ISSN 1000-1239 CN 11-1777/TP


    Default Latest Most Read
    Please wait a minute...
    For Selected: Toggle Thumbnails
    Journal of Computer Research and Development    2018, 55 (4): 673-674.  
    Abstract759)   HTML6)    PDF (457KB)(519)       Save
    Related Articles | Metrics
    Network Function Virtualization Technology Research
    Zhou Weilin, Yang Yuan, Xu Mingwei
    Journal of Computer Research and Development    2018, 55 (4): 675-688.   DOI: 10.7544/issn1000-1239.2018.20170937
    Abstract2170)   HTML20)    PDF (2839KB)(1501)       Save
    Network function virtualization (NFV) is a technology that realizes network functions based on virtualization technology and standard commercial servers, switches, and storage, so as to substitute traditional middleboxes that are dedicated devices. NFV is proposed to save construction and operation cost for network service providers. NFV can also improve flexibility and scalability of network services, and can motivate development and deployment of new network functions. Currently, NFV is still under developing, and problems remain in system performance, management and orchestration, reliability, availability, security, and programmability. There have been a number of studies to address these problems. In this paper, we review the NFV architecture and basic technologies, and summarize the key problems to be solved. Then, we categorize the existing studies by a “four quadrants” method. We analyze and compare a number of typical solutions, and summarize the advantages and overheads. Finally, we discuss the research trends of NFV.
    Related Articles | Metrics
    High Performance Load Balancing Mechanism for Network Function Virtualization
    Wang Yuwei, Liu Min, Ma Cheng, Li Pengfei
    Journal of Computer Research and Development    2018, 55 (4): 689-703.   DOI: 10.7544/issn1000-1239.2018.20170923
    Abstract908)   HTML3)    PDF (4709KB)(772)       Save
    Network load balancer is an important middle box component for NFV(network function virtualization) which can realize load distribution accurately, through which telecom operators can improve the business ability and extend capacity flexibly and efficiently. In this paper, we design and implement a high performance NFV-oriented network load balancing mechanism and system based on DPDK framework which is capable of running on common virtualization platforms. HVLB is designed based on the SDN (software defined networking) principle which splits the data forwarding and scheduling strategy effectively. It realizes the multi-cores and multi-queues based data processing architecture in user space, and ensures the data access isolation and task handling balance meanwhile. It uses the comprehensive abilities including network transmitting and computing ability to select the objective NF and forward the packet, which can guarantee the network performance and deliver the data packet accurately at the same time. Evaluations based on a prototype of KVM platform show that our mechanism significantly improves the performance of packet processing and forwarding compared with the LVS, and it also obtains the line speed processing with 64 byte UDP packet.
    Related Articles | Metrics
    Real-Time Link Fault Detection as a Service for Datacenter Network
    Wang Junxiao, Qi Heng, Li Keqiu, Zhou Xiaobo
    Journal of Computer Research and Development    2018, 55 (4): 704-716.   DOI: 10.7544/issn1000-1239.2018.20170941
    Abstract1192)   HTML3)    PDF (3319KB)(759)       Save
    In large scale datacenter network, link fault detection is an important way to guarantee network connectivity and the performance of large-scale online applications. Currently, the function of link fault detection is provided by middlebox or switches. With the development of software defined networking (SDN) and network function virtualization (NFV), many network functions are decoupled from the hardware devices, while being deployed in the cloud as services. However, the existing methods of link fault detection face some challenges, such as time consuming, high usage of bandwidth, and server overload. To tackle these challenges, we first analyze the existing work on link fault detection. Then we propose the concept of probe matrix and the probe matrix optimization based link fault detection method. We also design a service framework by combining the link fault detection controller and the SDN controller. Finally, the simulation results show that the proposed method significantly outperforms the existing work in detection period, usages of bandwidth and server CPU with tolerable computational overheads for probe matrix optimization.
    Related Articles | Metrics
    DrawerPipe: A Reconfigurable Packet Processing Pipeline for FPGA
    Li Junnan, Yang Xiangrui, Sun Zhigang
    Journal of Computer Research and Development    2018, 55 (4): 717-728.   DOI: 10.7544/issn1000-1239.2018.20170927
    Abstract811)   HTML3)    PDF (3237KB)(514)       Save
    In public cloud, flexible network functions are required to enforce network isolation, service-level agreement and security for multi-tenants. While software-based network functions are flexible, they have limited capacity with low processing throughput and induce high latency. FPGA has good programmability and high processing throughput, and it is appealing due to the balance between hardware performance and software flexibility. However, how to use FPGA to realize network function lacks a unified and reconfigurable architecture. This paper presents DrawerPipe, a reconfigurable pipeline model. This module abstracts the packet processing into five standard “drawers”. And operators can load their modules in these “drawers” which are combined as a packet processing pipeline. As the drawers are independent from each other, the modules loaded in different drawers can be excurted in parallel. Furthermore, we add a function-independent programmable interface between modules to adapt the communication format between different modules, which also helps to release the constraint imposed by the interface definition. Finally, we implement a variety of network functions based on DrawerPipe. The result shows that DrawerPipe not only has good scalability, but also has the advantages of wire-speed processing performance and high resource utilization, which can be used for rapid deployment of network functions.
