Computer Network Technology and Application

Chapter 10 Basic knowledge of local area network (2)
(1) FDDI
FDDI uses a double-ring structure, and the data on the two rings is transmitted in opposite directions, which can support 100 computers at a distance of more than 500km.The double ring is composed of a main ring and a backup ring. Under normal circumstances, the main ring is used for data transmission, and the backup ring is idle, thus improving the reliability of FDDI.The network has the characteristics of a timing token (only the station that gets the token can send data, otherwise it can only receive) protocol, supports multiple topologies, and the transmission medium is optical fiber.

(2) CCDI
CCDI is a variant of FDDI, which uses twisted-pair copper cables as the transmission medium, and the data transmission rate is usually 100Mbps.

(3) FDDI-2
FDDI‐2 is an extended protocol of FDDI, supporting voice, video and data transmission.

(4) FDDT
FDDT is another variant protocol of FDDI, known as full-duplex technology, using the same network structure as FDDI, but the transmission rate can reach 200Mbps.

3. Asynchronous Transfer Mode (ATM)

Integrated Services Digital Network (ISDN) is a multimedia communication network integrating voice, image and data.In the early 20s, the United States and Japan have started the research and use of ISDN. B‐ISDN is a broadband communication network that requires a new data transmission mode, so Asynchronous Transfer Mode (ATM) was born.

In 1990, the International Telegraph and Telephone Consultative Committee (CCITT) formally recommended ATM as a technical basis for realizing B-ISDN.In this way, the information transmission and exchange mode based on ATM has become the basis of telecommunications and computer network operations and one of the main bodies of communication in the 2st century.

ATM adopts cell-based asynchronous transmission mode and virtual circuit structure, fundamentally solves the real-time and bandwidth problems of multimedia: realizes point-to-point transmission for virtual links, and usually provides 155Mbps bandwidth.It not only absorbs the connection-oriented service and quality of service guarantee of circuit switching in traffic communication, but also maintains the flexibility of variable bandwidth and suitable for burst transmission in traditional networks such as Ethernet and FDDI, so it has become the scope of application so far. The most extensive, most technologically advanced, and most ideal means of network interconnection.

ATM technology has the following characteristics:
(1) Realized network transmission connection service and service quality assurance (QoS).

(2) Large switching throughput and high bandwidth utilization.

(3) It has a flexible networking topology and load balancing capabilities, with high scalability and reliability.

(4) ATM technology is the only network technology that can be applied to both LAN and WAN at the same time. It is the unification of LAN and WAN technologies and is widely used in the construction of MANs.

4. Other types of LAN
Token Ring is a network technology successfully developed by IBM in the early 20s.It is called a ring because the physical structure of this network has the shape of a ring.There are multiple stations on the ring connected to the ring one by one, and the adjacent stations are point-to-point links. Therefore, the Token Ring is different from the Ethernet in the broadcast mode. It is a LAN that broadcasts to the next station sequentially.Another feature that makes it different from Ethernet is that it is heavily loaded and still has a deterministic response time.

The standard followed by Token Ring is IEEE802.5, which specifies three operating rates: 1Mbps, 4Mbps and 16Mbps.

3.2 Reference model of local area network

Any network needs an open communication standard, and the local area network mainly adopts the IEEE802 standard.

3.2.1 Reference model of local area network

The architecture of a LAN generally only includes the lowest two layers of the OSI reference model: the physical layer and the data link layer.

1. Physical layer

The main function of the physical layer is to ensure the correct transmission of binary bit signals on the communication channel. Its main functions include signal encoding and decoding, generation and removal of synchronous preambles, sending and receiving of binary bit signals, and error checking (CRC checking). test) and other functions.

2. Data link layer

The main role of the data link layer is media access control.Since different LANs use different data formats, different transmission media, and different network topologies, which lead to different media access control methods, it is impossible to define a unified medium that is independent of the medium at the data link layer. Access control method.In order to simplify the complexity of protocol design, the LAN reference model divides the data link layer into two independent sublayers: Media Access Control sublayer (Media AccessControl, MAC) and Logical Link Control sublayer (Logical Link Control, LLC) .

Addressing is one of the tasks of the Mac sublayer in a local network, finding stations (computers or switches etc.) using Mac addresses.

