High-performance distributed data center monitoring framework

Abstract

The invention discloses a high-performance distributed data center monitoring framework. The framework structurally comprises monitoring core engines, monitoring and dispatching progresses, active monitoring and roll poling devices, passive monitoring receivers, warning engines and monitoring data processing centers. The monitoring framework is designed in a distributed manner, the involved processing processes in the framework are separated, detailed and modularized, the framework is divided into six modules for finishing work of each stage, and one monitoring core engine is reserved for finishing dispatching operation of each module, so that the consumed resource quantity during operation of the whole monitoring framework is reduced, equal distribution of the consumed resources in each module is realized, finally, the high performance during data center monitoring is realized, and the monitoring scale can be expanded to tens of thousands of nodes even hundreds of thousands of nodes.

Description

technical field [0001] The invention relates to the fields of distributed monitoring and data center monitoring, in particular to a high-performance, distributed, and large-scale monitoring data center monitoring architecture. Background technique [0002] At present, the scale of data centers is getting larger and larger, and the demand for high-performance data center monitoring is becoming stronger and stronger. However, because the traditional monitoring architecture only has a bloated and inefficient monitoring core, various data collection, processing and analysis, etc. The work efficiency is very low, and there are performance bottlenecks that cannot be solved, and it is impossible to monitor large-scale data centers. In practical applications, as data centers are built bigger and bigger, the requirements for data center monitoring are getting higher and higher, the traditional monitoring architecture can no longer meet the usage requirements, and the performance bott...

AI Technical Summary

Problems solved by technology

[0002] At present, the scale of data centers is getting larger and larger, and the demand for high-performance data center monitoring is becoming stronger and stronger. However, because the traditional monitoring architecture only has a bloated and inefficient monitoring core, various data collection, processing and analysis, etc. Work efficiency is very low, and there are performance bottlenecks that cannot be solved, and it is impossible to monitor large-scale data centers

In practical applications, as the data center is getting bigger and bigger, and the requirements for data center monitoring are getting higher and higher, the traditional monitoring architecture can no longer meet the usage requirements, and the performance bottleneck problem is also very serious

This traditional bloated monitoring architecture that has performance bottlenecks and integrates all processing, when monitoring a large number of equipment resources in the data center, has very low efficiency, serious resource consumption, and performance bottlenecks. It can only monitor 2,000 nodes. scale data center monitoring

Images