What Is a PCI Network Adapter?

The communication adapter is also called a network adapter or a network interface card (NIC), but now more people are willing to use the simpler name "network card". It is mainly used to connect shared resources and is a necessary part of a computer system.

Chinese name: Communication adapter
Foreign name: NIC, network interface controller

Alias: Network interface card
Function: Connect shared resources

Communication adapter physical connection

Insert a network interface board in the main box (or a PCMCIA card in a laptop) to connect the local computer to an external network.

Interface board

A network card is a network component that works at the data link layer. It is an interface between a computer and a transmission medium in a local area network. It can not only achieve physical connection and electrical signal matching with the transmission medium of the local area network, but also involves the sending and receiving of frames, and the frame. Packaging and unpacking, media access control, data encoding and decoding, and data cache functions.

Communication adapter network card function

The network card is equipped with a processor and memory (including RAM and ROM). The communication between the network card and the local area network is performed by serial transmission through a cable or twisted pair, and the communication between the network card and the computer is performed in parallel through the I / O bus on the computer motherboard. Therefore, an important function of the network card is to perform serial / parallel conversion. Because the data rate on the network and the data rate on the computer bus are not the same, a memory chip that caches data must be installed in the network card.

When installing the network card, the device driver that manages the network card must be installed in the computer's operating system. This driver will later tell the network card where to store the data blocks transferred from the local area network. The network card must also be able to implement the Ethernet protocol.

The network card is not an independent autonomous unit. Because the network card itself does not have a power source, it must use the power of the inserted computer and be controlled by the computer. Therefore, the network card can be regarded as a semi-autonomous unit. When the network card receives an error frame, it discards the frame without notifying the computer it is plugged into. When the network card receives a correct frame, it uses an interrupt to notify the computer and deliver it to the network layer in the protocol stack. When a computer wants to send an IP datagram, it is delivered from the protocol stack to the network card to be assembled into a frame and sent to the local area network.

With the continuous improvement of the integration, the number of chips on the network card has been continuously reduced. Although there are many types of network cards produced by various manufacturers, their functions are similar. The main functions of the network card are the following three:

1. Encapsulation and decapsulation of data: When sending, the data delivered by the previous layer is added to the header and tail to become the Ethernet frame. When receiving, the Ethernet frame is stripped of the head and tail, and then sent to the upper layer;

2. Link management: mainly the implementation of CSMA / CD protocol;

3. Encoding and decoding: Manchester encoding and decoding.

Communication adapter network card category

As far as compatible network cards are concerned, network cards are generally divided into ordinary workstation network cards and server-specific network cards. The server-specific network card is designed to adapt to many types of network services, and the performance is also different. It can be classified according to the following standards: According to the bandwidth supported by the network card, it can be divided into 10M network card, 100M network card, 10 / 100M adaptive network card, 1000M network card According to the type of network card bus, it is mainly divided into three categories: ISA network card, EISA network card, and PCI network card. Among them, ISA network card and PCI network card are more commonly used. The bandwidth of the ISA bus network card is generally 10M, and the bandwidth of the PCI bus network card is from 10M to 1000M. It is also a 10M network card. Because the ISA bus is 16-bit and the PCI bus is 32-bit, the PCI network card is faster than the ISA network card.

Interface type of network card: According to different transmission media, there are three interface types: AUI interface (thick cable interface), BNC interface (thin cable interface) and RJ-45 interface (twisted pair interface). So when choosing a network card, you should pay attention to the interface types supported by the network card, otherwise it may not be applicable to your network. Common 10M network cards on the market are mainly single-port network cards (RJ-45 interface or BNC interface) and dual-port network cards (RJ-45 and BNC interface). There are fewer network cards with AUI thick cable interface. The 100M and 1000M network cards are generally single-port cards (RJ-45 interface). In addition to the interface of the network card, we often pay attention to whether the network card supports diskless startup when selecting a network card. If necessary, consider whether the network card supports fiber optic connections.

