This paper teaches you how to set up a small LAN, so that you can master the process of setting up a small LAN from beginning to end. With the development of companies or enterprises, the leaders of some units began to realize the practical necessity of building local area networks internally. However, due to the constraints of key factors such as the shortage of funds, many leaders require the computer managers of their own units to undertake the task of LAN construction. Therefore, this paper introduces the detailed methods and operation skills of self-built LAN, hoping to play a role in attracting jade.
Office Network The main function of small office LAN is to realize network communication and enjoy network resources. After forming a small local area network, we can * * enjoy office equipment such as files, printers and scanners, and also use the same modem to surf the Internet * * * to enjoy Internet resources.
The role of small local area network Small local area network refers to a computer network with small space, small scale and low network construction cost. Commonly used in offices, school multimedia classrooms, game halls, Internet cafes, and even two computers at home can form a small LAN.
As we all know, primary and secondary schools have limited funds and need dozens of students to get on the computer at the same time. If every machine saves hard disk and CD-ROM, it can save a lot of money. Using a diskless workstation is a better method. This kind of network not only saves money, but also facilitates management and maintenance.
In addition, you can also set up an intranet-type local area network, and you can access the Internet only through a telephone line. Usually, some popular websites are "cloned" by offline browsers, stored on the server, and a home page is made to link these websites. Students visit the homepage on the server, and visit popular websites through the homepage, which becomes the information transmission in the local area network. Due to the high transmission speed of local area network, students will not feel the pain of an old cow pulling a broken car when using telephone lines. At the same time, because a lot of information is transmitted in the local area network, less data is transmitted through the modem, and students who occasionally visit Internet websites will not feel slow.
Network planning Network planning mainly refers to the choice of operating system and the determination of network structure. At present, there are two kinds of network planning widely used in small LAN: peer-to-peer network and client/server network.
Reasonable setting of switch
Switch is an important data exchange device in LAN, and the correct and reasonable use of switch can also improve the data transmission performance in the network. The author once configured the switch port as 100M full duplex, and installed an EISA network card with the model of Intel 100 m on the server. Everything is normal after installation, but the speed of data transmission becomes extremely slow under heavy traffic load. Finally, it is found that this network card does not support full duplex. After changing the switch port to half duplex, the fault disappeared. This indicates that the ports of the switch and the network card must have the same speed and duplex mode. At present, there are many adaptive network cards and switches, which should be able to adapt to the speed and duplex mode correctly in principle, but in fact, due to inconsistent brands, the full duplex mode cannot be realized correctly. It is clear that the server network card is set to full duplex, but the duplex light of the switch is not on, which can only be solved by manual forced setting. Therefore, when setting the parameters of network devices, we must refer to the parameters of network devices on servers or other workstations, so as to make each device work as well as possible.
Client/server network.
There is at least one dedicated server in the client/server network to manage and control the operation of the network. All workstations can * * * enjoy the software and hardware resources in the file server. Compared with peer-to-peer network, client/server network has the advantages of stable operation, safe information management, convenient expansion of network users and easy upgrade.
The disadvantage of client/server network is that it needs special file server and corresponding external connection equipment (such as HUB), and the network construction cost is high and the management is complicated. Client/server network is suitable for units with a large number of microcomputers, relatively scattered locations and large amount of information transmission.
2. Peer to Peer Network
Peer-to-peer network does not use dedicated servers, and each site is both a network service provider-server and a network service applicant-workstation, so it is also called peer-to-peer network. Peer-to-peer network is easy to build, low cost and easy to maintain, which is suitable for units with fewer microcomputers and concentrated layout.
In the peer-to-peer network, each microcomputer not only has all the independent rights of a single computer, but also enjoys the processing power and storage capacity of each computer in the network for information exchange. Peer-to-peer network has unique advantages in the case of small hard disk capacity and slow computer processing speed. But the disadvantage of peer-to-peer network is that the files stored in the network are very scattered, which is not conducive to the confidentiality of data. At the same time, the data bandwidth of the network is greatly limited and it is not easy to upgrade.
3. Network operating system
At present, there are three common LAN operating systems: NetWare, Windows NT Server and Unix.
The NetWare operating system has low requirements for network hardware (only 286 workstations are needed) and is compatible with DOS commands. Its application environment is similar to DOS, with rich application software and perfect and reliable technology, especially the installation of diskless workstations is more convenient. Therefore, it is best to choose NetWare for computers with low configuration or low overall grade when networking. At present, the commonly used versions are 3. 1 1, 3. 12, 4. 10 and other Chinese and English versions. NetWare server supports diskless stations and games very well, and is often used in teaching networks and game halls.
