Modern Computing Systems such as Web servers, databases, cloud environmentetc, are vulnerable to attackers in the web. One of the most commonly known type of threats is Denial-of- Service (DoS) attacks causes heavy loses and damages to these systems. In this paper, we present detection system to analyze DoS attack that uses Multivariate Correlation Analysis (MCA) for determining network traffic characterization by analyzing the distinct correlation between network traffic features. Our MCA-based approach employs the mechanism of widely used detection method anomaly-based detection in attack identification. Thus making it quite easy to detect known as well as unknown attacks by learning patterns of legitimate network traffic. Furthermore a triangle area based approach is used along to enhance the process of MCA. The proposed system is effectively mapped and calculated using KDD Cup 99 Dataset.