Collection
zero Useful+1
zero

Network segment

Announce Upload video
Information science terminology
Network segment generally refers to the part of a computer network that can communicate directly using the same physical layer devices (transmission media, repeaters, hubs, etc.) [1]
Chinese name
Network segment [1]
Foreign name
network segment [1]
Discipline
information science [1]
Physical layer equipment
Transmission media, repeaters, hubs, etc [1]
Application
Computer communication transmission [2]
Field
Computer network field [3]

brief introduction

Announce
edit
Almost all network products are divided into multiple physical segments, which is caused by technical limitations in network implementation. A network may also be divided into multiple segments to improve performance. Of course, if you segment physically, your workstation will not be able to access resources outside the local segment. At this time, routers and bridges will be introduced [4]

Router

Announce
edit
Routers can be used to physically combine multiple segments into one logical segment. Routers work at the network protocol layer. They retransmit packets from one network segment to another based on a unique address. Routers use algorithms to establish a list of network addresses of local network segments to provide routing capability for packets requiring their intervention. If there are multiple routers in the network and one is broken, other routers will be responsible for finding another path to ensure that the data reaches the destination [4]
Of course, routers are generally expensive. Although routers built on RISC CPUs are the fastest, many routers are built on Intel CPUs. Due to the time limit provided by the network implementation, the router needs a certain amount of processing capacity. For example, if an "acknowledgement" reaches the timeout limit, it is not good to route a packet to another machine. When the packet arrives at the machine, it has requested to resend the data from the source machine. For dedicated routers, an economical option is to use the routing capabilities of Windows NT to route IPX/SPX and TCP/P packets. Another option is to delete the packet if the router cannot identify a specific data type. But deleting data packets will cause many problems [4]

Bridge

Announce
edit
Not all network protocols, such as NetBEUI, can be routed, so another method of synthesizing network segments is needed, which leads to the concept of bridge. The bridge works on the MAC layer. It retransmits data regardless of whether there is a network address (Yes, it is transparent routing; No, it is source routing). Some bridges can combine these two functions in a box, which is commonly referred to as "transparent source routing bridge". The bridge is widely used in Lan Manager based networks, and it can still be used when such networks are upgraded to Windows NT Server [4]

switching hub

Announce
edit
The switching hub usually improves the performance of the network by providing each connected network client with its own network segment. In most cases, a switching hub is used for duplex operation between a server and several clients. If there is a 10Mb adapter on the network, the customer can perform duplex operation. For example, these customers can communicate at the upstream speed of 10Mb/s or downstream speed of 10Mb, so the total bandwidth is 20Mb/s. The switching hub can also connect two network segments with different transmission speeds. The server may have a 100Mb/s network adapter The customer has only a 10Mb/s adapter, which connects the server to the 100Mb/s port of the switching hub, and connects the customer (or other hubs) to the 10Mb/port, thus connecting different network segments into a physical network segment. This improves the speed of accessing services and increases the throughput from network to customer [4]

Repeater

Announce
edit
The first problem faced by network engineers is how far a signal can travel. The technology used to create a network can load a signal for a limited distance without losing the signal [5]
When you need to connect the system in a huge building, the signal has to travel hundreds of feet, resulting in weak signal. To solve this problem, designers have come up with a simple device - repeater, which can be placed on the wire to monitor network traffic [5]
In essence, a repeater is a simple device running on the physical layer. When receiving network data streams, the repeater will repeat these data streams on another network. In other words, all data traffic of each network is repeated on another network. As mentioned earlier, the network card (over Ethernet) must first monitor the network and confirm that the network is quiet (no traffic) before transmitting data. In this way, all data streams on the two networks drift there, waiting for the monitoring results to be transmitted, which makes data sharing more difficult and retransmission more and more common [5]

Communication between network segments

Announce
edit
The network segments in the network mainly include physically isolated network segments and utilization VLAN technology Isolated network segments, most of which are isolated by VLAN technology within the LAN [6]
There are two main ways to realize the communication between VLANs, one is to use a three-layer switch, the other is to use a router. In medium-sized networks, Layer 3 switches are usually used to realize the communication between VLANs. Layer 3 switching technology combines Layer 2 switching technology and routing technology, which is faster than ordinary routers. In small networks, it is generally unnecessary to configure large servers, and there is no large data flow between VLANs. Therefore, exit routers are often used to achieve mutual communication between VLANs [6]
In the three-layer switch, you can create an SV interface for each VLAN, configure an IP address on the interface, and the address will become the default gateway of the connected network segment [6]