Factors to consider when purchasing network adapters

Whether the network card can be correctly selected, connected, and set during networking is often a prerequisite and necessary condition for whether the network can be correctly connected. In general, the following factors should be considered when purchasing a network card:

Communication adapter network type

Optional types include Ethernet, token ring network, FDDI network, etc. When selecting, you should choose the corresponding network card according to the type of network. According to statistics, most LANs currently use Ethernet technology.

Communication adapter applications

Currently, there are 10M, 100M, 10M / 100M, and Gigabit Ethernet network card. For a large data network, the server should use a Gigabit Ethernet network card. This network card is mostly used for the connection between the server and the switch to improve the overall system response rate. The 10M, 100M, and 10M / 100M network cards are network devices that people often buy and commonly use. The prices of these three products are not much different. 10M / 100M adaptive means that the network card can automatically negotiate with the remote network device (hub or switch) to determine whether the current available rate is 10M or 100M. For ordinary file sharing and other applications, 10M network card is enough, but for possible future applications such as voice and video, 100M network card will be more conducive to the transmission of real-time applications. In view of the foundation that 10M technology already has (such as previous hubs and switches), the usual workaround is to purchase a 10M / 100M network card, which is not only conducive to protecting existing investments but also to further network expansion. For SMEs, 10M / 100M / 1000M network cards should be the first choice when purchasing.

Communication adapter transmission rate

The transmission rate of the network card should be selected according to the bandwidth requirements of the server or workstation and combined with the maximum transmission rate that the physical transmission medium can provide. Taking Ethernet as an example, the selectable speeds are 10Mbps, 10 / 100Mbps, 1000Mbps, and even 10Gbps. However, the higher the speed, the more suitable. For example, it is a waste to configure a 1000M network card for a computer connected to a twisted pair with only 100M transmission speed, because it can only achieve a transmission rate of 100M at most.

Communication adapter bus type

Common bus slot types in computers include ISA, EISA, VESA, PCI, and PCMCIA. The intelligent network card of PCI or EISA bus is usually used on the server, the ordinary network card of PCI or ISA bus is used for the workstation, and the PCMCIA bus network card or the portable network card with parallel interface is used in the laptop. At present, PCs basically no longer support ISA connection, so when buying a network card for your PC, do not buy an outdated ISA network card, you should choose a PCI network card.

Cable adapters supported by communication adapter network cards

The network card is ultimately connected to the network, so there must be an interface through which the network cable is connected to other computer network equipment. Different network interfaces are suitable for different network types. At present, the common interfaces are RJ-45 interface of Ethernet, BNC interface of thin coaxial cable, AUI interface of thick coaxial cable, FDDI interface, ATM interface, etc. In addition, some network cards provide two or more types of interfaces in order to be applicable to a wider range of application environments. Some network cards also provide RJ-45, BNC interface, or AUI interface.

1. RJ-45 interface: This is the most common type of network card and the most widely used interface type network card. This is mainly due to the popularity of twisted-pair Ethernet applications. Because this RJ-45 interface type network card is used in Ethernet with twisted pair as the transmission medium, its interface is similar to the common telephone interface RJ-11, but RJ-45 is an 8-core cable, and the telephone line The interface is 4-core, usually only 2-core cable (ISDN telephone line to 4-core cable). There are also two status indicator lights on the network card. The working status of the network card can be judged initially through the color of these two indicators.

2. BNC interface: This interface network card is used in Ethernet or token network with thin coaxial cable as the transmission medium. At present, this type of interface network card is rare, mainly because it uses thin coaxial cable as the transmission medium. Fewer networks.

(C) AUI interface: This type of interface card is used in Ethernet or token network with thick coaxial cable as the transmission medium. This type of interface card is currently rare.

