Grid load predicting method based on support vector regression machine

Abstract

The invention relates to a grid load predicting method based on a support vector regression machine. The method comprises the following steps: firstly, carrying out automatic regression (AR) modelingon history property data of nodes by a time sequence method; evaluating dimensions of the input vector in SVR according to orders of an AR model; performing SVR learning on the history data, and constructing a regression function of SVR; predicting the property of the node at the next time according to the regression function and the measured history property data, and regulating the regression function of SVR on line according to the regression function. The method can provide a data reference for dispatch, property optimization and the like of grid resources, avoid passive and blind task dispatch and enhance the efficiency of the grid environment.

Description

technical field [0001] The invention is a grid load prediction method. This method first uses the time series method to perform autoregressive (AR) modeling on the historical performance data of the node, estimates the dimension of the input vector in the SVR according to the order of the AR model, and then performs SVR learning on the historical data to construct the SVR According to the regression function, the performance of the node at the next moment is predicted according to the regression function and the measured historical performance data, and the SVR regression function is adjusted online according to the error of the prediction result. The method can provide data basis for grid resource scheduling and performance optimization, avoid passive and blind task scheduling, and improve the efficiency of the entire grid environment, belonging to the technical field of grid computing. Background technique [0002] As an important new field, Grid Computing has gained worl...

AI Technical Summary

Problems solved by technology

However, due to the autonomy of grid nodes, their resources are not controlled by the grid, which makes grid task scheduling and performance optimization difficult

A typical situation is that when a user submits a batch of tasks, the scheduler finds the grid node that is currently the most idle, and assigns the batch of tasks to it, but before they are fully accepted by the grid nodes, the node is in The scheduler runs a daily program unexpectedly and causes the CPU to be busy. At this time, this batch of tasks will increase the burden on the node, and it will be impossible to predict when they will be completed.

Images