1. Department of Electronic Engineering, Tsinghua University, Beijing 100084, China;
2. Information Technology Center, Tsinghua University, Beijing 100084, China
摘要越来越多的网络服务运行在世界各地的数据中心服务器中，并利用数据中心之间的物理链路构建覆盖网络（overlay network），以充分利用数据中心之间链路的网络资源。为了实现高效地构建覆盖网络，以及灵活地调度覆盖网络中传输流量的目的，该文提出了一种基于OpenFlow协议的覆盖网络路由器软件（software overlay router，SOR）。首先，SOR构建在被广泛使用的开源软件Open vSwitch和OpenVPN基础上，既能满足快速部署的要求，又能降低软件的开发和维护成本；其次，SOR使用基于OpenFlow协议的有状态隧道间流量调度（traffic engineering of multi tunnels，TEOFMT）算法，通过维护一张TCP/UDP报文的四元组映射表，依据该映射表来修改报文，最终根据内核路由表将报文转发到相应的隧道中来实现灵活调度流量的目的。该文在中国教育网的数据中心之间搭建覆盖网络，并利用带宽测量工具iPerf和DVTS-SASM（digital video transport system-scalable application specific measurement）来测量覆盖网络的链路带宽，测量数据表明利用SOR来构建覆盖网络能够极大地提高网络链路资源的利用率。
Abstract：More and more network services use overlay networks with physical links between geographically distributed data centers with various kinds of tunnelling technologies to make full use of link resources between data centers. This paper presents an OpenFlow based software overlay router (SOR) to construct a flexible traffic overlay network. The system uses open sourced and widely used software packages, Open vSwitch and OpenVPN, for rapid deployment and low cost software development and maintenance. The system then leverages an OpenFlow based stateful traffic engineering of multi tunnels (TEOFMT) algorithm, to maintain a four tuple mapping relationship between TCP/UDP packets, to modify the packets according to the mapping relationship, and to forward the modified packets to the desired tunnel based on a kernel routing table for flexible traffic transmission scheduling. An overlay network was built using this router with the bandwidth measurement tools, iPerf and DVTS-SASM (digital video transport system- scalable application specific measurement), used to measure the link bandwidth in the network. The results indicate that this router greatly improves the utilization rate.
Data Center Map APS. Colocation data centers.[Z/OL]. (2017-08-07). http://www.datacentermap.com/datacenters.html.
[2]
LIU Y, NIU D, LI B. Delay-optimized video traffic routing in software-defined interdatacenter networks[J]. IEEE Transactions on Multimedia, 2016, 18(5):865-878.
[3]
CHEN F, ZHANG C, WANG F, et al. Cloud-assisted live streaming for crowdsourced multimedia content[J]. Multimedia IEEE Transactions on, 2015, 17(9):1471-1483.
[4]
IZARD R, WANG Q, KRIBBS B, et al. OpenFlow-based live video streaming with GENI cinema[C]//IEEE Conference on Computer Communications Workshops. San Francisco, CA, USA:IEEE, 2016:1039-1040.
[5]
MCKEOWN N, ANDERSON T, BALAKRISHNAN H, et al. OpenFlow:Enabling innovation in campus networks[J]. Acm Sigcomm Computer Communication Review, 2008, 38(2):69-74.
[6]
CAI C X, LE F, SUN X, et al. CRONets:Cloud-routed overlay networks[C]//2016 IEEE 36th International Conference on Distributed Computing Systems (ICDCS). Nara, Japan:IEEE, 2016:67-77.
[7]
KAWASHIMA R, MATSUO H. Non-tunneling edge-overlay model using OpenFlow for cloud datacenter networks[C]//2013 IEEE 5th International Conference on Cloud Computing Technology and Science (CloudCom). Bristol, UK:IEEE, 2013, 2:176-181.
[8]
ALBERT L, FLORIN C, ALBERTO R, et al. Open overlay router[Z/OL]. (2017-04-12). https://github.com/Open-OverlayRouter/oor.
[9]
PFAFF B, PETTIT J, KOPONEN T, et al. The design and implementation of open vSwitch[C]//12th USENIX Symposium on Networked Systems Design and Implementation. Oakland, CA, USA:USENIX Association, 2015:117-130.
[10]
GERT D, DAVID S, JAMES Y, et al. OpenVPN[Z/OL]. (2017-04-12). https://github.com/OpenVPN/openvpn.
[11]
余坤, 包丛笑, 李星. 利用网站服务器测量链路性能[J]. 清华大学学报(自然科学版), 2014, 54(4):474-479. YU K, BAO C X, LI X. Internet path performance measurements using web servers[J]. Journal of Tsinghua University (Science and Technology), 2014, 54(4):474-479. (in Chinese)
[12]
FUJITA T, YAMAMOTO T, IWASE Y, et al. RYU[Z/OL]. (2017-04-12). https://osrg.github.io/ryu/.
[13]
OGAWA A, KOBAYASKI K, SUGIURA K, et al. Design and implementation of DV based video over RTP[C]//10th International Packet Video Workshop. Cagliari, Italy:University of Cagliari, 2000.
[14]
BAO C X, LI X, JIANG J P, et al. Scalable application-specific measurement framework for high performance network video[C]//17th International Workshop on Network and Operating Systems Support for Digital Audio & Video. Urbana-Champaign, IL, USA:ACM, 2007:87-92.
2018, Vol. 58 
