• 1

[CF FIBERLINK] Exchange working principle, detailed explanation!

1. What is a switch?

Exchange, switching is according to the needs of the transmission of information, the information to be transmitted by the manual or equipment to the corresponding route to meet the requirements. Broad switch switch is a kind of device that completes the information exchange function in the communication system. This process is an artificial exchange. Of course, now we have already popularized program-controlled switches, the exchange process is automatic. In a computer network system, the concept of exchange is an improvement of the shared working mode. We have introduced the HUB hub is a kind of sharing equipment, HUB itself cannot identify the address, when the same LAN host to B host data, data packets in the network is broadcast transmission, by each terminal, through the verification data Baotou address information to determine whether to receive. That is to say, in this way of working, only one set of data frames can be transmitted on the network at the same time, and if there is a collision, you have to try again. This way is to share the network bandwidth. The switch has a very high-bandwidth back bus and an internal exchange matrix. All ports of the switch are attached to the back bus. After the control circuit receives the packet, the processing port will find the address control table in memory to determine the NIC (network card) of the MAC (hardware address of the network card) to the destination port through the destination port, exchange the opportunity to "learn" the new address and add it to the internal address table. Exchange and switch originated from the telephone communication system (PSTN), we can now see in the old movie: chief (call user) picked up the microphone to a shake, bureau is a row of full wire machine, wearing a headset call lady after receiving connection requirements, put the thread in the corresponding exit, establish connection for two client end, until the end of the call. This can also "segment" the network, where the switch only allows the necessary network traffic through the switch. Through switch filtering and forwarding, it can effectively isolate broadcast storms, reduce the occurrence of false packets and wrong packets, and avoid shared conflicts. The switch can transfer data between multiple pairs of ports at the same time. Each port can be regarded as a separate network segment, and the network device connected to it alone enjoys the entire bandwidth, without having to compete with other devices. When node A sends data to node D, node B can send data to node C at the same time, and both transmissions enjoy the full bandwidth of the network and have their own virtual connections. If the 10Mbps Ethernet switch is used here, the total circulation of the switch is equal to 210Mbps=20Mbps, and the use of the shared HUB of 10Mbps, the total circulation of a HUB will not exceed 10Mbps. In short, the switch is a network device based on MAC address identification and can complete the function of encapsulating and forwarding data packets. The switch can"

2. What is the role of the switch?

"Exchange" is the most frequent word on the Internet today, from bridging to route to ATM to telephone system, it can be used, not exactly what is the real exchange. In fact, the word exchange first appeared in the telephone system, which refers to the exchange of voice signals between two different phones, and the device that completes the work is the telephone switch. So, as originally intended, the exchange is just a technical concept, that is, to complete the forwarding of the signal from the device entrance to the exit. Therefore, all devices as long as they are and meet the definition may be called switching devices. Thus, "exchange" is a broad term that actually refers to a bridging device when it is used to describe the second layer of a data network, and a routing device when it is used to describe the device of the third layer of a data network. The Ethernet switch we often talk about is actually a multi-port second layer network device based on bridge technology, which provides low latency and low overhead access for the forwarding of data frames from one port to another. Thus, there should be an exchange matrix inside the core of the switch that provides a path for communication between any two ports, or a fast exchange bus to send data frames received by any port from other ports. In practical devices, the function of the exchange matrix is often completed by a specialized chip (ASIC). In addition, ethernet switch in the design idea has an important assumption, namely the exchange of the core speed is very fast, so that usually large traffic data will not make its congestion, in other words, the ability to exchange relative to the information and infinite (on the contrary, ATM switch in the design idea is, that the exchange ability of the relative to the information is limited). Although the ethernet tier 2 switch is based on the multi-port bridge, switching has its richer features, which is not only the best way to get more bandwidth, but also makes the network easier to manage.

3 The switch application

As the main connection device of LAN, Ethernet switch has become one of the most popular network devices. With the continuous development of exchange technology, the price of Ethernet switch has dropped sharply, and exchange to desktop has been the general trend. If your Ethernet has a lot of users, busy applications, and a wide variety of servers, and you haven't made any changes to its structure, the entire network performance can be very low. One solution is to add a 10 / 100Mbps switch to Ethernet, which can not only handle regular Ethernet data streams at 10Mbps, but also support fast Ethernet connections at 100Mbps. If the network utilization exceeds 40% and the collision rate is greater than 10%, the switch can help you solve a bit. Switches with 100Mbps fast Ethernet and 10Mbps Ethernet ports can run in full duplex, with dedicated 20Mbps to 200Mbps connections established. Not only are the functions of switches different in different network environments, but also the effects of adding new switches and existing switches in the same network environment. Fully understanding and mastering the traffic mode of the network is a very important factor to play the role of the switch. Because the purpose of using the switch is as far as possible to reduce and filtering data flow in the network, so if a switch in the network due to improper installation location, almost need to forward all the received packets, switch cannot play the role of optimizing network performance, but reduces the data transmission speed, increased the network delay. In addition to the installation location, it may also have a negative impact if switches are also blindly added in networks with low load and low information. Influenced by the processing time of the packet, the buffer size of the switch and the need to regenerate new packets, using a simple HUB is better in this case. Therefore, we cannot simply think that switches have advantages over HUB, especially when the user's network is not crowded and there is a lot of available space, using HUB can make full use of the existing resources of the network.

