CN111767195A - Intelligent noise reduction processing method for alarm information - Google Patents
Intelligent noise reduction processing method for alarm information Download PDFInfo
- Publication number
- CN111767195A CN111767195A CN202010907421.8A CN202010907421A CN111767195A CN 111767195 A CN111767195 A CN 111767195A CN 202010907421 A CN202010907421 A CN 202010907421A CN 111767195 A CN111767195 A CN 111767195A
- Authority
- CN
- China
- Prior art keywords
- alarm
- merging
- strategy
- association
- alarm information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3065—Monitoring arrangements determined by the means or processing involved in reporting the monitored data
- G06F11/3072—Monitoring arrangements determined by the means or processing involved in reporting the monitored data where the reporting involves data filtering, e.g. pattern matching, time or event triggered, adaptive or policy-based reporting
- G06F11/3082—Monitoring arrangements determined by the means or processing involved in reporting the monitored data where the reporting involves data filtering, e.g. pattern matching, time or event triggered, adaptive or policy-based reporting the data filtering being achieved by aggregating or compressing the monitored data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/32—Monitoring with visual or acoustical indication of the functioning of the machine
- G06F11/324—Display of status information
- G06F11/327—Alarm or error message display
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2465—Query processing support for facilitating data mining operations in structured databases
Abstract
The invention belongs to the technical field of operation and maintenance monitoring systems, in particular to a processing method for intelligent noise reduction of alarm information, which comprises the steps of compressing the alarm quantity and merging the alarm association relation, and S1, compressing the alarm quantity; the compression of the alarm quantity refers to that the alarm information is compressed and integrated and then sent, alarm trend prediction is carried out through analysis of historical alarm data volume, and alarm compression is carried out when the alarm quantity exceeds a threshold value range; s2, merging alarm association relations; the merging of the alarm association relation refers to performing association rule data mining on the alarm information, and writing the association rules into a rule base after later-stage arrangement. The invention has the advantages that the monitoring burden of an operation and maintenance engineer is greatly reduced, so that the alarm can be effectively and timely analyzed and processed, and meanwhile, the relevance information among the alarms can be effectively ensured by analyzing and processing the convergence of the alarm elements.
Description
Technical Field
The invention relates to the technical field of operation and maintenance monitoring systems, in particular to a processing method for intelligently reducing noise of alarm information.
Background
The operation and maintenance monitoring system is the most important ring in the whole product life cycle, plays a very important role in the health condition of product operation, and is a standard for measuring the quality of a product. The monitoring system can comprehensively monitor and alarm the server, the operating system, the middleware and the application, can achieve the purposes of early warning and fault finding in advance, and provides detailed data for replying problems after the matters.
However, the monitoring system may generate a large amount of alarms in a short time, which not only causes a great pressure on the short message gateway, but also greatly increases the monitoring burden of the operation and maintenance engineer, so that the monitoring system may not be able to analyze and process the alarms timely and effectively. Furthermore, direct analytical processing of the alarm elements may result in ignoring the relevance information between alarms. Therefore, the alarm needs to be summarized and associated in a convergence manner. In the operation and maintenance monitoring system, an alarm convergence engineer refers to the process of analyzing, merging and discarding alarm information, and reduces the scale of the alarm information and the pressure of network operation and maintenance in the process.
Disclosure of Invention
The invention aims to provide an intelligent noise reduction processing method for alarm information, which not only solves the problem that the existing monitoring system can generate a large amount of alarms in a short time to cause huge pressure on a short message gateway, but also greatly increases the monitoring burden of an operation and maintenance engineer so that the alarms can not be analyzed and processed effectively in time. Moreover, the direct analysis and processing of the alarm elements may result in ignoring the relevance information between alarms, thereby effectively solving the problems proposed in the background art.
In order to achieve the above purpose, the present invention provides the following solutions: a processing method for intelligent noise reduction of alarm information comprises the following steps:
s1, compressing the alarm quantity; counting historical alarm data of a receiver in the previous month, performing trend prediction on the alarm quantity in the current time period by taking hours as a unit, and obtaining a predicted value;
s2, if the current alarm quantity is larger than the estimated value, judging that large-scale alarm occurs and compressing the alarm quantity in the time interval;
s3, merging alarm association relations; according to the compressed alarm information obtained in the step S2, mining and merging the alarm items with time sequence association relation by using a time sequence association rule mining algorithm Apriori, or mining and merging the alarm items with the same or similar service granularity with policy association relation by using a policy association rule mining algorithm;
and S4, merging all the alarm information merged in the S3 and sending the merged alarm information to the same alarm receiver.
