Software Defined Network is new network architecture. One of its components is the controller, which is the intelligent part of SDN. Many controllers such as Floodlight, Open Daylight, Maestro, NOX, POX and many others are released. The question is which controller can perform better in which situations. Many works were done to compare controllers regarding architecture, efficiency and controllers’ features. In this paper, two of the most popular controllers, Floodlight and OpenDaylight are compared in terms of Network QoS parameters such as delay and loss in different topologies and network loads. This paper can help researchers to choose the best controller in different use cases such as clouds and multimedia. The results with 95% confidence interval show that OpenDaylight outperforms Floodlight in low loaded networks and also for tree topology in mid loaded networks in terms of latency. Floodlight can outperform OpenDaylight in heavy loaded networks for tree topology in terms of packet loss and in linear topology in terms of latency. There is no significant difference in performance of Floodlight and OpenDaylight controllers in other cases.

Software Defined Networking (SDN) is an emerging architecture which decouples networks control plane and data plane physically. It makes control plane programmable trough a centralized controller, and builds intelligent and flexible networks. The OpenFlow is one of the most famous SDN protocols, which acts as a southbound interface between control plane and data plane. In this survey, SDN implementation approaches and different southbound interfaces, beside different version of OpenFlow, are introduced. In addition to general architecture of SDN, different wireless architectures are discussed. Here, also potential SDN's applications and research areas including hot topics such as Information Centric Networks, Cloud and datacenters, multimedia, wireless and mobile networks over SDN are reviewed.

In recent years with the improvement in the field of network technology and decreasing of technology cost, lots of data are produced. This massive amount of data needs mechanism for processing and mining information rapidly. In this paper a new Hadoop based network management system with policy approach which is considered hierarchical manager is presented. Storing and processing massive data efficiently are two capability of Hadoop technology by using HDFS and MapReduce. In this paper, processing time is considered as a main factor. As a result it is proved that this management system using policy approach increases the performance of entire system without putting on extra cost for implementation. This system in contrast with pure Hadoop and centralized system is several times more rapid.