PCA is a useful statistical technique that has found application in fields such as face recognition, image compression, dimensionality reduction, Computer System performance analysis etc. It is a common technique for finding patterns in data of high dimension. In this paper, we present the basic idea of principal component analysis as a general approach that extends to various popular data analysis techniques. We state the mathematical theory behind PCA and focus on monitoring system performance using the PCA algorithm. Next, an Eigen value-Eigenvector dynamics is elaborated which aims to reduce the computational cost of the experiment. The Mathematical theory is explored and validated. For the purpose of illustration we present the algorithmic implementation details and numerical examples over real time and synthetic datasets.