Preferably, the alarm trend prediction algorithm comprises the following steps:
s21, establishing an alarm quantity statistical model based on a large amount of historical alarm data to obtain the distribution rule of the data;
s22, obtaining a large-scale alarm threshold value by solving maximum likelihood estimation;
and S23, optimizing and adjusting through coefficient compensation, and outputting a rule file of large-scale alarm threshold values in hours.
Preferably, the input of the time sequence association rule mining algorithm is based on a time sequence alarm data sequence, a support candidate set with a time window is obtained firstly, then support counting is carried out according to the candidate set, confidence is calculated, and an association rule file obtained by judging according to a confidence threshold is output.
Preferably, the input of the strategy association rule mining algorithm is a service granularity alarm data sequence, the service granularity is a configuration window, ip and an index code from small to large, namely the execution priority of the algorithm is merging according to the configuration window, merging according to the ip and merging according to the index code, and finally a strategy rule file is output.
More preferably, an execution priority policy is formulated for the alarm convergence scheme, and the specific execution policy priority is as follows:
s51, if the alarm trend prediction algorithm obtains that the current condition of large-scale alarm is satisfied, then combining all the alarm information belonging to the same alarm receiver name into one piece to be sent to the alarm receiver;
s52, if a plurality of association strategies generate alarms at the same time, merging the time sequence association rules by adopting a time sequence association rule mining algorithm;
s53, if multiple machines have multiple strategies to generate alarms under the same service, adopting strategy association rule mining algorithm to merge the strategies according to the configuration window to merge the alarm information;
s54, if the condition that a plurality of instances generate alarms on the same machine is met, alarm information is merged according to an ip merging strategy in a strategy association rule mining algorithm;
s55, if multiple instances generate alarms under the same strategy, adopting a strategy association rule mining algorithm to merge the alarms according to a monitoring strategy;
s56, if multiple strategies on the same monitored object are met to generate alarms, alarm information is merged according to the index code merging strategy in the strategy association rule mining algorithm.
The beneficial effects of the invention are as follows: the alarm is intelligently summarized and associated in a convergence mode, so that the problem that the conventional monitoring system can generate a large amount of alarms in a short time and great pressure is caused to a short message gateway is solved, the monitoring burden of an operation and maintenance engineer is greatly reduced, and the alarm can be effectively analyzed and processed in time. Meanwhile, the convergence analysis processing of the alarm elements can effectively ensure the relevance information among the alarms.
Drawings
Fig. 1 is a frame diagram of an alarm convergence algorithm of a processing method for intelligent noise reduction of alarm information according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments.
As shown in fig. 1, a processing method for intelligent noise reduction of alarm information includes alarm convergence, which is composed of two parts, namely compression of alarm quantity and combination of alarm association.
The compression of the alarm quantity refers to that the alarm information is compressed and integrated and then sent, generally, the alarm trend prediction is carried out through the analysis of historical alarm data quantity, when the alarm quantity obviously exceeds the threshold range, the alarm compression is carried out, the problem that different servers report the same fault is emphasized, the solution is that the alarm is delayed firstly, then the alarm information is compressed and integrated and then sent, the fact that the information received by each receiver is essence is guaranteed, the alarm correlation is emphasized on solving the problem that the same fault triggers a plurality of alarms, the solution is that the alarm information is subjected to correlation rule data mining, and the rules are sorted at the later stage and then written into a rule base.
The invention relates to a processing method for intelligently reducing noise of alarm information, which comprises the steps of firstly, judging whether large-scale alarm occurs currently by adopting an alarm trend prediction algorithm, comprehensively utilizing a mathematical method to carry out modeling and parameter estimation, judging that large-scale alarm occurs and compressing alarm items in a time period if the current alarm quantity is far larger than a predicted value obtained by the algorithm, then mining and merging the alarm items with a time sequence association relationship by adopting a time sequence association rule mining algorithm Apriori, finally mining and merging the strategy association relationship of the alarm items with the same or similar service granularity by adopting a strategy association rule mining algorithm, and storing a newly added merging strategy in a strategy association relationship library.
