Addressing dynamic failures in Software-defined Networks

Legacy IP networks are currently a huge problem for Internet Service Providers, as the demand grows exponentially, the profit doesn’t follow. With the emergence of the Software-Defined Networks (SDN), providers are hoping to improve their service while lowing the operational expenses, without requiring massive investment. In this proposal, we focus our work towards self-healing SDNs, that requires fault-tolerant mechanisms and intelligent network management for enabling the system to perceive its incorrect states and acting to fix it. As fault tolerance is a huge issue, we narrow our project for only dynamic failures, as these are usually the best target for machine learning approaches as deterministic solutions are sub-optimal or too complex. Con- sidering this scenario, we use existing technologies such as ONOS platform and its applications for developing a self-healing solution considering existing proposals for automated network management, and also developing our methods for comparison and improvement.