Method and device for detecting network delay state

Abstract

The invention discloses a method and device for detecting a network delay state, and relates to the technical field of computers. A specific implementation of the method includes: inserting an INT header into a plurality of messages, and adding metadata encapsulated with the status information of the head node switch after the INT header, so as to obtain a plurality of detection messages including INT data. send multiple probe packets to the intermediate node, and add metadata encapsulated with the status information of the intermediate node switch; continue to forward multiple probe packets to the terminal node, and add a packet encapsulated with the status information of the terminal node switch Metadata; multiple detection packets are encapsulated and sent to the monitoring server to detect network delay status. In this implementation manner, by configuring a simple sampling rule, the head node switch actively sends an INT detection message, thereby achieving the technical effect of acquiring network delay status information and reducing the pressure on monitoring equipment at the same time.

Description

technical field [0001] The present invention relates to the field of computer technology, and in particular, to a method and device for detecting a network delay state. Background technique [0002] The current data center network carries more and more diversified services, and the traffic load is heavier. Therefore, how to solve or avoid network congestion is more and more critical for network forwarding equipment. INT (In-band Network Telemetry, in-band network telemetry) is a new type of Telemetry (telemetry technology) protocol, which can monitor the network status by collecting and reporting the network status at the data level. [0003] In the process of realizing the present invention, the inventor found that there are at least the following problems in the prior art: [0004] 1. When using INT, the traffic that may be carried on the data link needs to be predicted in advance, so the same sampling rules need to be configured on switches and service packets. For netwo...

AI Technical Summary

Problems solved by technology

[0004] 1. When using INT, it is necessary to predict the traffic that may be carried on the data link in advance, so the same sampling rules need to be configured on the switch and service packets. For networks with complex topological structures, the operation is complicated and difficult;

[0005] 2. When the configured sampling rules match randomly, all traffic in the entire network will need to insert INT data, which greatly increases the pressure on the monitoring server

Images