A business monitoring method based on real-time statistics and alarm formula

Abstract

The invention discloses a business monitoring method based on real-time statistics and alarm formulas, and relates to the technical field of data processing. The invention includes a variable statistics subsystem and a formula alarm subsystem running in parallel, and the relationship between the variable statistics subsystem and the formula alarm subsystem Sharing data through the cache sharing system, the business monitoring method includes the following steps: S1: The variable statistics subsystem periodically and parallelly counts the values ​​of each variable according to the preset time granularity, and stores the statistical values ​​of each variable in the cache sharing system; S2: The formula alarm subsystem concurrently extracts the statistical values ​​of the variables associated with the formula calculation from the cache sharing system according to the preset cycle, and concurrently calculates the function values ​​of each associated variable, and periodically substitutes them into the formula for calculation. When the calculation result If it is true, an alarm will be issued. The present invention can flexibly configure variables and self-defined formulas to monitor the system in a more flexible and accurate manner, and adapt to flexible and changeable multi-dimensional business monitoring requirements.

Description

technical field [0001] The invention relates to the technical field of data processing, and more specifically relates to a business monitoring method based on real-time statistics and alarm formulas. Background technique [0002] The prior art usually uses the database's own statistical functions or SQL (Structured Query Language, Structured Query Language) to perform statistics, and uses hard-coded methods to determine whether the statistical value reaches a certain threshold and issue an alarm. [0003] However, the existing statistical methods have the following problems: [0004] 1. Directly use SQL for query statistics. SQL cannot be reused for data. The number of queries is directly related to the number of variables counted. If you want to count historical data (such as statistics by day), the number of queries will be more. When the amount is relatively large, query Too many times will slow down the database, affect business operation, and slow down the statistics s...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is: in order to solve the existing data statistics method to use SQL to carry out query statistics, the number of queries is directly related to the variable number of statistics linearly, when the query volume is large, the database will be slowed down, resulting in slow statistics, real-time statistics The performance is reduced, resulting in the problem that the alarm is not timely. The present invention provides a business monitoring method based on real-time statistics and alarm formulas

Images