Windows NT Server is a very powerful network operating system, which can run almost all the newer popular software safely and simply. Windows NT also supports multiprocessor operation, which provides higher scalability for the network and more memory for applications that consume more memory. Commonly used versions of Windows NT are 4.0 and 5.0. It is suggested to use Windows NT Server operating system to set up office and industrial and commercial enterprise networks.
For ordinary network users, Unix is not easy to master, and Unix is basically not used in small LAN.
Network structure The interconnection of nodes in a network is called the topology of the network. Bus topology and star topology are commonly used in small LAN.
1. bus structure
The bus structure uses a single transmission line (bus) to connect all nodes (workstations and servers) in the network. Signals sent by any station can propagate along the bus and be received by all other nodes. Small LAN workstations and servers with bus structure usually use BNC interface network cards. T-type BNC interface connector and 50 ohm coaxial cable are used for series stations, and terminal resistors need to be installed at both ends of the bus.
Advantages: short cable length, simple structure, low cost and easy wiring. It is very easy to set up a small serial LAN, which is suitable for units with fewer computers and concentrated layout, such as small office networks and game networks.
Disadvantages: not centralized control; Fault detection is needed between nodes on the network, and the failure of one connecting node may lead to the communication failure of the whole network; When expanding the trunk length of the bus, it is necessary to reconfigure the repeater, cut off the cable and adjust the terminator. For networks with a large number of computers, relatively scattered locations and a large amount of information transmission, it is recommended not to use bus structure.
2. Star structure
There is a unique forwarding node (central node) in the star structure network, and each computer is connected with the central node through a separate communication line. The information transmission mode and access protocol are very simple. RJ-45 interface network card is often used in small LAN workstations and servers with star structure, with hub as the central node and twisted pair connecting hub with workstations and servers. At present, the scheme with high cost performance is to adopt 10M/ 100M adaptive hub, the workstation uses low-grade 10M network card, and the server uses small star network purchased from high-grade 100M network card. The communication rate between the server and the hub is100m; The communication rate between the workstation and the hub is 10M.
Advantages: using the central node can provide services and reconfigure the network conveniently; The failure of a single connection point only affects one device, not the whole network, so it is easy to detect and isolate the failure and facilitate maintenance.
Disadvantages: each site is directly connected to the central node, which requires a large number of cables, so the cost is high; If the central node fails, the whole network will not work, which requires the central node to have high reliability.
3. Office building network wiring
The network wiring of the office building can consider one hub on each floor (connecting all computers on this floor), and connecting all hubs on each floor to the central server with cables to form a segmented management small network. If there are many hubs in the network, consider using switches. In fact, a switch is a special kind of hub, which is mainly used to connect hubs, servers or distributed backbone networks. It is recommended to use switches between buildings. To build a small network managed by segments, the cost-effective scheme is to install multiple network cards (2-4 cards) on the server, and each hub is connected with several branches and stars, which can speed up the network.
The horizontal trunk line of the office building can use unshielded class 5 and super class 5 twisted pairs (there are even class 6 and class 7 twisted pairs at present), so that the speed can reach 100Mbps or more; Coaxial cables (or optical cables) are used for outdoor and vertical wiring.
In network construction, wiring can be said to be a real one-time investment, and a large range of network wiring needs to break ground, so it is basically impossible to increase investment to improve its performance. The performance of the cabling system largely determines the performance of the network, so the planning of the cabling system should be based on the highest possible performance at present. For large networks, we must resolutely use Category 5 or Super Category 5 lines.
When building a local area network, we should first consider the geographical scope of the network distribution, and the size of the scope determines the structure and wiring of the network. When the range is small, several workstations can be connected by a cable to form a small network. When the geographical distribution of the network is wide, it is necessary to consider whether to manage it in sections, and configure the transmission media and connection equipment needed between network segments and between network segments accordingly.
Network hardware devices The main hardware devices that make up a small LAN are network transmission media such as network cards and hubs, and network interconnection devices such as repeaters, bridges, routers and gateways. The following mainly introduces network transmission media such as network cards and hubs and LAN interconnection devices such as repeaters, bridges, routers and gateways.
A network interface card (NIC), also known as a network adapter, is a hardware device that connects a computer and a network. The network card is inserted into the expansion slot of the computer or server, and exchanges data with the network through the network cable (such as twisted pair, coaxial cable or optical fiber) to enjoy resources. When purchasing a network card, you should consider the following factors:
1. speed
The speed of the network card describes the speed at which the network card receives and sends data. The price of 10M network card is relatively low (tens of yuan per card), and it can meet the data transmission requirements of ordinary small LAN access. Users considering cost performance can choose 10M network card. 100M high-speed network card should be selected in signal transmission with wide bandwidth or switched LAN.
