SDN (Software Defined Network), NFV (Network FuncTIon Virtulizaiton), NV (Network VirtulizaTIon), three network terms are hot words in the telecommunications industry and IT industry, they are closely related to carrier networks, cloud computing, virtualization, so frequently In different fields and different occasions, many people confuse them. Here, I will talk about the SDN, NFV and NV that I understand in my previous study and experience.
Talking about SDNSDN is a software-defined network, and the Open Network Foundation (ONF) defines a software-defined network as three characteristics:
Control plane is separated from the forwarding plane
Control plane centralization
Network programmable
People who understand the existing network can understand at a glance that our current network is non-SDN, that is, hardware-defined, regardless of the carrier backbone network, metropolitan area network, access network, data center network, and enterprise campus network and office network. The network is characterized by:
The control component is integrated with the forwarding component, and is embodied in a device-specific, closed, tightly coupled control and forwarding plane;
The working mode of the control plane is distributed. Each device independently learns the network topology through Layer 2 and Layer 3 protocols to form a forwarding table and a routing table. The packet forwarding is determined autonomously.
Not programmable, the network deployment through the device manufacturer's unique command line, the command line is different, there is no standard.
The ideal network depicted by SDN should look like this:
Forwarding plane generalization (the goal of Openflow), the development, production, and manufacturing of forwarding devices no longer have a threshold. Manufacturers only need to assemble general-purpose chips into devices, and manufacturers' device differentiation disappears. The forwarding device is controlled by the controller, and the forwarding information is sent by the controller.
The controller, Controller, has high performance (controls a fairly large scale network), deploys complex services (advanced services), and supports high availability. The Controller can form a cluster group to support a larger network.
The open northbound interface, application or cloud platform can call the controller through this interface to implement the network services required by the service.
Here, the familiar definition of SDN and the ideal network for SDN implementation are because the industry has different interpretations of SDN. There is a saying that SDN is programmable in the third feature. The first and second features are not so important. My opinion is that if the control is not concentrated, a certain degree of programmability can be achieved in the existing equipment vendors, but the benefits are limited. The limitations are:
No global topologyConfiguration context consistency is not guaranteed
No unified network running status information
Therefore, the three characteristics of SDN are complementary to each other. Only the organic combination between the three can truly solve the constraint of the hardware-defined network, and the upper-layer service can call the network as much as it wants to serve it.
Google's B4 and Facebook's latest data center have proved this, and its core construction ideas are emphasizing Controller's full control of the network. It can be seen that it is necessary to emphasize the satisfaction of the first and second features.
Talking about SDN needs to mention Openflow (although it appears to be just a southbound interface protocol), because they are presented at the same time and closely related, here I talk about Openflow.
Since the first day of SDN, the academic community represented by Professor Nick of Stanford University has been working on the generalization of the forwarding plane. Openflow fully reflects this idea. But everyone knows that the progress of Openflow has not been smooth so far, for three reasons: First, the current commercial chip can not effectively support the multi-field matching defined by Openflow; Second: the support of commercial chip to match, action multi-stage pipeline is also very Limited; the third point is not directly related to Openflow. The interoperability between the controller+switch scheme introduced by various manufacturers in the industry and the existing network is too simple to meet the production requirements, making the network formed an island and can only be used for specific occasions ( Such as research and testing), it is not a general solution that can be fully promoted.
Despite the current situation, Openflow's efforts to achieve generalization at the forwarding level are of great significance. Assuming that Openflow's standards are fixed, the commercial production of OpenFlow-based chips is no less important to the network than to x86 for the server industry.
Talking about NFVNFV is an ETSI (the European Telecommunications Standards Institute) that aims to reduce the expensive cost of network construction by replacing the current dedicated hardware equipment provided by telecom equipment vendors to operators by adopting common hardware and popular virtualization technologies. . The supporters behind NFV are the giants in the telecommunications industry.
IT industry people don't know much about telecom networks, thinking that NFV's goal is to replace the data center switches, routers and other devices with general-purpose servers. In fact, NFV covers a wide range. In addition to data networks, there are fixed networks, mobile networks and transmission networks in telecom networks. There are more than one hundred types of network elements involved (the experts in the telecom industry are difficult to be accurate). Speaking of how many kinds of network elements are in the telecommunications network, are covered by NFV.
Since the goal of NFV is to replace dedicated telecommunications equipment with general purpose servers, then you need to understand their respective characteristics. Dedicated telecommunication equipment is characterized by high performance, high reliability and strong scalability. The strength of the general-purpose server is that it has strong computing power, standardization, and mature and open virtualization technology support.
High efficient charging speed for Acer laptop, stable current outlet can offer power for the laptop at the same time charge the laptop battery. The best choice for your replacement adapter. The DC connector is 5.5*1.7mm or 3.3*1.0mm. We can meet your specific requirement of the products, like label design. The plug type is US/UK/AU/EU. The material of this product is PC+ABS. All condition of our product is 100% brand new.
Our products built with input/output overvoltage protection, input/output overcurrent protection, over temperature protection, over power protection and short circuit protection. You can send more details of this product, so that we can offer best service to you!
Laptop Adapter For Acer,Charger For Acer,Acer Laptop Adapter ,Ac Adapter For Acer
Shenzhen Waweis Technology Co., Ltd. , https://www.waweisasdapter.com