Highly Accurate Prediction of Jobs Runtime Classes
This work addresses job scheduling efficiency for computing systems, but it is incremental as it builds on known techniques for separating short and long jobs.
The paper tackles the problem of improving scheduling performance by predicting job runtime classes, achieving an overall accuracy of 90% with sensitivity and specificity both above 90%.
Separating the short jobs from the long is a known technique to improve scheduling performance. In this paper we describe a method we developed for accurately predicting the runtimes classes of the jobs to enable this separation. Our method uses the fact that the runtimes can be represented as a mixture of overlapping Gaussian distributions, in order to train a CART classifier to provide the prediction. The threshold that separates the short jobs from the long jobs is determined during the evaluation of the classifier to maximize prediction accuracy. Our results indicate overall accuracy of 90% for the data set used in our study, with sensitivity and specificity both above 90%.