PCM A Real-Time Detection System And Method On Flush Reload Attack Using PCM

摘要

The present invention relates to a system for detecting a flush+reload cache subchannel attack in real time using a performance counter attack (PCM) and a method thereof. According to the present invention, the method comprises the following steps of: executing a detection module including a detection function for using a metafile learned through machine learning; generating a PCM module and a comma separated values (CSV) file after the detection module is executed; inputting data generated from the CSV file as an input value for machine learning; outputting two kinds of predication values corresponding to the input value through the metafile; and determining the presence of an attack based on the prediction values and providing a notification in accordance with the presence of the attack. The number of data of the CSV file exceeds twenty, the CSV file is initialized. Accordingly, due to simple execution, the method quickly detects the attack in real time to prepare a measure against the attack. Moreover, the method uses machine learning, thereby performing correct prediction during execution of other background modules.