How does network latency occur?

Network delay refers to the time elapsed since an IP packet is sent from the source host process and reaches the destination host process, in milliseconds.

The IP packet passes from the source host through multiple relay devices, such as switches and routers, and finally reaches the destination. The network delay is composed of the following three:

Network delay = transmission delay + processing delay + buffer queue delay

Propagation delay

This delay is well understood, referring to the delay of light and electrical signals on a wired medium, or the delay of radio signals in an air medium. This transmission delay is only related to the speed of light or the transmission speed of electrical signals. . For a selected end-to-end path, this value is constant.

Processing delay

Receive IP packets, query forwarding table, hardware or software forwarding, including encapsulation, decapsulation, encoding, decoding time, the more the device passes, the weaker the device throughput (how much Gb data can be forwarded in one second), the greater the delay But for a selected end-to-end path, this value is also constant.

Buffer queue delay

When there is no network congestion on the end-to-end path, the delay of the buffer queue is zero, and the network forwarding performance reaches line speed, as if the two ends are directly connected by a line.

However, network congestion often occurs, because the host's demand for bandwidth is endless, so congestion is the norm. When the network is congested, the router will buffer the traffic that cannot be forwarded in time in the egress queue. Weighted FairQueue, Low Latency Queue, or the default First In FirstOut may be used. This queue may cause a delay of hundreds of milliseconds or more. It is variable and can be close to 0 or a few seconds, depending on the degree of network congestion.

What is Weighted FairQueue?

Put different types of traffic into different queues, assign a weight or percentage to different queues in advance, and use the fair rotation method to send the buffered data of each queue in turn.

What is Low Latency Queue?

According to IP priority, high-priority voice traffic (Precedence = 5) is sent first and can be queued, VIP customers.

What is First In FirstOut?

First come first, then later.

Therefore, the excessive network delay is mainly caused by the excessive delay of the buffer queue!

Google's latest TCP BBR algorithm is to actively measure the delay of the network and use intelligent algorithms to learn the maximum available bandwidth of the network, so that the transmission rate ≤ the maximum available bandwidth, minimize the depth of the router's buffer queue, and minimize the delay .

In summary, the network delay is composed of transmission delays that are not controlled by humans, as well as processing delays and queue delays that humans can control. Obviously, if you want to reduce the overall delay of IP packets, you must start with the latter two, that is, how to reduce processing delay and queue delay.

Processing latency is a key indicator of whether a device is a bullish fork. Taking routers as an example, how to match the correct routing entry with the fastest speed for a router with nearly one million routing tables includes full black technology. With multi-level query, you only need to query hundreds or thousands of entries to hit the correct entry. This is a technical element that needs to be realized during product design.

For network administrators, the only thing that can control the delay size is the queue delay, which requires the administrator to be proficient in the queue queuing and scheduling mechanism.

The above is the news sharing from the PASSHOT. I hope it can be inspired you. If you think today' s content is not too bad, you are welcome to share it with other friends. There are more latest Linux dumps, CCNA 200-301 dumps and CCNP Written dumps waiting for you.