Multi-hierarchy load balancing cloud resource monitoring method

Abstract

The invention discloses a multi-hierarchy load balancing cloud resource monitoring method. The multi-hierarchy load balancing cloud resource monitoring method includes the steps that a distributed type multi-layer monitoring framework is set up; the resource states of all monitoring nodes are analyzed, and all resource monitoring scripts are automatically generated according to cloud platform user designated requirements; cluster center nodes integrate active and passive mainframe discovery strategies to continuously monitor local area networks which the cluster center nodes are responsible for; a monitoring center node adjusts monitoring strategies such as transmission protocols according to the load states of all the cluster center nodes, the transmission expenditure of monitoring data is reduced through a fast-compression and bidirectional-transmission technology, and the cluster center nodes are switched in time when the monitoring center node breaks down; the monitoring center node conducts mining and analyzing on historical monitoring data of all resources regularly and conducts warning and deep monitoring in advance. According to the multi-hierarchy load balancing cloud resource monitoring method, new cluster nodes are automatically found according to the features of the large-scale distributed type cloud environment, a load balancing technology is used, a powerful fault-tolerant mechanism and a powerful event engine are adopted, and therefore the high reliability and safety of the cloud environment are ensured.

Description

technical field [0001] The invention relates to cloud computing technology, in particular to a multi-level load balancing cloud resource monitoring method supporting active discovery. Background technique [0002] With the continuous deepening of the Internet era, the era of cloud computing and big data has arrived, the traditional IT architecture can no longer support the massive data generated, and the scale of data centers is showing an explosive growth trend. The traditional physical resource monitoring method can no longer meet the current needs, and a large amount of human operation and maintenance costs have brought a serious burden to the enterprise. In this situation, enterprises have replaced traditional IT infrastructure with cloud environments, but whether large-scale flat cloud resources are safe and reliable, and whether the overall performance of cloud services is as stable and reliable as traditional physical servers, requires a powerful The cloud monitoring...

AI Technical Summary

Problems solved by technology

OpenStack's Ceilometer, as a community open source project, has only recently been officially released to OpenStack in a stable version. The stability and reliability have yet to be verified. So far, Ceilometer is limited by OpenStack and has poor compatibility.

Nagios provides a powerful monitoring framework, a complete alarm system and fault-tolerant control, but there are fewer component monitoring items, and currently there is no persistent storage of alarm data and a friendly graphical interface display

Ganglia provides a powerful monitoring mechanism, rich monitoring items, flexible monitoring topology, and perfect monitoring chart display, but lacks an alarm function, and users can only find problems manually

Images