CN109213616A - A kind of micro services software systems method for detecting abnormality based on calling map analysis - Google Patents
A kind of micro services software systems method for detecting abnormality based on calling map analysis Download PDFInfo
- Publication number
- CN109213616A CN109213616A CN201811114173.0A CN201811114173A CN109213616A CN 109213616 A CN109213616 A CN 109213616A CN 201811114173 A CN201811114173 A CN 201811114173A CN 109213616 A CN109213616 A CN 109213616A
- Authority
- CN
- China
- Prior art keywords
- micro services
- node
- value
- feature
- abnormal
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/0703—Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation
- G06F11/0751—Error or fault detection not based on redundancy
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/0703—Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation
- G06F11/079—Root cause analysis, i.e. error or fault diagnosis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2201/00—Indexing scheme relating to error detection, to error correction, and to monitoring
- G06F2201/865—Monitoring of software
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Debugging And Monitoring (AREA)
Abstract
Invention is related to a kind of based on the micro services software systems method for detecting abnormality for calling map analysis.Each micro services feature is monitored, sliding window is established for each feature, is based on Single moving average method predicted characteristics value;The difference of comparison prediction value and monitor value detects off-note based on statistical method;The micro services of feature abnormalities are detected as exception, and the call graph of micro services is established according to the knowledge that system constructs;The intensity of anomaly of each micro services is assessed based on PageRank method and is ranked up.
Description
Technical field
The present invention relates to a kind of based on the micro services software systems method for detecting abnormality for calling map analysis, belongs to software technology
Field.
Background technique
Internet company's business is constantly expanded, and user demand frequently changes, and conventional software architectural has been unable to satisfy internet
Using the demand for design, exploitation, maintenance etc..Micro services are a kind of system architectures of high cohesion lower coupling, will be applied
One group of micro services that has a single function, can independently dispose is split into, each other by lightweight communication mechanism coordinated, is realized quick
Victory exploitation and deployment.But the component of micro services is numerous, dependence is complicated, increases abnormal odds and diagnosis
Difficulty.Have work and be normally based on monitoring metrics or log analysis, system manager combines domain knowledge investigation abnormal
Reason, but when in face of complicated micro services, then it can not quickly position abnormal cause.Particularly, when one of serviced component
It is abnormal, abnormal influences to spread as the mutual calling of inter-module is continuous, service quality decline or promise breaking are eventually led to,
Cause heavy losses.Therefore, it is the performance and reliability for ensureing micro services, effectively detects abnormal the problem of having become urgent need to resolve
One of.
Work at present can be divided into following a few classes, the method for detecting abnormality based on monitoring metrics, such as Hystrix (Latency
and Fault Tolerance for Distributed Systems. https://github.com/Netflix/
Hystrix), this method is realized simple, it is only necessary to easy configuration can monitoring objective system operating index, such as CPU, interior
It deposits, and alarm rule is set.When system occurs abnormal, then alarm is triggered, it is different that administrator in conjunction with domain knowledge carrys out analyzing and positioning
Normal reason.Method for monitoring and analyzing based on log, such as ELK (https: //www.elastic.co/), such methods are collected respectively
Log information in a component, administrator analyze abnormal initiation reason by information retrieval method or log mining.
What the monitoring method based on measurement or log was generally concerned with is the executive condition of single serviced component, and this mode is realized relatively
Simply, it but can not reflect that the integrality of micro services can not track Business Stream in complicated micro services interactive relation, manage
Reason person will take a substantial amount of time to search and position abnormal cause, sometimes or even cannot all diagnose and go wrong.Based on micro services
The analysis method for reliability of dependence, as App-Bisect (Rajagopalan S, Jamjoom H. App-Bisect:
autonomous healing for microservice-based apps, USENIX Conference on Hot
Topics in Cloud Computing, 2015.)、Gremlin (Heorhiadi V, Rajagopalan S,
Jamjoom H, Reiter M K, Sekar V. Gremlin: Systematic Resilience Testing of
Microservices. IEEE 36th International Conference on Distributed Computing
Systems. 2016. 57-66.).Such method is concerned with the operating status of micro services component-level, passes through Version Control
Or server resets guarantee the reliability of system.The rank that such method positions extremely is that micro services component equally can not
The state of system is portrayed from whole angle, when system occurs abnormal, it is still desirable to which administrator further checks.Separately
One kind is distributed method for tracing, as Dapper (Sigelman B H, Barroso L A, Burrows M,
Stephenson P, Plakal M, Beaver D, Jaspan S, Shanbhag C. Dapper: a large-scale
Distributed systems tracing infrastructure, Google Tecnical Report, 2010.), should
The execution information that method is called by forms monitoring methods such as log, pitching pile or notes, tracking request end to end is in system
In treatment process.Exception can cause processing track to be deviateed, by analysis processing track thus to achieve the purpose that diagnosis.
