A comparative study of various DDoS attacks there detection and discrimination approaches against a flow correlation approach

Distributed Denial of Service (DDoS) attack is a serious threat to the Internet Community.Eventhough the main intension of DDoS attacks is to access the web resources for illegitimate purpose, they posses a serious threat to the legitimate users. Various techniques are available to detect a DDoS attack. But the main challenge is how to discriminate a DDoS attack from a flash crowd. In this paper we had studied about the various type of DDoS attacking tools there behavior and the various DDoS detection method. The attacks we studied spanned from DoS attacks to the attacks from sophisticated Botnets(Hybrid Botnets).As each attack detection method will be broken by a new detection overcoming methods all the attacks and the detection approaches have been studied in a sequence. In many cases it was found that the attackers can mimic or imitate the characteristics of a legitimate traffic by the illegitimate traffic. So a novel approach is needed to discriminate this mimicking and discriminate the attacks from a legitimate access. We were able to compare the available techniques with the most comprehensive and updated technique using a flow correlation approach. This technology was found to be a promising technique for discriminating DDoS attacks from the legal crowd called as Flash Crowds