2. Bus type
Common network cards can be divided into ISA network cards and PCI network cards according to bus types. ISA network card transmits 16 bits of data, with a nominal speed of 10M. PCI network card transmits 32 bits of data, which is faster. At present, most PCI network cards on the market are 10M and 100M. It is recommended not to buy outdated ISA network cards unless your computer does not have a PCI slot.
Step 3 connect
Common network card interfaces are BNC interface and RJ-45 interface (similar to telephone interface), and there are also dual-port network cards with both interfaces. The choice of interface is related to the network wiring form. In a small * * * shared LAN, BNC interface cards are directly connected to other computers and servers through coaxial cables. RJ-45 port network card is connected to the hub through twisted pair, and then connected to other computers and servers through the hub.
4. Others
When selecting a network card, you should also check what operating system the driver of its program floppy disk supports; If the speed requirement is high, you can consider choosing a full-duplex network card; If a diskless workstation is installed, the supplier shall provide the boot ROM on the corresponding network operating system. At present, most network cards on the market are plug-and-play network cards with soft jumpers, and it is difficult to identify and install such low-grade network cards as Windows 98.
HUB is the connecting device between computer and server in LAN, and it is the star connection point of LAN. Each workstation is connected to the hub through twisted pair, and the hub centrally manages the workstations.
The simplest stand-alone hub has multiple user ports (8 ports or 16 ports), and each port is connected to a network site (workstation or server) through twisted pair. After data is sent from one network site to the hub, it will be relayed to all other ports in the hub for each user on the network. Independent hubs are usually the cheapest hubs and are most suitable for small independent workgroups, departments or offices. Independent hubs have BNC interfaces, which are usually used to connect network servers.
Hub selection mainly considers the number of ports (8 ports, 16 ports or 24 ports) from the network station capacity and the speed (10M, 100M) from the data flow. There are many models and types of hubs, which cannot be described in this article.
Network transmission media Network transmission media are entities that transmit data in the network and connect network sites, such as twisted pair, coaxial cable and optical fiber. Network information can also be transmitted by radio system, microwave wireless system and infrared technology. The following mainly discusses twisted pair and coaxial cable.
1. Twisted pair cable
Twisted pair cable (hereinafter referred to as twisted pair) is a transmission medium formed by encapsulating one or more twisted pairs in an insulating sheath, and it is the most commonly used wiring material in local area network at present. In order to reduce the degree of signal interference, each twisted pair in a cable is generally formed by twisting two insulated copper wires, hence the name twisted pair. Twisted pair is generally used for wiring connection of star networks. RJ-45 connectors (crystal connectors) are installed at both ends to connect the network card and the hub. The maximum network cable length is100m. If you want to expand the network, you can install repeaters between two twisted-pair cables, and you can install up to four repeaters. For example, if four repeaters are installed and connected to five network segments, the maximum transmission range can reach 500 meters.
Twisted pair can be divided into two categories: unshielded twisted pair (UTP) and shielded twisted pair (STP). Unshielded twisted pair in LAN can be divided into three categories, four categories, five categories and super five categories, and shielded twisted pair can be divided into three categories and five categories.
At present, twisted-pair cables commonly used in local area networks are generally 5 types and 4 pairs (i.e. 8 wires) unshielded cables. The transmission rate of these twisted pairs can reach 100Mbps.
The outer protective rubber of the three types of twisted pair sold in the market is thin, with the words "CAT 3" marked on the rubber and "Category 3" marked on the outer packaging carton, so the price is low; The outer protective rubber of category 5 twisted pair is thick, with the words "CAT 5" marked on the rubber and "Category 5" marked on the outer packaging carton, so the price is higher. When buying, don't buy inferior 5-level twisted pair for the sake of cheapness. These products can only be used as triple twisted pair.
Super Category 5 twisted pair is unshielded twisted pair. Compared with ordinary Category 5 twisted pair, Super Category 5 twisted pair has less attenuation and stronger anti-interference ability when transmitting signals. In the 100M network, the interference degree of the user equipment is only 1/4 of that of the common Category 5 line. It is a solution for future network applications and is rarely used at present.
2. Coaxial cable
Coaxial cable consists of a hollow outer cylindrical conductor and an inner conductor located on the central axis. The inner conductor and the cylindrical conductor are separated from the outside by an insulating material. According to different transmission frequency bands, coaxial cables can be divided into baseband coaxial cables and broadband coaxial cables. Coaxial cable can be divided into thick cable and thin cable according to different diameters.
Thin cable has developed rapidly in recent years, so it is generally used in computer local area network without special requirements. Thin cable is generally used for bus network wiring connection. Connect BNC interface network card with T-type BNC interface connector, and install terminal resistors at both ends. The maximum length of each trunk line of the thin cable network is185m, and each trunk line can access up to 30 users. If you want to broaden the scope of the network, you need to use repeaters, such as connecting five network segments with four repeaters, so that the maximum distance of the network can reach 925 meters.