4. Three switching modes of the switch

1. Straight-through type (Cut Through)
Ethernet switch in direct mode can be understood as a line matrix telephone switch between ports. When the input port detects a data package, it checks the header of the package, obtains the target address of the package, starts the internal dynamic search table to convert it into the corresponding output port, connects at the intersection of input and output, and connects the data packet to the corresponding port to realize the exchange function. With no storage required, the delay is very small and the exchange is very fast, which is its advantage. The disadvantage is that because the packet content is not saved by the Ethernet switch, it cannot check whether the transmitted packets are incorrect and cannot provide error detection capability. Since there is no cache, input / output ports with different rates cannot be directly connected and easy to lose packets.

2. Storage and forwarding (Store & Forward)
Storage and forwarding mode is the most widely used way in the field of computer network. It stores the packets of the input port first, and then conducts CRC (cyclic redundancy code check) check. After processing the error packet, the target address of the packet is removed, and sends the packet into the output port through the search table. Because of this, the storage and forwarding mode has a large delay in data processing, which is its deficiency, but it can detect the data packets entering the switch and effectively improve the network performance. In particular, it can support the conversion between ports at different speeds, maintaining the coordination between high speed ports and low speed ports.

3. Fragment isolation (Fragment Free)
This is a solution somewhere in between with the first two. It checks whether the packet is 64 bytes, and if it is less than 64 bytes, it is false; if it is more than 64 bytes, the packet is sent. This method also does not provide data verification. Its data processing speed is faster than the storage and forwarding mode, but slower than the straight-through mode.

5 Switch classification

Broadly speaking, switches are divided into two types: WAN switch and LAN switch. WAN switches are mainly used in the telecommunication field, providing the basic platform for communication. And LAN switches are applied to local area networks to connect terminal devices, such as PCs and network printers. From the transmission medium and transmission speed can be divided into Ethernet switch, fast Ethernet switch, Gigabit Ethernet switch, FDDI switch, ATM switch and token ring switch. From the scale application, it can be divided into enterprise level switch, department level switch and working group switch. The scale of each manufacturer is not completely the same. Generally speaking, enterprise level switches are rack type, while department level switches can be rack type (less slot number) or fixed configuration type, while working group level switches are fixed configuration type (relatively simple function). On the other hand, from the perspective of application scale, as backbone switches, switches for large enterprises with more than 500 information points are enterprise-level switches, switches for medium enterprises below 300 information points are departmental level switches, and switches within 100 information points are working group level switches.

6 Switch function

The main functions of the switch include
Physical site
Network topology structure
error check
Frame sequence as well as the flow control
VLAN (virtual LAN)
Link convergence
In addition to being able to connect to the same type of networks, switches can also interconnect between different types of networks (such as Ethernet and Fast Ethernet). Many switches today can provide high-speed connection ports that support fast Ethernet or FDDI, etc., to connect to other switches in the network or to provide additional bandwidth for critical servers with large bandwidth usage. In general, each port of the switch is used to connect a separate network segment, but sometimes to provide faster access speed, we can connect some important network computers directly to the switch port. In this way, the key servers and key users of the network will have faster access speeds and support greater information traffic.

About Us

640 (2)

Switch fault classification:

Switch faults can be generally divided into hardware faults and software faults. Hardware failure mainly refers to the failure of the switch power supply, backplane, module, port and other components, which can be divided into the following categories.

(1)Power failure:
the power supply is damaged or the fan stops due to unstable external power supply, or aging power line, static electricity or lightning strike, so it cannot work normally. Damage to other parts of the machine due to the power supply also often occurs. In view of such faults, we should first do a good job of external power supply, introduce independent power lines to provide independent power supply, and add voltage regulator to avoid instantaneous high voltage or low voltage phenomenon. Generally speaking, there are two ways of electric power supply, but due to various reasons, it is impossible to provide dual power supply for each switch. UPS (uninterruptible power supply) can be added to ensure the normal power supply of the switch, and it is best to use UPS that provides voltage stabilization function. In addition, professional lightning protection measures should be set up in the machine room to avoid the damage of lightning to the switch.

(2) Port failure:
this is the most common hardware failure, whether it is fiber port or twisted pair RJ-45 port, must be careful when plugging and plugging the connector. If the fiber plug is accidentally dirty, it may cause the fiber port pollution and can not communicate normally. We often see a lot of people like to live to plug the connector, in theory, it is ok, but this also inadvertently increases the incidence of port failure. Incare during handling may also cause physical damage to the port. If the size of the crystal head is large, it is also easy to destroy the port when inserting the switch. In addition, if a section of the twisted pair attached to the port is exposed outside, if the cable is struck by lightning, the switch port will be damaged or cause more unpredictable damage. In general, a port failure is a damage to one or several ports. Therefore, after eliminating the fault of the computer connected to the port, you can replace the connected port to judge whether it is damaged. For such failure, clean the port with an alcohol cotton ball after the power is switched off. If the port is indeed damaged, the port will only be replaced.

