Software-Defined Networking (SDN) opened up new opportunities in networking with its concept of the segregated control plane from the data-forwarding hardware, which enables the network to be programmable, adjustable, and reconfigurable dynamically. These characteristics can bring numerous benefits to cloud computing, where dynamic changes and reconfiguration are necessary with its on-demand usage pattern. Although researchers have studied utilizing SDN in cloud computing, gaps still exist and need to be explored further. In this article, we propose a taxonomy to depict different aspects of SDN-enabled cloud computing and explain each element in details. The detailed survey of studies utilizing SDN for cloud computing is presented with focus on data center power optimization, traffic engineering, network virtualization, and security. We also present various simulation and empirical evaluation methods that have been developed for SDN-enabled clouds. Finally, we analyze the gap in current research and propose future directions.

A Taxonomy of Software-Defined Networking (SDN)-Enabled Cloud Computing

DRL-R: Deep reinforcement learning approach for intelligent routing in software-defined data-center networks

In software-defined networking (SDN) paradigm, where the control and data plane are separated, the scalability of the SDN controller in the control plane is critical and can affect the overall network performance significantly. To improve controller scalability, efforts have been put into enhancing the capability of SDN switches in the data plane, to make them more autonomous in providing routine services without consulting the controller. In this regard, we investigate the service placement problem on SDN switches aiming at minimizing the average accumulated service costs for end users. To solve this problem, we propose a novel reinforcement-learning-based algorithm with guaranteed performance and convergence rate, called Q-placement. Comparing to traditional optimization techniques, Q-placement exhibits many appealing features, such as performance-tuneable optimization and off-the-shelf implementation. Extensive evaluations show that Q-placement consistently outperforms benchmarks and other state-of-the-art algorithms in both synthetic and real networks. Moreover, these evaluations reveal insights into how the network topological properties (e.g., density), servicing capacities, and controller's roles affect the accumulated service costs, which is useful in service planning tasks.

Q-Placement: Reinforcement-Learning-Based Service Placement in Software-Defined Networks

The considerable increase of multimedia services, such as video-on-demand (VoD) services, is a significant contributor to the total Internet traffic. Software-defined networking (SDN) and information-centric networking (ICN) are two promising technologies that can be combined to facilitate video delivery and to reduce network delays. In this paper, we first formulate the caching decision problem as a 0-1 integer linear programming (ILP) problem. Second, in contrast to existing approaches that solve the formulated ILP problem by assuming all future video requests are known, we consider the impact of the time scale, which transforms the static 0-1 ILP problem into a dynamic problem. By solving the dynamic 0-1 ILP problem, we find more accurate optimal solutions compared to existing approaches. Third, since the formulated 0-1 dynamic ILP problem is NP-hard, we leverage the in-network caching of ICN and the global view of the SDN controller to propose a novel SDN-based caching decision policy. Finally, extensive evaluations are performed, and the results demonstrate that the proposed SDN-based caching decision policy provides solutions that are close to the optimum in substantially less computation time. The SDN-based caching decision policy also outperforms existing practical ICN caching decision policies in terms of the cache hit ratio and the average number of hops, which are directly related to the video delivery latency. Moreover, the SDN-based caching decision policy can substantially reduce the number of generated and broadcasted interest packets, which is a shortcoming of the current ICN.

An SDN-Based Caching Decision Policy for Video Caching in Information-Centric Networking

In software-defined data-center networks, Elephant flow/Mice flow/Coflow coexist and multiple resources (bandwidth, cache and computing) coexist. However, the conventional routing methods cannot overcome the large gap between different performance requirements of flow and efficient resource allocation. Therefore, this paper proposes DRL-R (Deep Reinforcement Learning-based Routing) to bridge this gap. First, we recombine multiple resources (node’s cache, link’s bandwidth) by quantifying the contribution score of them reducing the delay. This actually converts the performance requirements of flow into resource requirements of this flow, hence, the routing problem can be converted into a job-scheduling problem in resource management. Second, a DRL agent deployed on an SDN controller continually interacts with the network for adaptively performing reasonable routing according to the network state, and optimally allocating network resources for traffic. We employ Deep Q-Network (DQN) and Deep Deterministic Policy Gradient (DDPG) to build the DRL-R. Finally, we demonstrate the effectiveness of DRL-R through extensive simulations. Benefitted from continually learning with a global view, DRL-R can improve throughput highest up to 40% and flow completion time highest up to 47% over OSPF. DRL-R can improve the load balance of link highest up to 18.8% over OSPF. Additionally, DDPG has better performance than DQN.

Intelligent Routing based on Deep Reinforcement Learning in Software-Defined Data-Center Networks

Cooperative cache has become a promising technique to optimize the traffic by caching big data in networks. However, controlling distributed cache nodes to update cached contents synergistically is still challenging in designing cooperative cache systems. This paper proposes an SDN-based Cooperative Cache Network (SCCN) for ISP networks, aiming to minimize the content transmission latency while reducing the inter-ISP traffic. Based on the proposed increment recording mechanism, the SCCN Controller can timely capture the change of content popularity, and place the most popular contents on the appropriate SCCN Switches. We formulate the optimal content placement as a specific multi-commodity facility location problem and prove its NP-hardness. We propose a Relaxation Algorithm (RA) based on relaxation-rounding technique to solve the problem, which can achieve an approximation ratio of $1/2$ in the worst case. To solve large scale problems for big data efficiently, we further design a Heuristic Algorithm (HA), which can find a near-optimal solution with three orders of magnitude speedup compared to RA. Specifically, HA can achieve a desirable tradeoff between the transmission delay and the Internet traffic. We implement a prototype based on Open vSwitch to demonstrate the feasibility of SCCN. Extensive trace-based simulation results show the effectiveness of SCCN under various network conditions.

Xuejun Cai

Papers

SDN-Based Big Data Caching in ISP Networks