wherein the topology information about the first data network is associated with a characterization of the first network condition and the first telemetry data;
- and
- training one or more machine learning models using the first telemetry data in association with the characterization of the first network condition and the topology information about the first data network.
1 more element
A non-transitory computer-readable storage medium storing one or more programs configured to be executed by one or more processors, the one or more programs including instructions for:
- receiving first telemetry data produced by a plurality of network components of a first data network that includes a first network condition that impairs operation of one or more network components of the first data network, wherein the first telemetry data was produced while the first network condition impairs the operation of one or more network components of the first data network;
- receiving topology information about the first data network, wherein the topology information about the first data network is associated with a characterization of the first network condition and the first telemetry data;
- and
- training one or more machine learning models using the first telemetry data in association with the characterization of the first network condition and the topology information about the first data network.
1 more element
Dependencies are inferred from the claim text.