Application performance bottleneck detection and diagnosis method based on cluster analysis

Abstract

The invention relates to an application performance bottleneck detection and diagnosis method based on cluster analysis. The method is implemented in three parts: the data collection and real-time display module, the off-line client analyzing and displaying module and the on-line statistical analysis module based on background under the real environment of the mobile terminal. The application performance bottleneck detection and diagnosis method based on cluster analysis has the advantages that the performance analysis and the problem diagnosis can be conveniently, quickly and accurately applied, and unnecessary development cost is reduced; the time and the cost of the optimization stage are saved, the development, the testing and the optimization analysis and the like of the application are further refined, so that unnecessary work can be distributed to the appropriate personnel, the development efficiency of the application is improved, and the stability of the application is improved; at the same time, along with the increasing of data scale, more data mining and analysis can be carried out on the basis of these data information.

Description

technical field [0001] The invention relates to a method for detecting and diagnosing application performance bottlenecks based on cluster analysis. Background technique [0002] Under the current environment, with the growing popularity of the Unity3D engine, the threshold for development has been continuously lowered, and a large number of developers of various levels have poured in. The number of applications, software, and games developed based on the Unity3D engine is increasing day by day. There may be various efficiency problems in the application developed by the developer, which leads to the product not being able to meet the market demand well. [0003] To meet the needs of developers, the usual method is to use Mac's xcode for auxiliary analysis, real-time online frame-by-frame manual data collection and analysis based on Unity's built-in Profiler, but the inevitable existence of increased unnecessary additional development costs (absolutely Most development plat...

AI Technical Summary

Problems solved by technology

[0002] Under the current environment, with the growing popularity of the Unity3D engine, the threshold for development has been continuously lowered, and a large number of developers of various levels have poured in. The number of applications, software, and games developed based on the Unity3D engine is increasing day by day. There may be various efficiency problems in the application developed by the developer, resulting in the product not being able to meet the market demand well

[0003] To meet the needs of developers, the usual method is to use Mac's xcode for auxiliary analysis, real-time online frame-by-frame manual data collection and analysis based on Unity's built-in Profiler, but the inevitable existence of increased unnecessary additional development costs (absolutely Most development platforms are still windows), difficult to find potential problems, difficult to locate bottlenecks, low efficiency of manual analysis, single pain point for solutions and other problems and deficiencies

Images