Considering that the alarm data is time sequence data, three aspects of 'trend prediction mining', 'time sequence association rule' and 'strategy association' are provided aiming at the time sequence data, and the three algorithms are ordered according to a certain priority to jointly complete an alarm convergence task. The core of alarm convergence is alarm compression and alarm association, an alarm trend prediction algorithm can realize alarm compression, and a time sequence association rule mining algorithm and a strategy association rule mining algorithm can realize alarm association.
The input of the alarm trend prediction algorithm is the alarm amount counted per hour, firstly, an alarm amount statistical model is established based on a large amount of historical alarm data to obtain the distribution rule of the data, secondly, a large-scale alarm threshold value is obtained by solving maximum likelihood estimation, then, the optimization adjustment is carried out through coefficient compensation, and finally, a rule file of the large-scale alarm threshold value counted per hour is output.
The input of the time sequence association rule algorithm is based on a time sequence alarm data sequence, firstly a support candidate set with a time window is obtained, then support counting is carried out according to the candidate set, then confidence coefficient is calculated, and an association rule file obtained by judging according to a confidence coefficient threshold value is output.
The input of the strategy association rule mining algorithm is a service granularity alarm data sequence, and the service granularity is a configuration window, an ip and an index code from small to large. Therefore, the execution priority of the algorithm is merging according to the configuration window, merging according to the ip, merging according to the index code, and finally outputting the strategy rule file. An execution priority strategy is drawn up for the three alarm convergence schemes, when corresponding conditions are sequentially met from front to back, the alarm information is merged according to the priority, so that the purpose of alarm convergence is achieved, and the specific execution strategy has the following priority:
if the current large-scale alarm condition is met by the alarm trend prediction algorithm, all the alarm information belonging to the same alarm receiver name is combined into one piece to be sent to the alarm receiver.
If a plurality of association strategies generate alarms at the same time, merging the time sequence association rules by adopting a time sequence association rule mining algorithm.
If multiple machines generate alarms by multiple strategies under the same service, alarm information is merged by adopting a strategy of merging according to a configuration window in a strategy association rule mining algorithm.
If the condition that a plurality of instances generate alarms on the same machine is met, alarm information is merged by adopting a strategy of merging according to ip in a strategy association rule mining algorithm.
If multiple instances generate alarms under the same strategy, alarm information is merged by adopting a strategy of merging according to monitoring strategies in a strategy association rule mining algorithm.
If the alarm is generated by a plurality of strategies on the same monitored object, the strategy of merging according to index code in the strategy association rule mining algorithm is adopted to merge the alarm information.
The alarm trend prediction algorithm principle of the invention is as follows: and counting historical alarm data of the alarm receiver in the last month, performing trend prediction on the alarm amount in the current time period by taking hours as a unit, judging that large-scale alarm occurs when the alarm amount exceeds a prediction threshold value, and merging all alarm information in the time period and sending the merged alarm information to the same alarm receiver. When the overall distribution type of the historical alarm data is known, the maximum likelihood estimation is the optimal point estimation method, so the maximum likelihood estimation in the point estimation is adopted in the node to realize the estimation of the alarm trend. The alarm prediction trend analysis process comprises the steps of denoising data by using the sub-sites, establishing a prediction model by using historical data, and carrying out maximum likelihood estimation according to the model to obtain a prediction value.
Statistical model
The alarm trend prediction needs to preprocess data to eliminate dirty data, and establish a prediction model after obtaining pure alarm quantity data. According to historical data, the alarm quantity received by each person in the same hour every day is drawn into a histogram and a line graph by taking a month as a unit, and the alarm receiving quantity of each person in the same hour every day can be preliminarily judged to be in accordance with normal distribution. And (4) according to the normal distribution model, adopting a maximum likelihood estimation method to obtain the estimated alarm quantity threshold value of each person in the same hour every day.
Maximum likelihood estimation
There are several possible results of a test, which can be expressed as:
if it is used
The frequency of Ai occurrence in n experiments is shown by Bernoulli's law of large numbers as formula (3). It can be seen that in equation (1), the time having the highest probability also tends to have the highest frequency per unit time.
