Chunming Wu

Mapping Multicast Service-Oriented Virtual Networks with Delay and Delay Variation Constraints

As a key issue of building a virtual network (VN), the VN mapping problem can be addressed by various state-of-the-art algorithms. While these algorithms are efficient for the construction of unicast service-oriented VNs, they are generally not suitable for multicast cases. In this paper, we investigate the mapping problem in the context of virtual multicast service-oriented network subject to delay and delay variation constraints (VMNDDVC). We present a novel and efficient heuristic algorithm to tackle this problem based on a sliding window approach. The primary objective of this algorithm is in two-fold: to minimize the cost of VMNDDVC request mapping, and to achieve load balancing so as to increase the acceptance ratio of virtual multicast network (VMN) requests. The numerical results obtained from extensive simulation experiments demonstrate the effectiveness of the proposed approach and superiority than existing solutions in terms of VN mapping acceptance ratio, total revenue and cost in the long term.

Multicast virtual network mapping for supporting multiple description coding-based video applications

Video applications have gradually become one of the mainstreams of Internet traffic along with the technological advances in recent years. In parallel, network virtualization technique allows Internet Service Providers (ISPs) to be decoupled from the physical network infrastructure operators to provide various video services to customers through tailored Virtual Networks (VNs). In a video multicast VN, the ISP could appropriately determine and configure the virtual nodes and links to deliver customized video service to the affiliated end users. However, available multicast tree mapping algorithms are based on the strong assumption that the locations of source and a collection of destination nodes are known a priori. Also, they have not taken any intermediate nodes between source and destinations into the VN mapping request. These assumptions greatly limit the adoption of advanced video coding techniques. In this paper, the multicast networks mapping for enabling Multiple Description Coding (MDC) based video applications is explored. A novel multicast mapping algorithm, MMPC, is proposed based on path convergence to flexibly identify and configure nodes (with automatic intermediate nodes identification) in multicast trees to meet certain criteria. The suggested approach is described in details and its key characteristics are analyzed through a set of analytical proves. Extensive simulation experiments are carried out to assess its performance (e.g. acceptance ratio, mapping cost and mapping time) against a range of multicast VN request scenarios. The simulation results demonstrate the superiority of MMPC algorithm in terms of multicast mapping efficiency and cost by using the conventional two-phase VN mapping algorithm as a comparison benchmark.

Evolving defense mechanism for future network security

The past few years have witnessed revolutionary advances in network technology. Along with new techniques such as SDN come lots of new network security challenges. Conventional network security mechanisms are incompetent to overcome these challenges, since they are built on a static network configuration that facilitates attackers in finding the weaknesses of a network. In this article, we conceive a novel conceptual network security mechanism, the evolving defense mechanism (EDM), to resolve current and future security problems. EDM is based on a bio-inspired idea of network configuration variations. According to the security requirements of the system, the user, and the network security state, EDM selects an efficient network configuration variation strategy to prevent corresponding security threats. Combined with SDN implementation, EDM resolves security problems from a new angle and is capable of evolving with new network security technology. We sketch a way to implement EDM and present its reference framework, which serves as an ecosystem and coexisting environment for various kinds of network configuration variations. The proposed mechanism avoids the deficiency of conventional mechanisms and has potential to cope with emerging security threats.

Robust dynamic bandwidth allocation method for virtual networks

Multiple virtual networks sharing an underlying substrate network is considered a promising tool to diversify and reshape the future inter-networking paradigm. As a simple and straightforward approach, the static bandwidth allocation in virtual networks (VNs) can often be inefficient in practice, and the adaptive allocation scheme with a small time-scale may lead to the transient network and service instability. Due to the fact that the traffic patterns vary over time, the challenge still remains to meet the expected resources allocation whilst promote the network scalability and robustness. In this paper, based on the robust optimization theory we present a robust dynamic approach which periodically identifies bandwidth allocation to VNs to work reasonable well for a range of traffic patterns over a period of time, rather than certain traffic pattern instance. This problem is formulated as a robust optimization problem using path-flow model aiming to compute the minimum-cost bandwidth allocation. Through the primal decomposition, we present a distributed algorithm which consists of two components running in individual VNs and the substrate network respectively. The numerical result obtained from simulation experiments demonstrates the strength and the effectiveness of the proposed algorithm in terms of convergence and acceptance ratio.

Prototype for customized multicast services in software defined networks

Multicast, transmitting packets from one source to a group of destinations, is a popular service in Internet and now in the datacenter network. However, a unified multicast algorithm cannot satisfy the diverse performance requirements from different users. Hence customized multicast services are proposed in software defined networks (SDN) in this paper. We tackle a few associated technical challenges, and introduce a prototype with OpenFlow1.3 and RYU controller. Our contributions are: (1) we present the modules in detail and explain why they are demanded to implement multicast services in SDN, independent of current Internet-based IGMP protocols and the multicast address; (2) our prototype could sliced multicast trees in the substrate network in accordance to different embedded multicast algorithms; (3) the last hop translation is proposed in access switches to guarantee the multicast flow to be forwarded through the calculated multicast tree and be consumed in end points. Simulation experiments evaluate these performances. And two applications (namely, multicast text transfer service and streaming media application) further validate the feasibility and operability of our customized prototype for multicast services in the real world.