But this method focuses on the treatment process for tracking single request end to end, and and is not provided with the analysis method of effect.In addition,
Micro services software systems are facing generally towards huge user, cause monitoring data amount huge, so that administrator is submerged in monitoring data
In.
Summary of the invention
The purpose of the present invention: according to the Historical Monitoring data configuration statistical model of micro services feature, pass through predicted characteristics value
To detect off-note and micro services automatically, sorted micro services intensity of anomaly to position abnormal root based on anomalous propagation network
This reason.
The principle of the present invention: monitoring each micro services feature, establishes sliding window for each feature, based on primary mobile flat
Equal method predicted characteristics value;The difference of comparison prediction value and monitor value detects off-note based on statistical method;Feature abnormalities
Micro services are detected as exception, and the call graph of micro services is established according to the knowledge that system constructs;It is commented based on PageRank method
Estimate the intensity of anomaly of each micro services and is ranked up.
The technology of the present invention solution: a kind of micro services software systems method for detecting abnormality based on calling map analysis,
Feature is to realize that steps are as follows:
The first step, feature monitoring:
The sliding window that length is N is established for each feature, collects timing monitoring data (yt,yt-1,…,yt-N+1) be sequentially put into
Sliding window, wherein yiFor i moment monitoring data;
Second step, feature prediction:
For each feature, the value at (t+T) moment is predicted,
;
Third step, off-note detection:
,, wherein yt+T' it is spy
Levy the predicted value at (t+T) moment, yt+T' it is characterized monitor value at (t+T) moment, it is different if being detected as off-note
Micro services where Chang Tezheng are identified as abnormal micro services.
4th step, the building of micro services calling figure:
According to the call relation between the building knowledge or monitoring micro services for understanding micro services software systems, in the form of digraph
Call relation between the abnormal micro services detected is described, wherein each micro services are expressed as a node, and micro services A is called
Micro services B is expressed as the directed edge from node A to node B.
5th step, abnormal root is because of positioning:
When node A has link to be directed toward node B, be considered as node B and obtain score, the score value number depend on node
The importance of the significance level of A, i.e. node A is bigger, and the score that node B is obtained is higher.The calculating of the score value is an iteration
Process, finish node will be ranked up it according to resulting score and search result delivered user, this point quantified
Number is exactly PR value.
(1) initial phase: the initial of all node is is givenPR i (0)(i=1,2,3,…,N) value, wherein N is node
Quantity meets;
(2) the more new stage:, wherein pjIndicate the node for being directed toward node i,
L(pj) it is directed to the number of nodes of node i, q is random walk coefficient q ∈ (0,1);
(3) abnormal positioning: each node of recursive calculationPRValue, is ranked up the PR value of each node, the PR value of micro services is got over
Height, intensity of anomaly are bigger.
The invention has the following advantages over the prior art:
1. being detected automatically by predicted characteristics value abnormal special according to the Historical Monitoring data configuration statistical model of micro services feature
Sign, without using the feature of domain knowledge manual analysis micro services;
2. constructing the call graph of micro services, model the interactive relation between micro services, can portray it is abnormal micro services it
Between propagation;
3. using the abnormality score of each micro services of PageRank algorithm evaluation, so that anomalous propagation factor is considered, it is accurate fixed
The basic reason of position micro services software systems exception.
Detailed description of the invention
Fig. 1 is the use environment of present invention method.
Specific embodiment
Below in conjunction with specific embodiments and the drawings, the present invention is described in detail, as shown in Figure 1, embodiment of the present invention side
Method process:
Piggy Metrics (https: //github.com/sqshq/PiggyMetrics) is had chosen as experimental subjects.
Piggy Metrics is one and provides the micro services application of personal finance function, with reference to the design principle of micro services,
Realize that code trustship at present is in Github using micro services Development Framework Spring Boot and Spring Cloud.Piggy
In Metrics software architecture, API Gateway is directly interacted with Client, is received Client request, is realized load balancing
And routing forwarding, call multiple back-end services to complete request processing, final polymerization result returns to Client;Account
Service provides account management function, including account registration, inquiry etc.;Statistics Service realizes Personal Finance
Statistical function, including income, expenditure and remaining sum etc.;The contact details and notice of Notification Service storage user are set
It sets, periodically transmits the message to booking reader;Auth Service realizes the access authority of service based on OAuth2 Tokens
Control;For application configuration using Spring Cloud Config as centralized management, serviced component passes through Config when starting
Service obtains configuration parameter;Each serviced component can run multiple examples, which uses Netflix Eureka and make
For service discovery component, the operating status of dynamic sensing Service Instance;Led between serviced component by Restful API
Letter is cooperated using Ribbon as client load balanced device with service discovery framework Eureka, neatly control load plan
Slightly.In addition, the data that the application uses decentralization store, each service uses individual MongoDB storing data.