    Related Articles | Metrics
    Packets Transmission with Multiple Levels of Credibility and Routing Calculation Based on Virtual Topologies
    Chen Wenlong, Zhao Yirong, Xiao Rong, Tang Xiaolan, Xu Ke
    Journal of Computer Research and Development    2018, 55 (4): 729-737.   DOI: 10.7544/issn1000-1239.2018.20170946
    Abstract628)   HTML4)    PDF (1977KB)(526)       Save
    The credibility of routers and forwarding paths in the Internet has been a popular topic. Not only network equipment of different vendors, but also the same one in different management environments has different credibility. The network flows with diverse credibility requirements are supposed to be transmitted along paths with the corresponding credibility levels (CR). In this paper, the credible transmission mechanism with multiple levels (CETML) is proposed, and the fundamental credible management strategies are suggested. Both routers and IP prefixes are associated with a CR, and the CR of a network flow is obtained according to its source and destination IP addresses. CETML constructs different virtual topologies for every transmission network with different CR, and insures that IP packets is forwarded by the routers whose CR is not less than the CR of these packets. Because the forwarding entries include multiple next hops, a small quantity of additional memory overhead is introduced in CETML. Analyzing the relevancy of the multi-level virtual topologies, we design a new routing calculating method based on Floyd algorithm in SDN environment. All the routing tables of virtual topologies can be achieved during the process of successive iterating calculation. Compared with current typical routing algorithms, the calculation time of CETML is significantly reduced.
    Related Articles | Metrics
    Delay-Aware Resource Scheduling Optimization in Network Function Virtualization
    Xu Ran, Wang Wendong, Gong Xiangyang, Que Xirong
    Journal of Computer Research and Development    2018, 55 (4): 738-747.   DOI: 10.7544/issn1000-1239.2018.20170926
    Abstract753)   HTML7)    PDF (3483KB)(1083)       Save
    Network function virtualization (NFV) aims to implement software-enable network functions so as to replace proprietary hardware devices in traditional networks. In response to the growing need for intensive resource, the software-oriented network functions bring issues such as the management of virtual network functions, low-latency scheduling, and efficient allocation of virtual network resources. The virtual network function scheduling problem itself is NP-hard. In order to ensure a good user experience on the specific issues of scheduling delays for virtual network function resources, it is necessary to ensure that network resources are reasonably allocated to prevent over-provisioning of resources and to maintain low end-to-end latency. In this paper,we formulate the problem as an integer linear programming model. In addition, to meet the characteristics of high dynamic network, we design a heuristic algorithm based on greedy. This algorithm firstly constructs the auxiliary graph, and then selects the node resources according to the analysis of the delay impact among different service chains, which considers the influence of network propagation delay and adopts multi-path transmission scheme for the multipoint processing function. The final experimental results show that our algorithm can effectively solve the model and reduce the overall network service latency by 5% to 15% compared with the previous research.
    Related Articles | Metrics
    A Hierarchical Method for Survivable Service Function Chain Embedding
    Liu Yi, Zhang Hongqi, Yang Yingjie, Chang Dexian
    Journal of Computer Research and Development    2018, 55 (4): 748-767.   DOI: 10.7544/issn1000-1239.2018.20170938
    Abstract684)   HTML4)    PDF (3734KB)(488)       Save
    Aiming at embedding the service function chain (SFC) in substrate network under the condition of single-node or single-link failure, this paper proposes a hierarchical method for survivable SFC embedding. The method takes into account both improving the survivability of SFC and reducing the resource consumption of substrate network. For the key service function chain (KSFC) delivering essential services, it pre-allocates backup resources for possible failures in the future. Meanwhile, for the normal service function chain (NSFC) delivering common services, it quickly re-embeds the failed nodes and links after a failure has occurred. Firstly, taking into account minimizing the service latency of SFC, we provide two mixed integer linear programming (MILP) formulations for the survivable embedding problem of KSFC and NSFC. Secondly, we propose two heuristic algorithms to solve the formulations quickly, namely, the primary-backup service path building algorithm (PBSP-Bld) for the KSFC and the failed service path restoring algorithm (FSP-Res) for the NSFC. The PBSP-Bld not only embeds nodes and links alternately based on greedy strategy to reduce the service latency of SFC, but also establishes bridging path between the primary and the backup service paths, which improves the speed of path switching and reduces the packet loss rate during path switching. Additionally, the FSP-Res finds the best re-embedding positions for failed nodes by means of the max-flow problem, which increases the number of recovered NSFCs. It also uses the modified version of Dijkstras shortest path algorithm to select re-embedding paths with low latency. Experimental results demonstrate the good performance of the heuristic algorithms under different network conditions. Moreover, the proposed method keeps the success ratio of running SFC above 592% under the simulated network environment.
    Related Articles | Metrics
    A Backup and Recovery Mechanism for Security Service Chain Fault in Network Function Virtualization Environment
    Huang Rui, Zhang Hongqi, Chang Dexian
    Journal of Computer Research and Development    2018, 55 (4): 768-781.   DOI: 10.7544/issn1000-1239.2018.20170942
    Abstract763)   HTML0)    PDF (3758KB)(570)       Save
    Considering the fault problem of security service chain (SSC) in a network function virtualization (NFV) environment, this paper proposes a backup and recovery mechanism for SSC based on proportional resource reservation. According to the proactive processing idea, it divides the resource proportionally in a substrate network and constructs a candidate set for each substrate node/link beforehand. When the node fault suddenly occurs, it chooses the fault recovery targets in the candidate set and allocates the reserved backup resources. It solves the node fault remapping problem immediately by using the improved discrete particle swarm optimization (DPSO) algorithm, decreasing the occupancy of resources while increasing the repair rate. When the link fault suddenly occurs, it redirects the affected traffic to the available links in the candidate set by changing the traffic splitting-rate of the substrate path. We design the dynamical path splitting algorithm to solve the link fault redirect problem effectively, maximizing the residual value of the underlying substrate network resources. The simulation experiment verifies the proposed algorithm from two aspects: one is the adaptability under different substrate network environments and the other is the validity under different fault models. In addition, we also make a preliminary explore to the appropriate value of the main proportion for the impact of our proposed backup and recovery mechanism.
    Related Articles | Metrics