In what situations are the following connected devices used respectively?

Physical layer – Repeaters and hubs both serve to amplify and relay digital signals. A repeater is an interconnection device used to extend the distance of a network. (The interconnection length in a local area network is limited, not unlimited; for example, in 10M Ethernet, the maximum number of repeaters allowed in the transmission path between any two data terminal devices is 5, or 4 repeaters). A REPEATER can amplify signals that have attenuated along a line; its two ends can be connected to the same transmission medium or to different media, such as coaxial cable on one end and twisted pair on the other. A hub is essentially a multi-port repeater; it has one port connected to the backbone network and multiple ports that connect to a group of workstations. It is used in networks with a star topology to connect multiple computers or network devices. A hub is a type of shared device; it cannot identify the destination address on its own. When Host A within the same local area network sends data to Host B, the data packets are transmitted in a broadcast manner over a network based on a hub architecture, and each terminal determines whether to receive the packet by checking the address information in its header. Logical Link Layer – Bridges and Switches. The functions of the Data Link Layer include establishing, maintaining, and terminating data links, framing, frame transmission, frame synchronization, frame error control, and flow control. A bridge operates at the data link layer, connecting two local area networks (LANs). It forwards frames based on MAC addresses (physical addresses), and can be considered a “low-level router” (routers operate at the network layer and forward data based on network addresses such as IP addresses). It can effectively connect two LANs, keeping local communications within the same network segment while forwarding corresponding signals to another segment; bridges are typically used to connect a small number of network segments of the same type. A switch is a network device that uses MAC addresses for identification and is capable of encapsulating and forwarding data packets. Switches can \"learn\" MAC addresses and store them in their internal address table, thereby establishing temporary switching paths between the sender and recipient of data frames, allowing the data frames to reach their destination directly from the source address. In fact, the predecessor of a switch was a bridge. A switch uses hardware to carry out the tasks of filtering, learning, and forwarding that were previously done by software in bridges. Switches are faster than HUBs, as HUBs do not know where the target address is and send data to all ports. The switch has a routing table; if it knows where the target address is, it sends the data to that location, and if it doesn’t know, it sends it to all ports. Such filtering can help reduce the data transmission volume across the entire network and improve efficiency. But the functions of a switch go beyond that; it can divide a network into various branches, separate network data streams, and isolate faults that occur within these branches. This allows for a reduction in the amount of data traffic in each network branch, making each network more efficient and improving the overall efficiency of the network. There is a trend currently to use SWITCHes in place of HUBs. 　Switch functions: An Ethernet switch knows the MAC address of the device connected to each port, and it maps these addresses to their respective ports, storing them in a MAC address table within the switch’s cache. 　　Forwarding/Filtering: When the destination address of a data frame has an entry in the MAC address table, it is forwarded to the port connected to the destination node rather than all ports (unless the data frame is a broadcast/multicast frame, in which case it is forwarded to all ports). 　　Loop elimination: When a switch includes a redundant loop, the Ethernet switch uses the Spanning Tree Protocol to prevent the formation of loops, while still allowing for backup paths to exist. 　　In addition to connecting networks of the same type, switches can also serve to interconnect different types of networks (such as Ethernet and Fast Ethernet). Today, many switches are equipped with high-speed connection ports that support Fast Ethernet or FDDI, etc., for connecting to other switches in the network or providing additional bandwidth for critical servers with high bandwidth demands. 　　Generally, each port on a switch is used to connect an independent network segment, but sometimes, in order to achieve faster access speeds, we can directly connect some important network computers to the switch ports. In this way, the key servers and important users of the network enjoy faster access speeds, enabling handling of larger data volumes. The network layer – routers – are used to connect multiple logically separate networks, as well as several networks that use different protocols and architectures. When data is transmitted from one subnet to another, a router can be used for this purpose. It has the capability to determine network addresses and select paths, as well as to filter and separate network data streams. On the one hand, it enables communication across different types of physical networks (DDN, FDDI, Ethernet, etc.); on the other hand, it logically divides the entire interconnected network into logically independent network units, thereby giving the network a certain logical structure. For networks of different sizes, the focus of a router’s role varies: 1. In backbone networks, the main function of a router is route selection. Routers on the backbone network must know the paths to all lower-level networks. This requires maintaining a large routing table and responding as quickly as possible to changes in connection status. Failures in the router can lead to serious problems with information transmission. 2. In a regional network, the main functions of routers are network connection and route selection; that is, they connect various lower-level network units—campus networks—and are responsible for forwarding data between these lower-level networks. 3. Within a campus network, the main function of routers is to separate subnets. The basic unit of the early Internet was the local area network (LAN), in which all hosts were part of the same logical network. As the scale of networks continues to expand, local area networks have evolved into campus networks composed of multiple subnets connected by high-speed backbones and routers. In this setup, each subnet is logically independent, and the router is the only device capable of separating them; it is responsible for forwarding packets between subnets as well as ensuring broadcast isolation. Routers located at the boundaries are in charge of establishing connections to higher-level networks. Application layer – Gateway: A gateway, also known as an inter-network connector or protocol converter. A gateway enables network interconnection at the transport layer; it is the most complex device for such interconnection, and is used solely for connecting networks that use different higher-layer protocols. The structure of a gateway is similar to that of a router, with the difference lying in the interconnection layer: switches operate at the data link layer, routers at the network layer, while gateways operate at the transport layer. Gateways can be used for WAN interconnection as well as LAN interconnection. A gateway is a computer system or device that serves as a converter. Between two systems that use different communication protocols, data formats, or languages, or even have completely different architectures, a gateway acts as a translator. Only by setting the IP address of the gateway can the TCP/IP protocol enable communication between different networks. So, which machine does this IP address belong to? The IP address of a gateway is the IP address of a device with routing capabilities; such devices include routers, servers with routing protocols enabled (which are essentially equivalent to routers), and proxy servers (which are also equivalent to routers).

Repeaters and routers typically serve as extensions and expansions of a network； Gateways are used to interconnect networks of the same type ； Bridges are used to interconnect heterogeneous networks.