Back to results
US 10,855,550 B2Granted
Network traffic prediction using long short term memory neural networks
AssigneeCisco Technology, Inc.
InventorsMehdi Nikkhah, Preethi Natarajan
Life of this patent
Estimating the term…
Abstract
A server uses an LSTM neural network to predict a bandwidth value for a computer network element using past traffic data. The server receives a time series of bandwidth utilization of the computer network element. The time series includes bandwidth values associated with a respective time values. The LSTM neural network is trained with a training set selected from at least a portion of the time series. The server generates a predicted bandwidth value associated with a future time value based on the LSTM neural network. The provisioned bandwidth for the computer network element is adjusted based on the predicted bandwidth value.
Claims17
A computer-implemented method comprising:
- receiving a time series of measured bandwidth utilization of a computer network element, the time series comprising a plurality of measured bandwidth values each associated with a respective time value;
- training a Long Short Term Memory (LSTM) neural network with a training set comprising at least a portion of the time series of measured bandwidth utilization;
- iteratively predicting a subsequent bandwidth value by:
- generating a predicted bandwidth value associated with a next respective time value in the time series, wherein the predicted bandwidth value is generated based on the LSTM neural network;
- generating a subsequent training set by updating the training set with the predicted bandwidth value associated with the next respective time value, the subsequent training set including the predicted bandwidth value in association with the next respective time value and at least a portion of the training set comprising the measured bandwidth utilization;
- and
- retraining the LSTM neural network with the subsequent training set;
- and
- adjusting a provisioned bandwidth for the computer network element based on the subsequent bandwidth value iteratively predicted using the retrained LSTM neural network.
6 more elements
An apparatus comprising:
- a network interface unit configured to communicate with computer network elements in a computer network;
- a Long Short Term Memory (LSTM) neural network configured to process a time series of measured bandwidth utilization received via the network interface unit, the time series comprising a plurality of measured bandwidth values each associated with a respective time value;
- and
- a processor configured to:
- train the LSTM neural network with a training set comprising at least a portion of the time series of measured bandwidth utilization;
- iteratively predict a subsequent bandwidth value by:
- generating a predicted bandwidth value associated with a next respective time value in the time series, wherein the processor generates the predicted bandwidth value based on the LSTM neural network;
- generating a subsequent training set by updating the training set with the predicted bandwidth value associated with the next respective time value, the subsequent training set including the predicted bandwidth value in association with the next respective time value and at least a portion of the training set comprising the measured bandwidth utilization;
- and
- retraining the LSTM neural network with the subsequent training set;
- and
- adjust a provisioned bandwidth for at least one of the computer network elements based on the subsequent bandwidth value iteratively predicted using the retrained LSTM neural network.
9 more elements
One or more non-transitory computer readable storage media encoded with computer executable instructions operable to cause a processor to:
- receive a time series of measured bandwidth utilization of a computer network element, the time series comprising a plurality of measured bandwidth values each associated with a respective time value;
- train a Long Short Term Memory (LSTM) neural network with a training set comprising at least a portion of the time series of measured bandwidth utilization;
- iteratively predict a subsequent bandwidth value by:
- generating a predicted bandwidth value associated with a next respective time value in the time series, wherein the predicted bandwidth value is generated based on the LSTM neural network;
- generating a subsequent training set by updating the training set with the predicted bandwidth value associated with the next respective time value, the subsequent training set including the predicted bandwidth value in association with the next respective time value and at least a portion of the training set comprising the measured bandwidth utilization;
- and
- retraining the LSTM neural network with the subsequent training set;
- and
- adjust a provisioned bandwidth for the computer network element based on the subsequent bandwidth value iteratively predicted using the retrained LSTM neural network.
6 more elements
Dependencies are inferred from the claim text.