3. FDDI interface: The network card of this interface is suitable for FDDI (Optical Fiber Distributed Data Interface) network. This network has a bandwidth of 100Mbps, but the transmission medium it uses is optical fiber, so the interface of this FDDI interface network card is also Fiber optic interface. With the advent of Fast Ethernet, its speed advantage has ceased to exist, but its disadvantages of using expensive optical fiber as the transmission medium have not changed, so it is currently very rare.

4. ATM interface: This type of interface card is used in ATM (Asynchronous Transfer Mode) fiber (or twisted pair) networks. It can provide physical transmission speeds up to 155Mbps

Communication adapter price and brand

The price of network cards with different speeds and brands varies widely.

How a communication adapter determines a network failure

Ping Communication adapter Ping

This command is a very useful tool for testing network connection status and packet sending and receiving status. It is the most commonly used command for network testing. Ping sends a return request data packet to the target host (address), asking the target host to reply after receiving the request, so as to determine the network response time and whether the machine is connected to the target host (address).

If the ping is unsuccessful, the failure can be predicted to occur in the following areas: network cable failure, incorrect network adapter configuration, and incorrect IP address. If the ping is successful and the network is still unavailable, the problem is likely to be in the software configuration of the network system. A successful ping can only ensure that there is a physical path for communication between the local machine and the target host.

Command format: ping IP address or host name [-t] [-a] [-n count] [-l size]

Parameter meaning:

-t keep sending data to the target host;

-a displays the network address of the target host in IP address format;

-n count specifies how many times to ping, the specific number is specified by count;

-l size Specify the size of the packet sent to the destination host.

For example, when your machine cannot access the Internet, you first want to confirm whether it is a local area network failure. Assume that the IP address of the proxy server of the local area network is 202.168.0.1. You can use the Ping avoid 202.168.0.1 command to check whether the machine is connected to the proxy server. As another example, the common command to test whether the local network card is installed correctly is ping 127.0.0.1.

Tracert Communication adapter Tracert

This command is used to display the path taken by the data packet to reach the target host and the time to reach each node. The command function is similar to Ping, but the information it obtains is much more detailed than the Ping command. It shows all the paths the data packets take, the IP of the node, and the time it takes. This command is more suitable for large networks.

Command format: tracert IP address or host name [-d] [-h maximumhops] [-j host_list] [-w timeout]

Parameter meaning:

-d does not resolve the name of the target host;

-h maximum_hops specifies the maximum number of hops to the target address;

-j host_list release the source route according to the address in the host list;

-w timeout Specifies the timeout interval. The default time unit of the program is milliseconds.

If we add some parameters to the Tracert command, we can also detect other more detailed information. For example, using the parameter -d, you can specify that the program also resolves the domain name of the target host while tracking the path information of the host.

The Netstat command can help network administrators understand the overall usage of the network. It can display detailed information about the currently active network connections, such as displaying network connections, routing tables, and network interface information, and can count which total network connections are currently running.

Using command parameters, the command can display the usage status of all protocols, including TCP, UDP, and IP protocols. In addition, you can select specific protocols and view their specific information, and display the port numbers of all hosts and the current host. Detailed routing information.

Command format: netstat [-r] [-s] [-n] [-a]

Parameter meaning:

-r display the contents of the local routing table;

-s display the usage status of each protocol (including TCP protocol, UDP protocol, IP protocol);

-n Display addresses and ports in a numeric table;

-a Displays the port numbers of all hosts.

The Winipcfg command displays the specific configuration information of the IP protocol in the form of a window. The command can display the physical address of the network adapter, the IP address of the host, the subnet mask, and the default gateway. You can also view the host name, DNS server, and node type. information. The physical address of the network adapter is very useful when detecting network errors.

Command format: winipcfg [/?] [/ All]

Parameter meaning:

/ all displays all configuration information about IP addresses;

/ batch [file] Write the command results to the specified file;

/ renew_ all Retry all network adapters;

/ release_all releases all network adapters;

/ renew N reset network adapter N;

/ release N Releases network adapter N.

