This paper proposes a model for detecting anomalous behavior in cloud computing. The model uses Software-Defined Networks to redirect virtual machine (VM) traffic so it can be captured and analyzed. The goal is to detect both known and unknown anomalous network behaviors between VMs. The model adopts hybrid techniques to analyze VM network behaviors and control network systems. Experimental results showed the approach was over 90% effective, proving the feasibility of the proposed anomalous behavior detection model.