Avoiding the evolved node B buffer overflow by using advertisement window control

In the Long Term Evolution (LTE) network, buffer overflows often occur in the evolved Node B (eNB) when dealing with the traffic bursts from the wired part due to restricted bandwidth or environmental interferences in the wireless part. In this paper, a novel approach is proposed to protect the buffer from overflow by appropriately controlling the advertisement window to enhance the Transmission Control Protocol (TCP) when the buffer becomes congested. The sender will slow down the data sending rate upon the receipt of reduced advertisement window, and hence the buffer overflow will be alleviated. To evaluate the suggested solution, a LTE network model is implemented and validated through extensive numerical experiments. It is demonstrated that the proposed approach can enhance the end to end system performance in terms of network throughput, average packet delay and jitter, as well as packet loss rate.

BABAC: An Access Control Framework for Network Virtualization Using User Behaviors and Attributes

With the application of network virtualization technique, a collection of autonomous and heterogeneous virtual networks (VNs) could co-exist on an underlying shared physical substrate. In such an environment, end users and network resources are distributed across different VNs in a dynamic fashion. From the service providers perspective, efficient access control mechanism needs to be deployed to VNs to guarantee global resource utilization efficiency whilst meet certain operational requirements, e.g. security, service level agreements (SLAs). As a response, this paper presents a scalable and flexible network access control framework within individual or across multiple VN domains: BABAC (Behaviors and Attributes Based Access Control). It is based on user behaviors and attributes where the access decisions are made through two phases: the preaccess control based on three categorized user attributes and the post-access control maintaining the knowledge of network resources utilizations based on three categories of user behaviors. This paper highlights the strength of the proposed approach to be applied in network virtualization environment (NVE) which exhibits a collection of unique characteristics, e.g. VN autonomy, user mobility and multi-homing. Through a comprehensive comparison with the state-of-the-art approaches, the evaluation outcome demonstrates its effectiveness for access control in NVE.

Performance evaluation for SDN deployment: an approach based on stochastic network calculus

The OpenFlow implementations (SDNs) have been deployed increasingly on varieties of networks in research institutions as well as commercial institutions. To develop an OpenFlow implementation, it is required to understand the performance of the network. A few benchmark tools (e.g., Cbench and OFlops) can be used to measure the network performance, while these tools take considerable time to simulate traffic behaviors and generate the required results, therefore extending the development time. In this paper, we present an analytical model, which is based on stochastic network calculus theory, for evaluating the performance of switch to controller. The previous studies show that stochastic network calculus can provide realistic emulation of real network traffic behaviors. Our model is evaluated by using both simulation tool and realistic testbed. The results show the stochastic network calculus based analysis model can realistically measure the network performance of the end-to-end properties between controller and switch.

Virtual network mapping for multi-domain data plane in Software-Defined Networks

Software-Defined Networking (SDN) separates the control plane from the data plane to improve the control flexibility, supporting multiple services with their isolated physical resources. In SDN, the virtual network (VN) mapping is required by network services for allocating these resources in the multidomain SDN. Such mapping problem is challenged by the NP-Completeness of the mapping and business privacy to protect the domain topology. We propose a novel multi-domain mapping algorithm for SDN using a distributed architecture to achieve a better efficiency and flexibility than the traditional PolyViNE approach, meanwhile protecting the privacy. By simulating on a large synthesized topology with 10 to 40 domains, our approach shows 25% and 15% faster than the PolyViNE in time, and 30% better in balancing load on multiple controllers.

A multipath resource updating approach for distributed controllers in software-defined network

Finding effective ways to collect the usage of network resources in all kinds of applications to ensure a distributed control plane has become a key requirement to improve the controller’s decision making performance. This paper explores an efficient way in combining dynamic NetView sharing of distributed controllers with the behavior of intra-service resource announcements and processing requirements that occur in distributed controllers, and proposes a rapid multipathing distribution mechanism. Firstly, we establish a resource collecting model and prove that the prisoner’s dilemma problem exists in the distributed resource collecting process in the Software-defined Network (SDN). Secondly, we present a bypass path selection algorithm and a diffluence algorithm based on Q-learning to settle the above dilemma. At last, simulation results are given to prove that the proposed approach is competent to improve the resource collecting efficiency by the mechanism of self-adaptive path transmission ratio of our approach, which can ensure high utilization of the total network we set up.摘要创新点在分布式 SDN 环境下, 针对各类应用有效地收集网络资源已成为改善控制器决策能力及其性能的关键因素。 论文通过研究分布式控制器收集到的资源请求如何结合快速路径分发机制在域间资源的通告, 更新生成共享资源的动态网络视图, 并能为控制器决策提供有益帮助的方法。 首先, 建立资源收集模型并证明收集过程中存在着囚徒困境问题。 然后, 提出了基于 Q 学习的旁路路径选择算法和分流资源传输算法。 最后, 仿真结果表明自适应路径传输机制能有效地改善分布式控制器的资源收集及决策效率, 并提高全网的资源利用率。