Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

A KPI abnormal early warning method for intelligent IT operation and maintenance system

An operation and maintenance system and abnormal technology, which is applied in the direction of network data indexing, instrumentation, and other database retrieval, can solve problems such as limiting the application of decision tree algorithms, and achieve the effect of reducing manual workload and ensuring accuracy

Active Publication Date: 2022-03-04
CHENGDU SOBEY DIGITAL TECH CO LTD
View PDF15 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the decision tree algorithm is very dependent on a large number of feature engineering. The effect of feature engineering often determines the quality of the abnormality prediction effect, so it also limits the application of the decision tree algorithm in abnormality prediction.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A KPI abnormal early warning method for intelligent IT operation and maintenance system
  • A KPI abnormal early warning method for intelligent IT operation and maintenance system
  • A KPI abnormal early warning method for intelligent IT operation and maintenance system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0067] Embodiment 1: as image 3 As shown, a KPI abnormal early warning method for an intelligent IT operation and maintenance system includes steps:

[0068] S1, determine the abnormal data of adjacent time, and form the abnormal data of adjacent time into an abnormal data group;

[0069] S2, determining the abnormal data transition process time group;

[0070] S3, calculating the trend fluctuation point within the initial time point of the abnormal data transition process time group in step S2;

[0071] S4, based on the data in steps S2 and S3, calculate and record trend information and maximum trend information;

[0072] S5, using the trend information and the maximum value trend information in step S4, to determine whether the real-time data is abnormal in the real-time abnormal warning.

[0073] Among the above five steps, step S1 to find historical abnormal data belongs to the data preparation stage. Steps S2, S3, and S4 perform feature engineering based on the histo...

Embodiment 2

[0075] Such as Figure 1~Figure 2 shown. On the basis of Embodiment 1, step S1 to determine the adjacent time abnormal data group refers to searching for data greater than or equal to the threshold from the historical KPI data, and forming adjacent abnormal data into an abnormal data group, including sub-steps:

[0076] S101, setting a threshold K, comparing the historical KPI data with the threshold K, marking the ones greater than or equal to K as 1, and marking the ones smaller than K as 0;

[0077] S102, traverse the historical KPI data, find the data whose current time point is 1, which is the abnormal data, and the previous time point is 0, which is the entry point time of the normal data, and record it in the time series time1_pre;

[0078] S103, traverse the historical KPI data, find out that the current time point is 1, which is the abnormal data, and the next time point is 0, which is the exit time of the normal data, and record it in the time series time1_last;

...

Embodiment 3

[0082] On the basis of Embodiment 2, step S2 determines the transition process time point of the abnormal data set based on the step S1 abnormal data set time1_pair, and the purpose of determining the transition process time point of the abnormal data set is for subsequent learning of the transition from the normal state to the abnormal state Prepare data for trend information, including sub-steps:

[0083] S201, traverse the abnormal entry and exit time pairs in the abnormal data group time1_pair in S105, push forward N time points for each abnormal entry point time, and record it in the time sequence time2_pre;

[0084] S202, traverse the abnormal entry and exit time pairs in the abnormal data group time1_pair, push back N time points for each abnormal exit point time, and record it in the time sequence time2_last;

[0085] S203, according to the time series obtained in S201 and S202, one-to-one correspondence is formed according to the index to form the abnormal data transi...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a KPI abnormal early warning method for an intelligent IT operation and maintenance system, comprising the steps of: S1, determining the abnormal data of adjacent time, and forming the abnormal data of adjacent time into an abnormal data group; S2, determining the abnormal data transition process time group ; S3, calculate the trend fluctuation point in the initial time point of the abnormal data transition process time group in step S2; S4, calculate and record the trend information and the most value trend information based on the data of steps S2 and S3; S5, use the trend information in step S4 The trend information and the most value trend information can be used to judge whether the real-time data is abnormal in the real-time abnormal warning. The present invention can not only accurately learn the trend information of historical abnormal data, but also avoid a large number of feature engineering, which plays a very important role in ensuring the accuracy of IT equipment abnormal prediction and reducing manual workload.

Description

technical field [0001] The present invention relates to the field of KPI abnormality prediction of an intelligent IT operation and maintenance system, and more specifically, relates to a KPI abnormality early warning method for an intelligent IT operation and maintenance system. Background technique [0002] In recent years, with the continuous development of business in various industries, more and more IT equipment has been put into the production environment of various industries. The normal operation of IT equipment is directly related to the normal operation of business systems. At present, many KPI (KeyPerformance Indicators) anomaly detection algorithms have been applied to IT equipment fault detection. However, the KPI anomaly detection algorithm can only function after an anomaly occurs, and cannot predict whether an anomaly may occur in the future. Although the occurrence of the fault can be correctly detected at this time, the business has already been affected....

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06F16/951G06F16/9537
CPCG06F16/951G06F16/9537
Inventor 张诚刘进杨瀚
Owner CHENGDU SOBEY DIGITAL TECH CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com