LED Communication adapter LED indicator

Generally speaking, each network card has more than one LED (Light Emitting Diode) indicator, which is used to indicate the different working status of the network card, so that we can check whether the network card is working properly. Typical LED indicators are Link / Act, Full, Power, etc. Link / Act indicates the active status of the connection, Full indicates whether it is Full Duplex, and Power is the power indicator.

Communication adapter main chip

The main control chip of the network card is the core component of the network card. The performance of a network card mainly depends on the quality of the chip. The main control chip of the network card generally adopts a low-energy design of 3.3V and a chip process of 0.35 m, which enables it to quickly calculate the data flowing through the network card, thereby reducing the burden on the CPU. The following are the commonly used network card control chips.

1. Realtek 8201BL: It is a common motherboard integrated network chip (also known as PHY network chip). PHY chip refers to handing over the computing part of the network control chip to the processor or southbridge chip to simplify the circuit design and thus reduce the cost.

2. Realtek 8139C / D: It is one of the most used network cards. The 8139D mainly adds power management functions, and the others are basically the same as the 8139C chip. The chip supports 10M / 100Mbps.

3. lntel Pro / 100VE: lntel's entry-level network chip.

4. nForce MCP NVIDIA / 3Com: nForce2 has two built-in network chip functions, Realtek 8210BL PHY network chip and Broabcom AC101L PHY network chip.

5. 3Com 905C: C supports 10 / 100Mbps speed.

6. SiS900: Originally a single network control chip, but now it has been integrated into the south bridge chip. Support 100Mbps.

Communication adapter remote wake-up function

Wake-on-LAN (WOL) is a technology in which a network card cooperates with other software and hardware to remotely boot through a local area network. No matter how far away the accessed computer is from us and where it is, as long as it is on the same local area network Within, they can be activated at any time. This technology is very suitable for environments with remote network management requirements. If there is such a requirement, you should pay attention to whether you have this function when purchasing a network card.

Computers that can be remotely awakened have certain requirements on hardware, which are mainly manifested in network cards, motherboards, and power supplies.

1. Network card: Whether to realize remote wake-up, one of the main components is a network card that supports WOL. The network card of the remotely awakened computer must support WOL, and the network card used to wake up other computers does not have to support WOL. In addition, when multiple network cards are installed in a computer, only one of them is set to be capable of remote wake-up.

2. Motherboard: It is also necessary to support remote wake-up. You can confirm by checking if there is a "Wake on LAN" item in the "Power Management Setup" menu of the CMOS. In addition, motherboards that support remote wake-up usually have a dedicated 3-pin socket to provide power to the network card (PCI2.1 standard). Because the current motherboards usually support the PCI 2.2 standard, it can directly provide + 3.3V Standby power to the network card through the PCI slot. Even if the WOL power cable is not connected, remote wake-up can be achieved. Therefore, 3-core sockets may no longer be provided. Whether the motherboard supports the PCI2.2 standard can be confirmed by checking whether there is a "Wake on PCI Card" item in the "Power Management Setup" menu of the CMOS.

3. Power supply: To achieve remote wake-up, the computer must be installed with an ATX power supply that complies with the ATX 2.01 standard, and the + 5V Standby current should be at least 600mA or more.

Communication adapter wireless network card

The working principle of the wireless network card is microwave radio frequency technology. The notebook currently has several wireless data transmission modes such as WIFI, GPRS, and CDMA to access the Internet. The latter two are implemented by China Telecom and China Unicom. The former Telecom or Netcom participates, but does not Mostly, they have a WIFI base station (in fact, a WIFI router, etc.) that accesses the Internet, and a WIFI network card for laptops. To say that the basic concepts are similar, data transmission is performed by wireless. Wireless Internet access follows the 802.1q standard. Through wireless transmission, a wireless access point sends out a signal and uses a wireless network card to receive and send data.

