Distributed Denial of Service (DDoS) attack constitutes one of the prominent cyber threats and among the hardest security problems in modern cyber world. This paper is to review the DDoS detection techniques and developing a numeric stable theoretical framework used for detecting various DDoS attacks in cloud. This paper is devoted to review and analysis of algorithms used for detection of DDoS attacks. The framework theorized here deals with the variability calculation method in conjunction with sampling, searching methods to find a current state of a particular parameter under observation for detecting DDoS attacks.