Frontend storage port rebalancing and allocation using machine learning techniques

A method, comprising:

• inputting, to a prediction model by a system comprising at least one processor, port metrics data monitored from datacenter frontend storage ports coupled to a storage array;
• inputting, to the prediction model by the system, historical port-related data captured over a prior timeframe;
• and
• in response to the inputting of the port metrics data and the inputting of the historical port-related data, obtaining, by the system, port rebalancing information from the prediction model,
• wherein the prediction model comprises an autoregressive integrated moving average (ARIMA) model, a linear regression model, and a locally estimated scatterplot smoothing (STL) model, and wherein the obtaining of the port rebalancing information from the prediction model comprises obtaining a weighted combination of predictions from the ARIMA model, the linear regression model, and the STL model.

A non-transitory machine-readable medium, comprising executable instructions that, when executed by a processor, facilitate performance of operations, the operations comprising:

• inputting, to a prediction model, port metrics data monitored from datacenter frontend storage ports coupled to a storage array;
• inputting, to the prediction model, historical port-related data captured over a prior timeframe;
• and
• in response to the inputting of the port metrics data and the inputting of the historical port-related data, obtaining port rebalancing information from the prediction model,
• wherein the prediction model comprises an autoregressive integrated moving average (ARIMA) model, a linear regression model, and a locally estimated scatterplot smoothing (STL) model, and wherein the obtaining of the port rebalancing information from the prediction model comprises obtaining a weighted combination of predictions from the ARIMA model, the linear regression model, and the STL model.