The composition of the MAC address of different manufacturers is also different. The addressing of the MAC address can determine which port or network card of which device is on the local network. For example, after the data arrives at the destination network (local network), the destination computer is determined by the MAC address.

Common manufacturers' MAC addresses are as follows.

Cisco: 00‐00‐0c
Novell：00‐00‐1B／ 00‐00‐D8
3Com：00‐20‐AF／00‐60‐8C
IBM: 08‐00‐5A
典型的Ethernet地址：00‐60‐8C‐01‐28‐12
Tip: You can type the GETMAC command in the DOS environment to get the MAC address of the machine.

The logical link control sublayer (LLC) constitutes the upper half of the data link layer, adjacent to the network layer and the MAC sublayer, and provides services to the network layer with the support of the MAC sublayer.

3.2.2 IEEE802 standard
In February 1980, IEEE established the LAN Standards Committee (referred to as the IEEE2 Committee), specializing in the standardization of LANs, and formulated the IEEE802 standard.

The IEEE802 standard mainly includes:

(1) The IEEE802.1 standard defines the LAN architecture, network interconnection, network management and performance testing.

(2) The IEEE802.2 standard defines the functions and services of the logical link control (LLC) sublayer.

（3）IEEE802.3标准，定义了CSMA／CD总线介质访问控制子层和物理层规范。在物理层，还定义了四种不同介质的10Mbps以太网规范，包括10Base‐5（粗同轴电缆）、10Baes‐2（细同轴电缆）、10Base‐F（多模光纤）和10Base‐T（无屏蔽双绞线UTP）。

The IEEE802.3u standard, the 100M Fast Ethernet standard, has now been incorporated into IEEE802.3.

IEEE802.3z standard, fiber optic media Gigabit Ethernet standard specification.

IEEE802.3ab standard, the transmission distance of 100m Category 5 unshielded twisted pair Gigabit Ethernet standard specification.

IEEE802.3ae standard, [-] Gigabit Ethernet standard specification.

(4) The IEEE802.4 standard defines the token bus (Token Bus) media access control sublayer and physical layer specifications.

(5) The IEEE802.5 standard defines the token ring (Token Ring) media access control sublayer and physical layer specifications.

(6) The IEEE802.6 standard defines the metropolitan area network (MAN) media access control sublayer and physical layer specifications.

(7) IEEE802.7 standard, broadband network technology.

(8) IEEE802.8 standard, optical fiber transmission technology.

(9) IEEE802.9 standard, integrated voice and data local area network (IVDLAN) technology.

(10) The IEEE802.10 standard defines an interoperable local area network security specification (SILS).

(11) IEEE802.11 standard, which defines the wireless LAN medium access control method and physical layer specification, mainly including:
IEEE802.11a defines a wireless local area network standard operating in the 5GHz frequency band with a transmission rate of 54Mbps.

IEEE802.11b defines a wireless local area network standard operating in the 2.4GHz frequency band with a transmission rate of 11Mbps.

IEEE802.11g defines a wireless local area network standard operating in the 2.4GHz frequency band with a transmission rate of 54Mbps.

(12) The IEEE802.12 standard defines the 100VG‐AnyLAN fast LAN access method and physical layer specifications.

(13) IEEE802.14 standard, interactive TV network (Cable Modem) technology.

(14) IEEE802.15 standard, wireless personal area network (WPAN) technology.

(15) IEEE802.16 standard, broadband wireless LAN technology.

These protocol standards are very mature for different transmission media and local area network technologies.

3.3 Switched LAN
Shared Ethernet and switched LAN are two important types of LAN types, which will be introduced respectively below.

3.3.1 Shared Ethernet
Typical representatives of shared Ethernet are bus network using 10Base-2 and 10Base-5 and star network with hub as the core.

In Ethernet using a hub, the hub connects Ethernet devices collectively to a central device.In essence, this kind of Ethernet with the hub as the core is not fundamentally different from the traditional bus-type Ethernet.

The biggest weakness of shared Ethernet is that all nodes are connected in the same collision domain, no matter which node is the source and destination of a data frame, the data is widely disseminated to all nodes, that is, all nodes All nodes can receive this frame, then with the increase of nodes, a large number of conflicts will lead to a sharp decline in network performance.