According to the IEEE802.11 protocol, the wireless LAN card is divided into a media access control (MAC) layer and a physical layer (PHY Layer) between the two, and a media access control-physical (MAC-PHY) sublayer (Sublayers) is also defined. . The MAC layer provides an interface between the host and the physical layer, and manages external memory, which corresponds to the NIC unit of the wireless network card hardware.

The physical layer specifically implements the reception and transmission of radio signals, which corresponds to the spread-spectrum communicator in the wireless network card hardware. The physical layer provides free channel estimation CCA information to the MAC layer in order to determine whether it can send signals. The MAC layer controls the CCSMA / CA protocol of the wireless network. The MAC-PHY sublayer mainly implements data packaging and unpacking. The necessary control information is placed in front of the data packet.

The IEEE 802.11 protocol states that the physical layer must have at least one method to provide a free channel estimation CCA signal. The working principle of the wireless network card is as follows: When the physical layer receives the signal and confirms that it is correct, it submits it to the MAC-PHY sublayer. After unpacking, the data is turned over to the MAC layer, and then it is determined whether it is data sent to the network card. Turn it in, otherwise, discard it.

If the signal received by the physical layer and sent to the network card is wrong, the sender needs to be notified to resend the packet information. When the network card needs to send data, it must first determine whether the channel is idle. If it is empty, it will be sent after a random backoff period, otherwise, it will not be sent for the time being. Because the network card works for time division duplexing, it cannot receive when sending, and it cannot send when receiving.

Development history of communication adapter network card

In the 1980s, with the development of microcomputer technology, the technology and products of microcomputer residential area network have developed rapidly. In the late 1980s, the foreign computer industry had predicted that the environment used by the computer in the 1990s was the network. In fact, this is indeed the case. The development of microcomputer residential area networks has a considerable impact in the entire computer network field. Thousands of microcomputer network users are distributed in various application fields to promote the development of network application technology, thereby also accelerating the computer network Development of technology.

In the past, foreign personal computer home area network products have occupied the network market. Among them, the number of network building users is mainly products from companies such as NOVELL, 3COM, IBM, BANYAN, and SUN. With the development of the network, Taiwanese manufacturers have developed rapidly due to their strong production capacity and more factories in the Mainland. Brands such as D-LINK and TP-LINK have gradually matured. In addition, domestic computer product manufacturers have faithfully Tak and Lenovo have also produced their own network products.

Network card: The most important connection device in a computer's local area network. The computer is mainly connected to the network through a network card. In the network, the work of the network card is twofold: on the one hand, it is responsible for receiving the data packets uploaded from the network, unpacking them, and transmitting the data to the local computer through the bus on the motherboard; Into the network.

Computer network: It is the product of the development of computer technology and communication technology. It has developed along with the society's requirements for information sharing and information transmission. The so-called computer network is to use communication equipment and lines to interconnect multiple computer systems with different geographical locations and independent functions, and to implement the network with well-functioning network software (that is, network communication protocols, information exchange methods, and network operating systems). A system for resource sharing and information transfer.

Computer network consists of three parts: resource subnet, communication subnet, and communication protocol.

Resource subnet: It is the user-oriented part of the computer network. It is responsible for the application-oriented data processing of the entire network. Its main body is all the main computers connected to the computer network, as well as the external equipment, software and Data available for sharing, etc.

Communication subnet: The part of the four computer network responsible for data communication. The communication transmission medium can be twisted pair, coaxial cable, radio communication, microwave, optical fiber, etc.

Communication protocol: In order to make the communication between the computers in the network reliable and effective, the rules and conventions that both parties of the communication must abide by are called communication protocols.

Resource sharing: including hardware and software resources. Hardware resources such as high-performance processing components with special functions, high-performance input and output devices (laser printers, plotters, etc.) and large-capacity auxiliary storage devices (such as tape drives, large-capacity hard drives, etc.), their sharing can save Hardware overhead. Software resources such as software and data.