Thin cable is easy to install and low in cost, but its daily maintenance is not very convenient due to the limitation of network wiring structure. Once a user fails, it will affect the normal work of other users.
Thick cable is suitable for the network trunk of large LAN, with long wiring distance and good reliability. Users usually use external transceivers to connect to the network backbone. The length of each section in the thick cable LAN can reach 500 meters, and the maximum length can reach 2500 meters after connecting five network segments with four repeaters. If the network is directly connected to the network card through thick cable, the network card must have AUI interface (15-pin D interface). Although the LAN built with thick cable has high performance and long transmission distance, it is difficult to install and maintain the network and the cost is high.
3. Optical cable
It is a thin and flexible transmission medium, which consists of a group of optical fibers and is used to spread light beams. Compared with other transmission media, optical cable has good electromagnetic insulation performance, small signal attenuation, wider frequency band and longer transmission distance. Optical cables are mainly used to connect backbone networks when the transmission distance is long and the wiring conditions are special.
In optical cable communication, an optical transmitter generates a light beam, converts an electrical signal into an optical signal, and then introduces the optical signal into an optical fiber. At the other end of the optical cable, the optical receiver receives the optical signal transmitted on the optical fiber, converts it into an electrical signal, and decodes it before processing. Optical fiber cable is the best transmission medium in LAN because of its long transmission distance and fast transmission speed. The installation and connection of optical cables need professional technicians to complete.
LAN interconnection devices commonly used LAN interconnections include repeaters, bridges, routers and gateways.
1. Repeater
Repeaters are used to extend the same type of LAN, connect two networks at the physical layer, and transmit information between networks. Repeaters transmit information between networks for signal amplification, shaping and transmission. When the physical distance of the local area network exceeds the allowable range, repeaters can be used to expand the range of the local area network. In many networks, the number of repeaters added between workstations is limited, such as using up to four repeaters in Ethernet.
2. Bridge
The data layer of the bridge connects two LAN segments, the inter-network communication is transmitted by the bridge, and the intra-network communication is isolated by the bridge. When the network load is too heavy to cause performance degradation, dividing others into two network segments with a bridge can alleviate the busy degree of network communication to the maximum extent and improve communication efficiency. For example, a network distributed in two layers is divided into a network segment at each layer and connected by a bridge. At the same time, the bridge plays an isolation role, and the fault on one network segment will not affect the other network segment, thus improving the reliability of the network.
Step 3: Router
Routers are used to connect networks with different protocols in network layer, data layer and physical layer, and protocol conversion is completed by routers, thus eliminating the differences between network layer protocols. Routers are suitable for connecting complex large networks. Routers have strong interconnection ability, can execute complex routing algorithms, and process more information than bridges, but the processing speed is slower than bridges.
Step 4: doors
You can save a lot of money by choosing the latest and best network hardware equipment that suits your current needs without thinking ahead.
Gateway is used to connect subnets of different protocols above the network layer to form heterogeneous Internet. Gateway can realize communication between heterogeneous devices and translate and convert different transport layer, session layer, presentation layer and application layer protocols. Gateway has the function of converting incompatible high-level protocols, such as interconnecting PC workstations in NetWare with SUN network.
Factors to be considered when purchasing network equipment The purpose to be paid attention to when purchasing network hardware equipment is to determine the main factors of network equipment and network wiring. Because network wiring is not easy to upgrade, you should choose the latest and best materials. However, you don't have to choose the latest and best network hardware devices. First, although the equipment using the latest technology has good performance, it is still unable to play its high performance; Second, from the perspective of cost performance, ordinary equipment with low cost and stable performance may not be much worse than the so-called new technology products in overall performance.
Use a new network card with good quality and fast speed.
In a local area network, it is normal that computers can't communicate with each other, and there may be many reasons for the failure. According to the author's statistics, most of the faults in the LAN are related to the network card, or the network card is not installed correctly, or the network cable is in poor contact. It is also possible that the network card is old and the computer can't recognize it correctly. In addition, some network cards were installed in the server, which could not withstand the impact of large-capacity data and was eventually scrapped. Therefore, in order to avoid the above phenomenon, we must be willing to invest. If the server is equipped with a network card, you must use a good quality network card, because the server generally runs continuously, and only a good quality network card can "work" for a long time. In addition, due to the large data transmission capacity of the server, the capacity of the network card we bought must also match it, so as to achieve "a good horse with a good saddle."
For example, at present, the price difference between the low-grade 10M network card and the high-grade 100M network card of PCI is at least 10 times. As far as workstations in a small LAN are concerned, there is little difference in speed between 100M network card and high-end 100m network card. At present, you don't need the speed of 100M network card, but when you need high-speed transmission,