Network segment and port

Announce
edit
In IP, the situation shown in Figure 1 is illegal. In this router, Port 1 and Port 2 have been configured on the same Class G network. Although this configuration is conducive to network scalability, most routers cannot handle this situation. The reason is that it is related to the routing table. Take a look at the example of router configuration in Figure 1 [3]
Figure 1 Error Port Example
In this case, the outbound group will arrive at the next station 192.32.17.147 [3]
The mask to be transmitted is equal to: 255.255.255.0 [3]
The router performs binary AND operation on the relay station and the mask [3]
Result of binary and: 192.32.17.0 [3]
The router looks up an internal table to see which port matches the network ID. To do this, the router will perform an AND operation on the mask configured for each port. To save time, this calculation is performed in advance in the internal table [3]
Port 1: 192.32.17.5 and 255.255.255.0=192.32.17.0 [3]
Port 2: 192.32.17.6 and 255.255.255.0=192.32.17.0 [3]
In this way, the router cannot distinguish between these output ports. In theory, the router can divide the load between matching ports equally on the basis of each packet. In fact, this will cause packets to arrive at the receiving host out of order, resulting in running conflicts [3]

Create virtual network segment

Announce
edit
Separate network access can greatly reduce the threat to management services. Then the previous example introduced that the external Internet oriented public service virtual machine should also not be in the same network segment as the trusted internal virtual machine. Xen provides this capability. The network service of a virtual machine is given as the IP address of the global routing of the public network segment as the interface, and a virtual network segment is created for internal virtual machines to use. Firewall or NAT VM can be installed on both internal and external networks [7]
Xen also has the ability to completely isolate the virtual machine from the public network. For example, the virtual machine sharing files can be placed on the second virtual network segment, which has no firewall or NAT support. Any internal virtual machine request to access the Internet and internal shared files requires two virtual network interfaces. One is on the isolated file sharing network, and the other is on the internal firewall network segment. Similarly, internally accessed virtual machines, such as data mining virtual machines or payroll computing virtual machines, must be allocated to isolated virtual machine network segments. Xen provides this strong network isolation support [7]

IP address

Announce
edit
In order to identify computers in the network and ensure the accuracy of computer communication on the Internet, each computer must have a unique identification address, just like each of us has a unique ID number, which is an IP address [2]
IP addresses are assigned by the Internet Network Number Assignment Authority (IANA for short) and cannot be used in the Internet at will, IP address is a 32-bit binary address. In order to facilitate memory, they are divided into four groups. Each group is divided into eight digits by a decimal point. Each group is represented by a corresponding decimal number between 0 and 255. This format of address is called dotted decimal address, such as 202.102.192.68 [2]
IP addresses are divided into five categories: A, B, C, D and E. The first byte in the range of 1-127 is Class A, 128-191 is Class B, 192-223 is Class C, 224-239 is Class D, and 240-255 is Class E [2]
Among them, Class A, Class B and Class C addresses are host addresses, Class D addresses are multicast addresses, and Class E addresses are reserved for future use [2]
Among the above addresses, there are also some addresses for internal use in the LAN, which are called private addresses or intranet addresses [2]
Class A reserved address: 10.0.0.0-10.255.255.255 [2]
Class B reserved address: 172.16.0.0~172.31.255.255 [2]
Class C reserved address; 192.168.0.0~192.168.255.255 [2]
What the Internet uses is TCP/IP protocol IP protocol is the core of TCP/IP protocol cluster. At present, the version number of the IP protocol is 4 (IPv4 for short), which has been used for more than 30 years. The number of IPv4 address bits is 32, that is, up to 2 thirty-two Computers can be connected to the Internet. Due to the vigorous development of the Internet, the demand for IP addresses is growing, and IP address resources are about to dry up [2]
In order to reduce the waste of IP address, people use subnet mask, network address translation (NAT) and other technologies. In order to expand the address space, it is proposed to redefine the address space through the new IP protocol IPv6. IPv6 is Next generation Internet Protocol, with 128 bit address length, can provide addresses almost without restrictions. At present, IPv6 is gradually replacing IPv4 [2]