Local area network: It is a communication system that allows several independent computers to communicate directly at an appropriate transmission rate within an appropriate range. Generally, networks can be classified according to their size. Most of the networks we use in offices or homes belong to local area networks. This type of network feels fast because the distance between computers is short and there is no need to go through too many network devices. However, the scope of application is therefore small.

Wide area network (WAN, Wide Area Network): In contrast to the local area network, anything that exceeds the local area network can be counted as a wide area network.

Metropolitan area network (MAN, Metropolitan ARea Network): A network operating within a city, or a network that physically uses urban telecommunication facilities (such as underground cable systems), is sometimes distinguished from the WAN and is called a metropolitan area network .

Network architecture: The overall design of the communication system, which provides standards for network hardware, software, protocols, access control, and topology. It widely uses the reference model of the Open System Interconnection (OSI-Open System Interconnection) proposed by the International Organization for Standardization (ISO) in 1979. The OSI reference model describes the network structure in seven layers: physical layer, data link layer, network layer, transport layer, dialogue layer, presentation layer, and application layer. Its specifications are open to all manufacturers and have knowledge of the international network structure. And the role of open systems. It directly affects the performance of buses, interfaces, and networks. Currently common network architectures include FDDI, Ethernet, Token Ring, and Fast Ethernet. From a network interconnection perspective, the key elements of a network architecture are protocols and topology.

Protocol: A formal description of the data format and rules that must be followed when exchanging data between computers. To put it simply, if computers on the network can communicate with each other smoothly, they must speak the same language, which is equivalent to a protocol. It is divided into Ethernet, NetBEUI, IPX / SPX, and TCP / IP protocols.

Topology structure: refers to the form in which various sites in the network are connected to each other. There are bus topology, star topology, ring topology, and their hybrids.

FDDI / CDDI: Developed by American National Standards Institute ANSI X3T9.5. The speed is 100Mbps; CDDI is FDDI based on copper cable (twisted pair). FDDI technology is mature, the network can extend 100 kilometers, and because of the use of ring structure and excellent management capabilities, it has high reliability. Expensive, complicated installation, complete standards, mature technology, rich software and hardware products.

IEEE802.5 / Token Ring Network: It is commonly used in IBM systems, and its supported rates are 4Mbps and 16Mbps. Currently Novell and IBM LAN Server support 16Mbps IEEE802.5 / token ring network technology.

Switched Ethernet: The supported protocol is still IEEE802.3 / Ethernet, but provides multiple separate 10Mbps ports. It is fully compatible with the original IEEE802.3 / Ethernet and overcomes the network efficiency drop caused by sharing 10Mbps.

100BASE-T Fast Ethernet: The difference from 10BASE-T is that the speed of the network is increased ten times, that is, 100M. FDDI's PMD protocol is used, but the price is cheaper than FDDI. The 100BASE-T standard is formulated by IEEE802.3. It is easy to integrate with 10BASE-T by using the same media access technology, similar step rules and the same pinouts as 10BASE-T. Only two repeaters are allowed per network segment, and the maximum network span is 210 meters.

IEEE802.3 / Ethernet (Ethernet): The most widely used media access technology currently operates at the physical and data link layers of the OSI model. It is the main media access technology adopted by Novell, Widows NT, IBM, UNIX network LANServer, DECNET and other low-level layers. The networking method is flexible, convenient, and supports many software and hardware products. Its speed is shared 10Mbps. According to different media can be divided into: 10BASE-2 (coaxial thick cable), 10BASE-5 (coaxial thin cable), 10BASE-T (twisted pair) and 10BASE-FL (optical fiber).

NETBIOS / NETBEUI: NETBIOS is the industry standard for LAN software interfaces, which can support multiple transmission media. NETBEUI is an extended user interface for NETBIOS, used by Microaoft Windows NT and IBM's LAN Manager. NETBIOS was developed earlier and is relatively simple. It does not consider the interconnection between networks. Its naming scheme is not suitable for multiple operating systems.

