Monitoring data adaptive compression method and system

Abstract

The invention provides a self-adaptive monitoring data compression method and system. The method comprises the following steps that: (1), a monitoring system is constructed; (2), a threshold value is set for monitoring data collected by a monitoring server; and whether the change value of the monitoring data collected by the monitoring server is greater than the threshold value or not is judged, such that corresponding operations are performed; and (3), a function relationship among a plurality of hosts existing in a cluster or a cloud platform is regularly found and calculated; the monitoring data are sent to the monitoring server only by optionally selecting one host from the hosts having the function relationship if the function relationship exists among the hosts; and corresponding various hosts need to send the monitoring data to the monitoring server if the function relationship does not exist among the hosts. According to the invention, the number of the hosts in need of sending the monitoring data is reduced by reducing the sending data volume of a single host and finding the function relationship existing among the hosts; and thus, the problem that the monitoring data in a network are explosively increased due to the fact that the number of the hosts in a monitoring platform is large can be effectively solved.

Description

technical field [0001] The invention relates to the management of monitoring data, in particular to a monitoring data self-adaptive compression method and system. Background technique [0002] Monitoring systems are usually used to collect runtime status information of hosts in computing clusters or cloud platforms. By implanting sensors in the host, the host can communicate with the monitoring server through a predetermined protocol. The sensor can work according to the default configuration of the local configuration file, and the server can also remotely configure the sensor on any host to work in a specific mode, such as setting the sampling frequency, starting the collection of new information or pausing the sensor. Generally speaking, computing clusters or cloud platforms contain hundreds of hosts, and each host needs to collect a large amount of monitoring data and send it to the monitoring server, resulting in an explosive growth of data traffic in the network, maki...

AI Technical Summary

Problems solved by technology

Generally speaking, computing clusters or cloud platforms contain hundreds of hosts, and each host needs to collect a large amount of monitoring data and send it to the monitoring server, resulting in an explosive growth of data traffic in the network, making the monitoring server face a heavy burden

Therefore, there is an urgent need to solve the problem of data explosion in surveillance systems

Images