What is QoS)

Detailed description of QoS traffic control management!
More and more routers use QoS. As a network administrator. QoS is becoming increasingly popular in our ears. What is QoS traffic control? Let's take a look at it today.
To put it simply. QoS is a technology used to solve network latency and congestion problems! Take the sea spider as an example: Generally, it includes priority (packet priority, IP address segment priority, port priority, residual bandwidth preemption priority), elastic bandwidth management, and so on.

Packet priority: A packet is usually smaller than 64 KB, for example, a ping packet. Request package and response package. When surfing the Internet. There will be many requests and responses. After these packages are given priority, we will feel much faster.

IP address segment priority: source IP address and destination IP address can be used. Or segment to set priority. For example. Lan. First, ensure the bandwidth of my web server. The client bandwidth is guaranteed, and the server bandwidth is guaranteed. Can be set.

Port priority: you can set a priority level for the source port and target port. In general. Internet cafes are dominated by games. Then I can target ports of some mainstream games. Give priority to the bandwidth of these games.

Remaining bandwidth preemption priority: As the name implies, it indicates that the instance is in use. When the network bandwidth is available. The remaining bandwidth can be allocated to some IP addresses and some ports. In this way, you can make full use of the bandwidth. Not a waste!

Elastic bandwidth management:
Take the download speed limit as an example. For example, the total bandwidth is 10 Mbit (the total maximum download speed is 1000 kbytess ). The download speed is 100 kbytess and the maximum download speed is 200 kbytess. The details are as follows:
1. When the bandwidth is idle, the speed can reach the "maximum speed"
If the bandwidth is idle (the number of Internet users is relatively small, and the bandwidth usage is around 50%, for example, only 10 online users), the download speed can reach up to 200 kb, 200 k is the peak download speed. Even if only one person is using the bandwidth, it will not exceed the peak speed. The 200 K in the 100 k is borrowed from another person for the time being, and will be automatically withdrawn when others need it.
2. There is a certain bandwidth usage rate between "guaranteed speed" and "maximum speed ".
If the bandwidth has a certain usage rate (with a certain number of Internet users, the bandwidth usage rate is around 80%, for example, 15 online users ), the download speed is reduced to 100 kb to kb ~ 200 K.
In general, this situation leads the majority.
3. high bandwidth usage or full usage. The speed is equal to or less than the "guaranteed speed"
If the bandwidth usage is relatively high (there are more people accessing the internet, and the bandwidth usage is above 90%, for example, if all 20 people are online), will the download speed exceed 100 kb? That is: if the total bandwidth cannot meet the "guaranteed speed", the speed of the end user will be less than the "guaranteed speed" (after the average allocation ).
For example, if 20 people download all at the same time, the total download speed of 1000 kbytess is 50 K for each person.

In addition, there is often a global bandwidth setting. This is used to count your total bandwidth. If this parameter is not set. The route cannot calculate your total bandwidth, and you cannot perform elastic bandwidth management or other settings. QoS. Global settings must be set. And. It is best to use real bandwidth. Do not be too small or too large.