Back to results
US 10,540,605 B2Granted
Traffic-based inference of influence domains in a network by using learning machines
AssigneeCisco Technology, Inc.
InventorsGrégory Mermoud, Jean-Philippe Vasseur, Sukrit Dasgupta
Life of this patent
Estimating the term…
Abstract
In one embodiment, techniques are shown and described relating to traffic-based inference of influence domains in a network by using learning machines. In particular, in one embodiment, a management device computes a time-based traffic matrix indicating traffic between pairs of transmitter and receiver nodes in a computer network, and also determines a time-based quality parameter for a particular node in the computer network. By correlating the time-based traffic matrix and time-based quality parameter for the particular node, the device may then determine an influence of particular traffic of the traffic matrix on the particular node.
Claims18
A method, comprising:
- computing, by a learning machine, a time-based traffic matrix indicating traffic between pairs of transmitter and receiver nodes in a computer network;
- determining, by the learning machine, a time-based quality parameter for a particular node in the computer network;
- performing, by the learning machine, a linear regression between the traffic matrix and the quality parameter for a given time at the particular node;
- determining, by the learning machine, based on an output of the linear regression, an amount of influence particular traffic of the traffic matrix has on the particular node in the computer network by correlating the time-based traffic matrix and time-based quality parameter for the particular node, wherein datapoints of the time-based traffic matrix associated with the particular traffic are used as input variables to determine the amount of influence the particular traffic has on the particular node on a per node basis;
- identifying, by the learning machine, one or more nodes in the computer network that affect performance of the particular node, wherein the performance includes a quality of service at the particular node based on the amount of influence;
- and
- setting, by the learning machine, the one or more nodes in the computer network that affect performance to inactive while the particular node is transmitting.
4 more elements
An apparatus, comprising:
- one or more network interfaces to communicate with a computer network;
- a processor coupled to the network interfaces and adapted to execute one or more processes;
- and
- a memory configured to store a process executable by the processor, the process when executed operable to:
- compute, as a learning machine, a time-based traffic matrix indicating traffic between pairs of transmitter and receiver nodes in the computer network;
- determine, as the learning machine, a time-based quality parameter for a particular node in the computer network;
- perform a linear regression between the traffic matrix and the quality parameter for a given time;
- determine, as the learning machine, based on an output of the linear regression, an amount of influence particular traffic of the traffic matrix has on the particular node in the computer network by correlating the time-based traffic matrix and time-based quality parameter for the particular node, wherein datapoints of the time-based traffic matrix associated with the particular traffic are used as input variables to determine the amount of influence the particular traffic has on the particular node on a per node basis,
- identify, by the learning machine, one or more nodes in the computer network that affect performance of the particular node, wherein the performance includes a quality of service at the particular node based on the amount of influence;
- and
- set, by the learning machine, the one or more nodes in the computer network that affect performance to inactive while the particular node is transmitting.
8 more elements
A tangible, non-transitory, computer-readable media having software encoded thereon, the software when executed by a processor operable to:
- compute, as a learning machine, a time-based traffic matrix indicating traffic between pairs of transmitter and receiver nodes in a computer network;
- determine, as the learning machine, a time-based quality parameter for a particular node in the computer network;
- perform a linear regression between the traffic matrix and the quality parameter for a given time;
- and
- determine, as the learning machine, based on an output of the linear regression, an amount of influence particular traffic of the traffic matrix has on the particular node in the computer network by correlating the time-based traffic matrix and time-based quality parameter for the particular node, wherein datapoints of the time-based traffic matrix associated with the particular traffic are used as input variables to determine the amount of influence the particular traffic has on the particular node on a per node basis,
- identify, by the learning machine, one or more nodes in the computer network that affect performance of the particular node, wherein the performance includes a quality of service at the particular node based on the amount of influence;
- and
- set, by the learning machine, the one or more nodes in the computer network that affect performance to inactive while the particular node is transmitting.
5 more elements
Dependencies are inferred from the claim text.