Identification of software phases using machine learning

    公开(公告)号:US10089207B2

    公开(公告)日:2018-10-02

    申请号:US14317485

    申请日:2014-06-27

    Abstract: A computing device executes an application having a number of phases. The computing device collects performance data indicative of a number of performance attributes of the computing device during execution of the application. The performance attributes include page swap data, page fault data, and process queue data. The computing device merges data collected from a processor performance monitoring unit with data collected from an operating system of the computing device. The computing device partitions the performance data into a number of cluster models, applies a classification algorithm to each cluster model, and selects the cluster model with the lowest misclassification rate. The computing device associates each cluster of the cluster model to a phase of the software application. Compatible phases of software applications are scheduled based on the selected cluster model.

Patent Agency Ranking