IPX / SPX: The main protocol of the NOVELL network. At present, there are many software and hardware that support IPX / SPX, and I / O devices. In the OSI reference model, it is equivalent to the third and fourth layers (network layer and transport layer). In the NOVELL network, the IP protocol NETBIOS protocol can be loaded on IPX.

TCP / IP: IP is widely deployed in UNIX and has become the de facto international industry standard. IP is also the main protocol of the Internet. The IP protocol can span the LAN and WAN. Almost all LAN and WAN devices support the IP protocol, which is the best protocol for unified media transmission. The IP protocol is a data-type protocol, which has a better response time for transmission and less protocol interaction, which is more suitable for the needs of high-speed transmission.

Bus-type topology: A single transmission line is used as the transmission medium, and all stations are directly connected to the trunk cable, that is, the bus, through the corresponding hardware interface.

Star topology: All sites are connected to a central point, which is called the hub of the network.

Ring topology: All stations are connected to each other in series, like a chain, forming a loop or called a ring.

Hybrid topology: After interconnection between residential area networks, there will be a mixed form of several topologies, that is, a hybrid topology.

Transmission medium: It is the physical path between the sender and receiver in the communication network. Currently, the commonly used network transmission media include twisted pair, coaxial cable and optical cable.

Twisted pair: It is the most commonly used transmission medium in integrated wiring systems. Especially in the star network topology, twisted pair is an essential wiring material. Twisted-pair cables are packaged with one or more twisted pairs. In order to reduce the level of signal interference and to reduce the level of signal interference, each pair of twisted-pair cables is generally wound with two insulated copper wires. Twisted pair can be divided into two categories: unshielded twisted pair (UTP) and shielded twisted pair (STP). Among them, STP is divided into 3 types and 5 types, and UTP is divided into 3 types, 4 types, 5 types, and super 5 types. At the same time, 6 and 7 twisted pairs will also be in the near future. A wiring system for computer networks.

RJ-45 connector: The two ends of each twisted pair are connected to the network card and hub (or switch) by installing RJ-45 connectors (commonly known as crystal heads).

Coaxial cable: It consists of a hollow cylindrical mesh copper conductor and a copper conductor located on the central axis. The copper conductor, the hollow cylindrical conductor and the outside are separated by an insulating material. Compared with twisted pair cables, coaxial cables have strong anti-interference ability and good shielding performance, so they are often used for device-to-device connection or for bus-type network topologies. According to the diameter, it can be divided into two types: thin cable and thick cable.

BNC connector: BNC connectors are installed at both ends of the thin cable and connected to the network card and hub (or switch) through a dedicated T-type connector.

Optical fiber: Optical fiber is a light guide fiber. It is a small, flexible medium that can transmit optical signals. An optical cable consists of multiple optical fibers. Compared with twisted pair and coaxial cables, fiber optic cables meet the current network requirements for long-distance transmission of large-capacity information and play a very important role in computer networks.

Half-duplex: It means that although the network card can receive and send data, it can only do one action at a time, and cannot send and receive at the same time.

Full-duplex: It is able to receive and send signals "simultaneously". For example, a telephone is a full-duplex transmission device. While listening to the other party, we can also send a call to the other party. In theory, full-duplex transmission can improve network efficiency, but it is actually only useful in conjunction with other related equipment. For example, a twisted pair network cable must be selected for full-duplex transmission, and the hub (HUB) connected in the middle must also be capable of full-duplex transmission. Finally, the network operating system used must also support full-duplex In this way, the power of full-duplex transmission can be truly exerted.

Programmed I / O: This is an effective transmission method from its early use so far. At that time, the company s NE 2000 network card that was popular around the world adopted this method. This transmission method is not easy to improve the transmission efficiency. Once it encounters a large amount of data, it becomes the bottleneck of transmission.

