ABSTRACT In an SDN environment, one of the most serious risks that might emerge is a distributed denial of service (DDoS) attack. It is a form of attack in which a large number of bogus packets are delivered into the network from various sources in order to drain network resources. In this research, we employed an SDN environment to generate and collect DDoS and normal traffic.SYN flooding, UDP flooding, and ICMP flooding attacks were produced using the hping3 tool, while normal traffic was produced using the ping and iperf. The produced traffic was assembled into a dataset and classified using machine learning models, namely Naive Bayes, Logistic Regression, SVM, KNN and Gradient Boost models. The evaluations findings demonstrate that KNN and Gradient Boost classifiers outperform other models in terms of accuracy, Precision, Recall, F1-score, AUC-ROC, training time and testing time. Keywords: SDN, DDoS attacks, Mininet, Iperf, hping3, Machine Learning Models.