The hub can only transmit one data frame at a time, which means that all ports of the hub must share the same bandwidth.If a port of a 100M switch is only connected to one node, then this node can monopolize the bandwidth of 100Mbps. This type of port is usually called a "dedicated 100Mbps port"; If the Ethernet hub has 100 ports connected to 100 computers, then this port will be shared by multiple nodes in the Ethernet. This type of port is called "shared 16Mbps port".If more machines are connected, each machine will get less speed.

3.3.2 Basic Structure of Switched LAN

The bandwidth of the shared Ethernet outlet limits the transmission speed of the connected downlink computer, and the more computers are connected, the more obvious the speed will drop.Switched LAN solves the shared problem.

A typical switched LAN is a switched Ethernet (Switched Ethernet), and its core component is an Ethernet switch (Ethernet Switch).It has multiple ports, and each port can be connected to a node or a shared device independently, and can also be cascaded with another Ethernet switch.

事实上，1000Mbps和10Gbps以太网都是全双工网络，在1000Mbps和10Gbps层次不使用集线器而使用交换机，因此在1000Mbps和10Gbps层次的以太网都是交换式以太网。

Since the switched LAN realizes concurrent transmission of data between multiple nodes through the Ethernet switch, it can greatly increase the network bandwidth and improve the performance and quality of service of the LAN.The port types of Ethernet switches are divided into half-duplex ports and full-duplex ports.

(1) For a 10Mbps port, the half-duplex port bandwidth is 10Mbps; the full-duplex port bandwidth is 20Mbps.

(2) For a 100Mbps port, the half-duplex port bandwidth is 100Mbps, while the full-duplex port bandwidth is 200Mbps.

3.3.3 Working principle of LAN switch
The LAN switch works at the second layer of the OSI model, the data link layer, and is responsible for receiving and sending data frames.

交换机有8个端口，其中端口1、2、4、7分别连接了结点A、结点B、结点C和结点D。根据以上端口号与结点MAC地址的对应关系，可以建立交换机的“端口号／MAC地址映射表”。

If node A and node D send data at the same time, respectively fill in the destination address of the frame in the destination address field (DA) of the Ethernet frame.If node A wants to send a frame to node C, the destination address of the frame is DA=node C.

Node D wants to send a frame to node B, then the destination address DA of the frame=node B.When node A and node D transmit Ethernet frames through the switch at the same time, the switching control part of the switch finds out the output port number corresponding to the destination address of the frame according to the corresponding relationship of the "port number/MAC address mapping table", and then assigns the frame to node A A connection from port 4 to port 7 is established to node C, while a connection from port 2 to port [-] is established for node D to node B.

Multiple connections between such ports can be established at the same time as required. Establishing multiple concurrent connections between multiple ports is a highly efficient working method, which avoids data transmission conflicts.

3.3.4 Switching mode of Ethernet switch

1. Ethernet frame structure

2. Switching method of Ethernet switch

The switching methods of Ethernet switches can be divided into the following three categories.

(1) Straight-through mode

In the cut-through mode, once the switch receives and detects the destination address field, it immediately forwards the frame, regardless of whether the data in the frame is wrong.The frame error detection task is completed by the node host.

(2) store and forward exchange method

In the storage and forwarding (Store &; Forward) mode, the switch first receives the frame completely and performs error detection.If the received frame is correct, it will be forwarded according to the output port number determined by the destination address of the frame, otherwise it will not be forwarded.

The cut-through and store-and-forward adaptive methods decide which of the two methods to use according to the probability of an error frame per unit time.When the probability of error frames per unit time is less than a certain threshold, the cut-through switching method is adopted; when the probability of error frames per unit time is greater than this value, the store-and-forward switching method is adopted.

(3) Improved straight-through switching method

The improved cut-through switching method is also called the fragment isolation method. It checks whether the length of the received frame is enough for 64 bytes. If it is less than 64 bytes, it means that it is a fake packet, and discards it; if it is larger than 64 bytes, it forwards it.

3.3.5 Features of Switched LAN

Switched LAN mainly refers to switched Ethernet, which has the following technical characteristics.

1. Low transmission delay

The reason is the use of a LAN switch, which has the shortest transmission delay time among network devices.

2. High transmission bandwidth

Each port of the switch has exclusive network bandwidth.For a port of a 100Mbps switch, the half-duplex port bandwidth is 100Mbps, while the full-duplex port bandwidth is 200Mbps.For Gigabit and [-] Gigabit Ethernet switches, the bandwidth per port is higher.

(End of this chapter)