And reversely carrying out reverse reasoning on the reasoning mode: when a certain event Ai occurs in a test, it is reasonable to recognize that P (Ai) is the largest one of P (a1), P (a2),. and P (Ak), i.e., Ai is the event which is "most likely to occur" in a1, a2, and Ak, and the inference mode is the so-called maximum likelihood principle, and the estimation based on the maximum likelihood principle is called maximum likelihood estimation.
Research on timing association rule mining algorithm
The Apriori algorithm is a classic algorithm which has the most influence on mining a frequent item set of a Boolean association rule, and the core idea of the Apriori algorithm is to mine the frequent item set, namely the frequent item set, also called the item set, through two stages of candidate set generation and downward closed detection, and the Apriori algorithm is a two-item set. The data mining step based on Apriori algorithm can be divided into the following two steps:
1) finding out all frequent item sets, namely frequency, according to the support degree;
2) an association rule, i.e. strength, is generated based on the confidence.
Algorithm core principle and process
The Apriori algorithm accumulates the count of each item by traversing the database, collects the items satisfying the minimum support degree, finds out the set L1 of frequent item sets, and is an iterative method of layer-by-layer search. The Apriori nature of the Apriori algorithm can be used to compress the search space, thereby increasing the efficiency of layer-by-layer generation of frequent sets of terms. The two prior properties of the frequent item set are as follows:
a priori property 1 all non-empty subsets of the frequent itemsets must also be frequent;
a priori property 2 a superset of the infrequent item set must be infrequent.
The core idea of the algorithm is as follows:
finding a frequent item set, scanning, counting, comparing, generating the frequent item set, connecting and pruning, generating a candidate item set, and repeating the steps until a larger frequent item set cannot be found;
an association rule is generated, and the process bits, based on the definition of confidence, for each frequent item set L,
all non-empty subsets S of L are generated, and if P (B | A) ≧ min _ conf, the rule L → S is output.
Research on strategy association rule mining algorithm
The strategy association rule mining algorithm is closely related to the monitoring service, abnormal alarm information generated by the strategy is firstly filtered, if the abnormal alarm information is not filtered, the merging strategy is triggered under the condition that a merging window is met, the merging window is a time period for merging the alarm information, and once the abnormal alarm information is found, the system sends the merged alarm information to an alarm receiver.
Currently available merging strategies include merging by ip, merging by code, merging by configuration unit, merging by monitoring strategy, and the like.
When multiple instances trigger alarms on one machine, all alarms on the machine can be merged to achieve the aim of alarm convergence. If a certain strategy meets the condition of merging windows and a plurality of other strategies generate alarms simultaneously on the host of the strategy, the alarms are merged and sent to reduce the number of alarms.
Merging according to codes refers to merging a plurality of strategy alarms under the same monitored object. If the alarms generated by a plurality of rule belong to the same code, the alarms are preferably combined according to the code. Merging by configuration unit means merging alarms under one service together for transmission. When a plurality of strategies of a plurality of machines or instances simultaneously need to be alarmed to trigger service alarm combination under one service, the service is positioned in a configuration unit in the operation and maintenance monitoring system. For example, the policies are set in three sections, such as service-ff.rule.all: ip: CPU _ IDLE, if the first two fields are the same, the policies are considered to be combinable, and if a certain policy satisfies merge _ window, the alarms generated by all the associated rules are combined and sent out.
Merging according to a monitoring strategy means that if a plurality of machines or instances generate alarms under one monitoring strategy, redundant alarms under the strategy are merged together. And if one alarm meets the merge _ window condition and is not sent out by the merging strategy, independently sending alarm information.
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art should be considered to be within the technical scope of the present invention, and the technical solutions and the inventive concepts thereof according to the present invention should be equivalent or changed within the scope of the present invention.
