Shulou(Shulou.com)06/03 Report--

1. Background:

With the rapid development of the Internet and the wide application of real-time services such as live video and network teaching, multiple receivers need to receive the same streaming media data from one or more source nodes at the same time, which greatly increases the information capacity of network transmission and takes up a lot of network bandwidth. For the demand of these applications, the traditional VOD technology not only consumes a lot of source node resources and network bandwidth, but also limits the expansion of the number of users. Comparatively speaking, multicast is a good transmission scheme. Because routers in traditional networks need to be pre-configured before they can dynamically support the joining and leaving operations of multicast subscribers and the generation of multicast trees, and routers in traditional networks do not dynamically choose the transmission path according to the large bandwidth needs of users, it is easy to cause link congestion, can not provide users with better quality of service, and is difficult to deploy on a large scale in the traditional network.

The software-defined network (SDN) framework based on OpenFlow technology has the function of centralized control, can perceive the change of network topology, and has natural advantages in fine-grained path selection, access control and load balancing, which provides a good solution for the realization of IPv6 multicast. In order to solve the problem of IPv6 multicast in SDN network, three functional modules are proposed: group member management, bandwidth topology maintenance and multicast tree construction in SDN controller, so that it is no longer necessary to deploy distributed multicast routing protocols.

II. Brief introduction of SDN

SDN is from the Clean Slate project team of Stanford University, they have a grand goal, is to rebuild the Internet, change the existing rigid network architecture model, in order to build a scalable high-performance modern network architecture. In 2009, SDN concept was selected as one of the top ten cutting-edge technologies of Technology Review. In April 2012, the ONF organization released the SDN white paper, which put forward an idea similar to the operating system: treat all network devices in the network as managed resources, and the controller is equivalent to an operating system to manage these resources. The controller abstracts the network device, maintains the network device, and provides the network device information to the upper application. The upper layer applications configure and manage these network devices through a unified programmable interface to achieve related network application functions, without having to care about the changes of the underlying network topology. The three-layer model of SDN (physical device layer, control layer and application layer) has been widely recognized in the industry.

In the practice of SDN network, scientific research organizations such as OFELIA, Internet2, FIRE and GENI have deployed SDN network in the real environment. Huawei, Ruijie, Cisco, Pica8 and other manufacturers have actively invested human and material resources in research and developed SDN controllers or SDN switches that support OpenFlow protocol. In the aspect of enterprise deployment of SDN network, Google deploys SDN network based on OpenFlow technology in data center on a large scale, which significantly improves the utilization of network resources and reduces the complexity of network operation and maintenance.

The network architecture diagram of SDN is shown as follows: the application layer mainly completes all kinds of upper-layer applications intended by users and unifies the management of network resources. The core function of the control layer is to achieve the network internal switching path calculation and boundary service routing calculation, flow table control and distribution and other functions. The forwarding layer is mainly composed of links between switches to form the basic forwarding network. The forwarding table item needed in the forwarding process is the flow table sent by the controller, and the switch forwards according to the flow table, which does not have the function of logical judgment.

(SDN network architecture diagram)

3. ONOS controller

SDN controller plays a decisive role in the performance of the whole SDN network architecture. At present, there are more than 20 controllers developed by different languages and different institutions, especially many controllers provided by the open source community, such as Nox,RYU,Floodlight,OpendayLight,ONOS and so on. Among them, the ONOS controller is the first commercial grade controller for operators. Supports a variety of southbound interface protocols, abstractly shields protocol differences, and is famous for its high reliability and high availability, which is more suitable for operator scenarios. The design of ONOS is highly hierarchical, modular and abstract. The kernel of ONOS is composed of many subsystems that follow the same architecture design. The core layer follows the object-oriented design principle of "programming for interfaces, not for specific implementation". The service functions provided by subsystems are abstracted into interfaces and presented to top-level applications and underlying protocol plug-ins. The structure of the subsystem is shown in the following figure.

(ONOS controller subsystem structure)

App Component: applications aggregate messages through AdminService and other service interfaces, which are used and manipulated by Manager Component.

Manager Component: the abstraction of the network is protocol-independent and provides a unified northbound interface. Mainly including Manager and Store,Store are responsible for data storage, query, update and east-west synchronization, all data-related operations from Manager will be completed through Store. Manager will also throw events in Store and implement the ListenerService interface, and other applications can listen for events through the ListenerService interface.

Provider Component:Provider is protocol-related and mainly provides abstract data types for the core layer. Provider injects network information into the core layer through the ProviderService interface provided by the core layer. Provider will also expose the Provider interface to the core layer and receive command messages from the core layer. Each Provider needs to be registered with ProviderRegistry before it can be recognized by ProviderService. IV. Architecture implementation

The function of IPv6 multicast is developed and realized in the adaptation layer, core layer and application layer of ONOS controller. It includes the maintenance of switch port state in the adaptation layer, the maintenance of subscriber information and subscriber directly connected switch information in the core layer, and the maintenance of multicast path selection in the application layer. The architecture implementation diagram is shown in the following figure.

(implementation architecture diagram)

The bandwidth topology adapter component maintains the status of the switch and its ports. The OpenFlowDeviceProvider class is an abstract class of switching devices that already exists in the ONOS controller, but it does not provide a way to obtain the real-time port bandwidth. In order to obtain real-time port available bandwidth information, the PortStatsCollector class is designed in the OpenFlowDeviceProvider class.

The group member management component needs to maintain the multicast subscriber and the subscriber switch information, and inform the multicast routing module to choose the path for the multicast subscriber. The implementation of group member management component depends on device management subsystem, packet management subsystem and host management subsystem. This component consists of two parts: multicast subscriber information maintenance and subscriber switch maintenance.

In the multicast routing component, when a multicast subscriber joins a multicast group, the multicast routing component chooses the transmission path for the multicast subscriber according to the current network topology and link bandwidth information. and to consider whether multicast subscribers join a new multicast group or an existing multicast group, there are different routing algorithms for the two situations. If you join a new multicast group, the multicast traffic is transmitted from the multicast sender to the receiver; if you join an existing multicast group, the multicast traffic is copied and forwarded from the switch that forwards the multicast traffic.

5. Experimental results

The data plane simulates six switches with Mininet simulator, and Mininet opens three hosts through xterm command in Mininet simulator. It is a network platform simulator that can create virtual hosts, switches, controllers and links. Mininet hosts run standard Linux network software. Mininet virtual hosts, switches, links and controllers are created by software to make it look like a complete network. In the Mininet simulator, three hosts are opened through the xterm command, the IPv6 address configured for the multicast sender is fc00::1/64, and the IPv6 addresses configured by the two subscribers are fc00::2/64 and fc00::3/4. The three hosts run their own programs to receive multicast traffic and output the source and time of receiving multicast traffic. The experimental results are shown below, where two subscribers can receive the same data at the same time.

Author: Li Dan

Source: Yixin Institute of Technology

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.