Four elements of Enterprise Core switch
The core switch shoulders the important task of information exchange center, so it must be a full-speed, non-blocking switching equipment. With the increase of port number and load, its performance should be on the rise, which requires the core switch to have high design parameters. For example, "the backplane bandwidth is 128G, the forwarding rate can be up to 96 million packets / s, and the load balance can be achieved on two control engines of a chassis", and so on. Such devices can enable enterprise users to realize full-line speed switching at full load. In terms of performance improvement, excellent core switching devices are no longer simply pursuing high parameters, but more close to the user's design.
Exchange structure At present, the network product structure design, there are many different ideas and ways. One is a centralized switching system, such as a control engine module with very strong performance and function in the center, and a simple interface card on the interface module. All the data are transmitted to the central card through the backplane for processing. After processing, it will be sent back by the backplane. The advantage of this design idea is that it can help users reduce the cost of the network from the ports of each module. All the data exchange depends on the data processing card of the core card. With the increase of the number of boards, the performance of the whole switch tends to decline, that is, the number of ports of the chassis is inversely proportional to the performance. In addition, the latest distributed switching architecture can be adopted. On each interface card, all data exchange, routing, forwarding and security filtering can be handled by the local computer, and local control and forwarding can be handled by multiple boards. The advantages are that the performance of the whole machine increases linearly with the increase of the number of boards and ports, and the disadvantage is that each card needs to add the corresponding processing chips, and the cost will be higher.
Bandwidth restriction In the past, most routing methods were based on the target address, while more complex applications needed more advanced policy routing. Take the setting up of a campus network as an example. There is also an education network export in the campus network. When deciding what flow to use and what kind of exit to use, the school needs to apply strategic routing to solve the problem. This is more flexible than routing based on the destination address.
From the actual situation, almost no network construction in large and medium-sized enterprises is still a blank, most of them are upgrading the network, which inevitably put forward the expansibility of the core switching equipment. Can you be compatible with the various manufacturers and brands of network products already available at the access end?
Can we keep up with the current development trend of network equipment, adapt to the development of network and realize smooth upgrade? Ten-megabit Ethernet standards were introduced, and in the near future there will be 100, 000 megabytes or even millions of megabytes of Ethernet released. Then, whether the core switching equipment that enterprises spend a lot of money to buy can play different roles smoothly and stably in the network of different periods, which must be considered when the enterprise carries on the basic cost investment.
In terms of hardware, are all parts of the hardware completely based on the design without a single point of failure, and can it guarantee continuous reliability? In software, can the reliability and fault tolerance of the network be guaranteed? These are issues of great concern to users. For example, the chassis is designed in accordance with the "five nine" telecommunications level standards and does not have a single point of failure per se; the backplane is designed with a passive backplane structure (that is, the backplane itself is passive and no power supply is required; and the power source uses a N1 redundancy mechanism, It is redundant in the switching and control engine, and the modules support live hot plug and plug. These characteristics are helpful to improve the reliability of the equipment. The small part of the switch that is worth mentioning is the design of the power supply. At present, in some telecom or operator projects, many places require to support 48V DC power supply. In the past, Chinese users often needed higher cost to support 48V DC modules when purchasing network equipment. In addition to the fault-tolerant mechanism in hardware, taking some measures on the protocols of each layer can help users improve their reliability, such as layer 2 spanning tree mechanism, which supports 802.1W and 802.1S, which can make convergence faster and more efficient. On port redundancy, 802.3ad. it can realize load balance between physical ports, which is a redundant backup mechanism, and support VRRP standard routing redundancy protocol in layer 3. At the same time, using the characteristics of some routing protocols, it can help users to establish multiple equivalent paths to the same destination, and realize load balancing and redundant backup of traffic.
How can network switching devices help network managers play a role in security features when an attack virus or similar hacking occurs? Some core switches add state-based firewall features to help users prevent and reject viruses and malicious attacks by hackers such as fragment attacks. In addition, some switches also support VPN in security, can prevent address embezzlement on the port, all security filtering mechanism can be reflected by hardware. At present, the focus of the user is QoS, VoIP requires the most QoS of the network. With the continuous development of Internet and client applications, the network becomes more and more abundant. The core switching devices must meet all kinds of requirements on the same network hardware platform, and ensure the quality of service to the users at the same time. In this respect, the correct design idea of the core switch is to provide different guaranteed bandwidth and service capacity according to the user's different consumption situation.
Related Industry Knowledge