Prediction of total execution time for MapReduce applications

摘要

In this paper, we estimate the total execution time of two applications under different input data size with linear regression model and error correction neural network model respectively. From the prediction results, we can conclude that error correction neural network model can predict the total execution time of these applications under different input data size accurately and the prediction accuracy of error correction neural network model for CPU-intensive workloads is much higher compared to linear regression model.