Abnormity detection early warning method and system

An anomaly detection and model technology, applied in the field of information processing, can solve problems such as poor versatility, false positives and false negatives, and achieve the effects of reducing detection delays, good versatility, and reducing the rule configuration process

Pending Publication Date: 2020-03-06
携程旅游信息技术(上海)有限公司
View PDF3 Cites 21 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0002] Traditional operation and maintenance exception monitoring is basically based on rules, and threshold alarms are set according to the experience of business experts. Configure the corresponding rules separately on each curve. Such a scenario also brings huge challenges to the operation and maintenance department.

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
  • Abnormity detection early warning method and system
  • Abnormity detection early warning method and system
  • Abnormity detection early warning method and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0059] An anomaly detection and early warning method, which is used to monitor whether the index data of business indicators is abnormal. The anomaly detection and early warning method adopts a client 1, a monitoring platform 2, an offline model training module 3, a distributed file system 4, and a distributed message queue system 5 and distributed stream processing system 6 to achieve, figure 1 A schematic diagram of the anomaly detection and early warning method is shown, and the anomaly detection and early warning method includes:

[0060] The client 1 configures the monitored service indicators on the monitoring platform 2, and the monitoring platform 2 stores the configuration information of the service indicators in the database 7; wherein, the monitoring platform 2 can be run on the client software system on terminal 1;

[0061] The monitoring platform 2 triggers the offline model training module 3, the offline model training module 3 trains the model offline and uploa...

Embodiment 2

[0068] This embodiment is a further refinement of Embodiment 1, which provides the specific process of offline model training of the offline model training module 3 . figure 2 A flow chart of the offline model training module 3 offline training model described in this embodiment is shown, including the following steps:

[0069] Step 31: Collect historical data curves of the service indicators, where the historical data curves include historical indicator data values ​​of the service indicators at different times. The historical data curve is used as a sample data set for subsequent training models. The historical data curve can come from various sources, such as the public data set provided by the operation and maintenance platform, or monitoring the enterprise's own data.

[0070] Step 32: Perform preprocessing on the historical data curve to obtain a curve with complete data and / or no abnormal value. The historical data curve is essentially a time series. Considering the ...

Embodiment 3

[0128] This embodiment provides an anomaly detection and early warning system. The abnormal detection and early warning system includes the client 1, the monitoring platform 2, the offline model training module 3, the distributed file system 4, the distributed message queue system 5 and the distributed stream processing system 6 in the first embodiment.

[0129] The client 1 is used to configure monitored service indicators on the monitoring platform 2;

[0130] The monitoring platform 2 is used to store configuration information of the service indicators in a database;

[0131] The monitoring platform 2 is also used to trigger the offline model training module 3;

[0132] The offline model training module 3 is used for offline training model and uploads the trained model to the distributed file system 4, and the model is used to predict the monitoring index;

[0133] The client 1 is also used to push the index data of the business index to the distributed message queue syst...

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 an anomaly detection early warning method and system. The method comprises the following steps: configuring monitored service indexes on a monitoring platform, and storing configuration information of the service indexes into a database; triggering an offline model training module, training a model offline, and uploading the trained model to a distributed file system, wherein the model is used for predicting monitoring indexes; pushing the index data to a distributed message queue system in real time; loading the configuration information and then storing the model, consuming the index data from the distributed message queue system in real time, taking the index data as input of the model to obtain a prediction result, and judging whether the index data is abnormalor not; and when the index data is abnormal, writing back alarm information of monitoring abnormality to the distributed message queue system, and reading the alarm information from the distributed message queue system in real time by the client. According to the invention, the monitored index data can be consumed in real time and whether abnormality exists or not can be detected, the detection delay problem of the monitoring index is reduced, and the early warning efficiency is greatly improved.

Description

technical field [0001] The invention belongs to the field of information processing, and in particular relates to an abnormal detection and early warning method and system. Background technique [0002] Traditional operation and maintenance exception monitoring is basically based on rules, and threshold alarms are set according to the experience of business experts. To configure the corresponding rules separately on each curve, such a scenario also brings huge challenges to the operation and maintenance department. [0003] Especially for large-scale mobile Internet companies, there are a large number of releases and changes in the production environment every day. The stability of web services is definitely very important. However, the foundation of this stability needs to be guaranteed by operation and maintenance. Operation and maintenance personnel judge whether the system is abnormal by monitoring the business indicators in the system. However, due to the diversity of...

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 Applications(China)
IPC IPC(8): G06F11/34G06F11/30G06F9/54G06F11/32
CPCG06F11/3447G06F11/3476G06F11/3093G06F9/546G06F11/327G06F2209/548
Inventor 王文进潘国庆陈剑明张翼
Owner 携程旅游信息技术(上海)有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap