Network traffic prediction method and device

The invention discloses a network traffic prediction method and device, and the method comprises the steps: enabling the prediction device to obtain a prediction network flow sequence of a second timeperiod after a first time period according to all prediction models pre-stored in a historical network flow sequence training model library of exit gateway equipment in a first time period; and updating the prediction model of at least one sub-time period according to the actual network flow of at least one sub-time period in the second time period of which the network flow error value is greaterthan the preset threshold when judging that the predicted network flow sequence does not meet the expected target based on a comparison result between the actual network flow sequence in the second time period and the predicted network flow sequence. Through the mode, the prediction device can train the prediction model corresponding to the sub-time period according to the historical network flowcorresponding to different