(3) Module failure:
the switch is composed of a lot of modules, such as stacking module, management module (also known as control module), expansion module, etc. The probability of failure of these modules is very small, but once there is a problem, they will suffer huge economic losses. Such failures can occur if the module is being accidentally plugged in, or the switch is being collided, or the power supply is not stable. Of course, the three modules mentioned above all have external interfaces, which is relatively easy to identify, and some can also identify the fault through the indicator light on the module. For example, the stacked module has a flat trapezoidal port, or some switches have a USB-like interface. There is a CONSOLE port on the management module for connecting with the network management computer for easy management. If the expansion module is fiber connected, there is a pair of fiber interfaces. When troubleshooting such faults, first ensure the power supply of the switch and module, then check whether each module is inserted in the correct position, and finally check whether the cable connecting the module is normal. When connecting the management module, it should also consider whether it adopts the specified connection rate, whether there is parity check, whether there is data flow control and other factors. When connecting the extension module, you need to check whether it matches the communication mode, such as using full-duplex mode or half-duplex mode. Of course, if it is confirmed that the module is faulty, there is only one solution, that is, you should immediately contact the supplier to replace it.

(4) Backplane failure:
each module of the switch is connected to the backplane. If the environment is wet, the circuit board is damp and short circuit, or the components are damaged due to high temperature, lightning strike and other factors will cause the circuit board can not work normally. For example, the poor heat dissipation performance or the ambient temperature is too high, resulting in the temperature in the machine, ordering the components to burn out. In the case of normal external power supply, if the internal modules of the switch can not work properly, it may be that the backplane is broken, in this case, the only way is to replace the backplane. But after the hardware update, the circuit plate of the same name may have a variety of different models. In general, the functions of the new circuit board will be compatible with the functions of the old circuit board. But the function of the old model circuit board is not be compatible with the function of the new circuit board.

(5) Cable failure:
the jumper connecting the cable and the distribution frame is used to connect modules, racks and equipment. If a short circuit, open circuit or false connection occurs in the cable core or jumper in these connecting cables, a failure of the communication system will form. From the above perspective of several hardware faults, the poor environment of the machine room is easy to lead to various hardware failures, so in the construction of the machine room, the hospital must first do a good job of lightning protection grounding, power supply, indoor temperature, indoor humidity, anti-electromagnetic interference, anti-static and other environment construction, to provide a good environment for the normal work of network equipment.

Software failure of the switch:

Software failure of a switch refers to the system and its configuration failure, which can be divided into the following categories.

(1)system mistake:
Program BUG: There are defects in the software programming. The switch system is a combination of hardware and software. Inside the switch, there is a refreshing read-only memory that holds the software system necessary for this switch. Due to the design reasons at that time, there are some loopholes, when the conditions are appropriate, it will lead to the switch full load, bag loss, wrong bag and other conditions. For such problems, we need to develop the habit of often browsing the websites of device manufacturers. If there is a new system or a new patch, please update it timely.

(2) Improper configuration:
Because to different switch configurations, network administrators often have configuration errors when configswitches. The main errors are: 1. System data error: system data, including software setting, is used to define the whole system. If the system data is wrong, it will also cause the comprehensive failure of the system, and has an impact on the whole exchange bureau.2. Bureau data error: The bureau data is defined according to the specific situation of the exchange bureau. When the authority data is wrong, it will also have an impact on the entire exchange office.3. User data Error: The user data defines the situation of each user. If the user data is set incorrectly, it will have an impact on a certain user.4, the hardware setting is not appropriate: the hardware setting is to reduce the type of the circuit board, and a group or several groups of switches set on the circuit board, to define the working state of the circuit board or the position in the system, if the hardware is not set correctly, will lead to the circuit board does not work properly. This kind of failure is sometimes difficult to find, need a certain amount of experience accumulation. If you cannot determine if there is a problem with the configuration, restore the factory default configuration and then step by step. It is best to read the instructions before the configuration.

(3) External factors:
Due to the existence of viruses or hacker attacks, it is possible that a host may send a large number of packets that do not meet the encapsulation rules to the connected port, resulting in the switch processor is too busy, resulting in the packets too late to forward, thus leading to buffer leakage and packet loss phenomenon. Another case is the broadcast storm, which not only takes up a lot of network bandwidth, but also takes up a lot of CPU processing time. If the network is occupied by a large number of broadcast data packets for a long time, the normal point-to point communication will not be conducted normally, and the network speed will slow down or paralyze.

In short, software failures should be more difficult to find than hardware failures. When solving the problem, it may not need to spend too much money, but need more time. The network administrator should develop the habit of keeping logs in their daily work. Whenever a fault occurs, timely record the fault phenomenon, fault analysis process, fault solution, fault classification summary and other work, in order to accumulate their own experience. After solving each problem, we will carefully review the root cause of the problem and the solution. In this way can we constantly improve ourselves and better complete the important task of network management.

Post time: May-15-2024