DDoS (Distributed Denial of Service) is one of the cyber-attacks that make the
network service unavailable. SDN (Software Defined Network) has tools to defeat
the DDoS, because SDN has good features in defeating DDoS such as logically
centralized controller, separation control plan, and programmability network.
This thesis offers the SVM (Support Vector Machine) that combined with Ryu
controller that can predict the incoming packet based on the learned traffic, whether
it is a normal packet or DDoS packet, and blocking the packet if the packet indicating
the DDoS traffic.
The purpose of this simulation is to evaluate the capability of the SVM
controller in handling the various size packet size and number of attackers of DDoS
attack. The result of the simulation the smallest attack damage that DDoS does is
at the 300 PPS rate with 1,2,3, and 4 attackers with an average throughput 9 Mbps.