Shared Memory: This type of network card puts the data to be transferred to the card's memory, and this memory must occupy an end address in advance (mostly occupying an address between 640-1024KB). With this address, this memory is It can be regarded as a part of the memory of the motherboard: when the host requests data from the network card, it goes directly to this memory to retrieve it; conversely, putting the data into the memory is also equivalent to passing it to the network card. If the PROGRAMMED I / O method is likened to spooning water with a spoon, then SHARED MEMORY fetches water with a bucket, which can highlight its efficiency when there is a large amount of transmission.

Bus Master: This type of network card has a control chip (CONTROLLER), which is used to control the entire transmission process and the use of the bus. Because the control action is performed by this chip, data can be directly transmitted from the network card to the motherboard without I / O PROT does not have to go through the CPU. Because it does not take up valuable CPU time, it can effectively reduce the burden on the system, so it is especially suitable for servers. Most EISA, MCA, and PCI network cards support communication with the motherboard using this BUS MASTER method.

802.3x flow control: Improved performance due to more efficient data transmission. The network card communicates with the switch to establish optimal data transmission.

Parallel Tasking technology: 3COM's patented technology, this technology can maximize data transmission speed when 10Mbps or 100 Mbps connection.

Parallel Tasking II technology: 3COM's patented technology, which can reduce CPU usage and improve application performance because data is more efficiently transmitted on the PCI bus. In the past, the network card allowed only 64 bytes of data to be transmitted on the PCI bus at a time during a bus master operation cycle. In order to transfer a 1514-byte data packet to the PC host, 24 separate bus master operation cycles are required, which makes the bus inefficient. With Parallel Tasking II technology, the network card can transmit the entire Ethernet data packet on the bus in one bus master operation cycle, which greatly improves the efficiency of the PCI bus. The result is faster transfers and improved system performance, making desktop and server application software work better.

32-bit bus master DMA: Wide data path and high-speed transmission and low CPU utilization provide the best system performance.

Interactive access technology: The network card can dynamically analyze network information flow to adjust network performance.

Remote wake-up: Enables network managers to order remote PCs to be powered on at a central location, making it easy to update and maintain desktops during off-hours (PC motherboards must be equipped with a 3-pin remote wake-up connector; it also requires Desktop Management Application software, which can Generate Magic Packet TM remote wake-up signal).

DMI2.0: Enables remote PCs to record and report PC status to improve desktop management.

3Com DynamicAccess software: It is an integral part of the 3Com Fast EtherLink XL series, adding various intelligences to the network card. It includes 1. Prioritize data flow by service category. Assign high priority to time-critical data to improve the performance of multimedia and critical business applications; 2. Distributed RMON (dRMON) SmartAgent TM software. The software can provide comprehensive and inexpensive network management in switched and high-speed network environments, including support for all types of remote monitoring; 3. Fast IP software. The software minimizes all kinds of bottlenecks that the router may produce, thereby improving the interconnection performance between networks; 4. Effective multicast control. This control can automatically filter out unnecessary multicast streams before the multicast data stream floods the LAN, thereby expanding the useful bandwidth of the network.

100VG-ANYLAN: Developed by HP, AT & T, and developed by IEEE802.12. Its advantages are that it can be based on the current three types of 8-core twisted pair networks, and supports priority scheduling, which is suitable for transmitting multimedia information and is cheap. The disadvantage is that the standard is immature, lacks the backbone of fault tolerance, has limited confidentiality, and supports fewer products.

ATM: High-speed packet-based network, which will be the main communication transmission method of the information highway in the future. ATM standards are developed by the ATM Forum (more than 150 countries participate). Data exchange based on 53-byte cells, the rate can reach 25M, 34M, 45M, 50M, 155M, 622M, and can reach several Gbps. ATM supports more and more products, but the price is higher.