Claims (5)
1. A processing method for intelligent noise reduction of alarm information is characterized by comprising the following steps:
s1, compressing the alarm quantity; counting historical alarm data of a receiver in the previous month, performing trend prediction on the alarm quantity in the current time period by taking hours as a unit, and obtaining a predicted value;
s2, if the current alarm quantity is larger than the estimated value, judging that large-scale alarm occurs and compressing the alarm quantity in the time interval;
s3, merging alarm association relations; according to the compressed alarm information obtained in the step S2, mining and merging the alarm items with time sequence association relation by using a time sequence association rule mining algorithm Apriori, or mining and merging the alarm items with the same or similar service granularity with policy association relation by using a policy association rule mining algorithm;
and S4, merging all the alarm information merged in the S3 and sending the merged alarm information to the same alarm receiver.
2. The intelligent noise reduction processing method for alarm information according to claim 1, wherein the alarm trend prediction algorithm comprises the following steps:
s21, establishing an alarm quantity statistical model based on a large amount of historical alarm data to obtain the distribution rule of the data;
s22, obtaining a large-scale alarm threshold value by solving maximum likelihood estimation;
and S23, optimizing and adjusting through coefficient compensation, and outputting a rule file of large-scale alarm threshold values in hours.
3. The intelligent noise reduction processing method for alarm information according to claim 1, wherein the input of the time-series association rule mining algorithm is based on a time-series alarm data sequence, a support candidate set with a time window is obtained first, then support count is counted according to the candidate set, then confidence is calculated, and an association rule file obtained by judgment according to a confidence threshold is output.
4. The intelligent noise reduction processing method for the alarm information according to claim 1, wherein the input of the strategy association rule mining algorithm is a service granularity alarm data sequence, and the service granularity is a configuration window, an ip and an index code from small to large, that is, the execution priority of the algorithm is merging according to the configuration window, merging according to the ip and merging according to the index code, and finally a strategy rule file is output.
5. The intelligent noise reduction processing method for alarm information according to claim 4, wherein an execution priority policy is formulated for the alarm convergence scheme, and the specific execution policy priority is as follows:
s51, if the alarm trend prediction algorithm obtains that the current condition of large-scale alarm is satisfied, then combining all the alarm information belonging to the same alarm receiver name into one piece to be sent to the alarm receiver;
s52, if a plurality of association strategies generate alarms at the same time, merging the time sequence association rules by adopting a time sequence association rule mining algorithm;
s53, if multiple machines have multiple strategies to generate alarms under the same service, adopting strategy association rule mining algorithm to merge the strategies according to the configuration window to merge the alarm information;
s54, if the condition that a plurality of instances generate alarms on the same machine is met, alarm information is merged according to an ip merging strategy in a strategy association rule mining algorithm;
s55, if multiple instances generate alarms under the same strategy, adopting a strategy association rule mining algorithm to merge the alarms according to a monitoring strategy;
s56, if multiple strategies on the same monitored object are met to generate alarms, alarm information is merged according to the index code merging strategy in the strategy association rule mining algorithm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010907421.8A CN111767195A (en) | 2020-09-02 | 2020-09-02 | Intelligent noise reduction processing method for alarm information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010907421.8A CN111767195A (en) | 2020-09-02 | 2020-09-02 | Intelligent noise reduction processing method for alarm information |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111767195A true CN111767195A (en) | 2020-10-13 |
Family
ID=72729262
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010907421.8A Pending CN111767195A (en) | 2020-09-02 | 2020-09-02 | Intelligent noise reduction processing method for alarm information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111767195A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112737839A (en) * | 2020-12-28 | 2021-04-30 | 上海联蔚盘云科技有限公司 | Method and equipment for self-adaptive fault repair in multi-public cloud environment |
| CN112887310A (en) * | 2021-01-27 | 2021-06-01 | 华南理工大学 | Method, device and medium for improving network attack risk assessment efficiency |
| CN113778783A (en) * | 2021-07-26 | 2021-12-10 | 北京芬香科技有限公司 | Intelligent alarm method and system based on monitoring data |
| CN113868207A (en) * | 2021-10-09 | 2021-12-31 | 北京中水科水电科技开发有限公司 | Compression display method and system for information redundancy jitter characteristics |
| CN115412422A (en) * | 2022-08-08 | 2022-11-29 | 浪潮云信息技术股份公司 | Dynamic window adjusting system |
| CN117527523A (en) * | 2023-11-23 | 2024-02-06 | 广东堡塔安全技术有限公司 | Cloud computing-based server security monitoring system |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111563524A (en) * | 2020-03-18 | 2020-08-21 | 宁波送变电建设有限公司永耀科技分公司 | Multi-station fusion system operation situation abnormity monitoring and alarm combining method |
-
2020
- 2020-09-02 CN CN202010907421.8A patent/CN111767195A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111563524A (en) * | 2020-03-18 | 2020-08-21 | 宁波送变电建设有限公司永耀科技分公司 | Multi-station fusion system operation situation abnormity monitoring and alarm combining method |
Non-Patent Citations (1)
| Title |
|---|
| 国悦婷: "运维监控系统告警收敛的算法研究与应用", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112737839A (en) * | 2020-12-28 | 2021-04-30 | 上海联蔚盘云科技有限公司 | Method and equipment for self-adaptive fault repair in multi-public cloud environment |
| CN112887310A (en) * | 2021-01-27 | 2021-06-01 | 华南理工大学 | Method, device and medium for improving network attack risk assessment efficiency |
| CN112887310B (en) * | 2021-01-27 | 2022-09-20 | 华南理工大学 | Method, device and medium for improving network attack risk assessment efficiency |
| CN113778783A (en) * | 2021-07-26 | 2021-12-10 | 北京芬香科技有限公司 | Intelligent alarm method and system based on monitoring data |
| CN113778783B (en) * | 2021-07-26 | 2023-12-12 | 北京芬香科技有限公司 | Intelligent alarm method and system based on monitoring data |
| CN113868207A (en) * | 2021-10-09 | 2021-12-31 | 北京中水科水电科技开发有限公司 | Compression display method and system for information redundancy jitter characteristics |
| CN115412422A (en) * | 2022-08-08 | 2022-11-29 | 浪潮云信息技术股份公司 | Dynamic window adjusting system |
| CN115412422B (en) * | 2022-08-08 | 2024-02-20 | 浪潮云信息技术股份公司 | Dynamic window adjusting system |
| CN117527523A (en) * | 2023-11-23 | 2024-02-06 | 广东堡塔安全技术有限公司 | Cloud computing-based server security monitoring system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111767195A (en) | Intelligent noise reduction processing method for alarm information | |
| US10877863B2 (en) | Automatic prediction system for server failure and method of automatically predicting server failure | |
| CN109597836B (en) | Communication equipment alarm association rule mining method based on weighting matrix | |
| CN111885012B (en) | Network situation perception method and system based on information acquisition of various network devices | |
| CN112073208B (en) | Alarm analysis method, device, chip system and storage medium | |
| CN108446184B (en) | Method and system for analyzing fault root cause | |
| CN109189736B (en) | Method and device for generating alarm association rule | |
| CN112118141B (en) | Communication network-oriented alarm event correlation compression method and device | |
| WO2021016978A1 (en) | Telecommunication network alarm prediction method and system | |
| CN111309565B (en) | Alarm processing method and device, electronic equipment and computer readable storage medium | |
| CN111212038A (en) | Open data API gateway system based on big data artificial intelligence | |
| TWI684139B (en) | System and method of learning-based prediction for anomalies within a base station | |
| CN110503247A (en) | Alarm of telecommunication network prediction technique and system | |
| CN110135603B (en) | Power network alarm space characteristic analysis method based on improved entropy weight method | |
| CN114511112A (en) | Intelligent operation and maintenance method and system based on Internet of things and readable storage medium | |
| CN113206834B (en) | Unknown protocol fuzzy test automation method based on reverse technology | |
| CN112328425A (en) | Anomaly detection method and system based on machine learning | |
| CN115514619A (en) | Alarm convergence method and system | |
| KR20210115991A (en) | Method and apparatus for detecting network anomaly using analyzing time-series data | |
| CN116436659A (en) | Quantitative analysis method and device for network security threat | |
| EP2882139B1 (en) | System and method for IT servers anomaly detection using incident consolidation | |
| CN110399278B (en) | Alarm fusion system and method based on data center anomaly monitoring | |
| CN110647086B (en) | Intelligent operation and maintenance monitoring system based on operation big data analysis | |
| CN107666403B (en) | Index data acquisition method and device | |
| CN116187423A (en) | Behavior sequence anomaly detection method and system based on unsupervised algorithm |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20201013 |
|
| RJ01 | Rejection of invention patent application after publication |