Experimental situation includes micro services application Piggy Metrics, workload generator, abnormal injector and abnormity diagnosis system
System.Wherein, Piggy Metrics carries out container volume by Docker Compose using Docker as basic running environment
Row;Workload generator is asked using testing tool Apache Jmeter (http://jmeter.apache.org/) analog subscriber
It asks, generates load;Abnormal injector will be injected into system, by preset script to test abnormity diagnostic system extremely
Diagnosis effect;Abnormity diagnostic system as designs the micro services abnormity diagnostic system with realization herein.Using Apache Jmeter
Analog subscriber request loads to generate.A test plan is created first, then creates one group of thread, setting for each scene
HTTP request parameter and result exhibition method, every group of Thread Count 100, cycle-index 100, scheduler duration are 90s, finally
Starting thread generates load to simulate.
The validity of proposed method is verified in such a way that injection is abnormal.It, successively will be abnormal single in experimentation
It solely is injected into system, abnormity diagnosis is then carried out using mentioned method herein.Each load duration curve is 90s, in 30s
Injection is abnormal, restores after continuing 30s, while the execution information of abnormity diagnostic system collection system, and carry out abnormity diagnosis.Specifically
Experimental procedure is as follows:
1) reset system: each exception is individually monitored and is diagnosed, influences each other in order to prevent, tests it every time
Afterwards, the cleaning of environment is carried out first, deletes junk data, then reinstalls the unloading of micro services application cluster.This process can
Easily to pass through container programming facility Docker Compose (https: //github.com/docker/compose)
It realizes.
2) start abnormity diagnostic system: starting abnormity diagnostic system, and debugged, check whether normal operation.
3) starting load generator: starting Apache Jmeter is arranged load configuration, initiates load.
4) injection is abnormal: for the exception that can restore automatically, such as pause container etc., injecting, holds after load continuous 30s
Cancel after continuous 30s abnormal;It for expendable exception, is then just injected when system starts, is continued until that load is completed.
Micro services abnormity diagnosis specifically includes the following steps:
1) distributed monitoring software Zabbix (https: //www.zabbix.com/) is used, the resource of collection vessel uses special
Sign, including CPU, memory, network utilization etc., setting sliding window size are 10, the monitoring data of collection in every 30 seconds, each
Feature establishes a sliding window;
2) using the formula of the second step in invention description, CPU, memory, network are predicted according to the monitoring data in sliding window
The characteristic value in the next stage of the features such as utilization rate;
3) using the formula of the third step in invention description, the variance of the features such as CPU, memory, network utilization is calculated, for every
The difference of a feature calculation monitor value and predicted value, if it is greater than 3 times of variances, then the micro services are detected as exception;
4) call relation between the abnormal micro services detected is described in the form of digraph, each micro services are expressed as a section
Point, micro services A call micro services B to be expressed as the directed edge from node A to node B;
5) using five step in invention description the step of, each micro services PR value is calculated, and sort from large to small to PR value,
The forward micro services that sort be abnormal root because.
Claims (1)
1. a kind of based on the micro services software systems method for detecting abnormality for calling map analysis, method characteristic is to realize step such as
Under:
The first step, feature monitoring: monitoring each micro services feature, establishes the sliding window that length is N for each feature, when collection
Sequence monitoring data (yt,yt-1,…,yt-N+1) it is sequentially put into sliding window, wherein yiFor i moment monitoring data;
Feature prediction: second step for each feature, predicts the value at (t+T) moment,
Wherein,;
Third step, off-note detection:,,
Wherein, yt+T' it is characterized predicted value at (t+T) moment, yt+T' it is characterized monitor value at (t+T) moment, if, then it is detected as off-note, the micro services where off-note are identified as abnormal micro services;
4th step, micro services calling figure building: according between the building knowledge or monitoring micro services for understanding micro services software systems
Call relation, the call relation between the abnormal micro services that detect is described in the form of digraph, wherein each micro services table
It is shown as a node, micro services A calls micro services B to be expressed as the directed edge from node A to node B;
5th step, abnormal root is because of positioning: giving the initial of all node isPR i (i=1,2,3,…,N) value, wherein N is number of nodes
Amount meets;The each node of recursive calculationp i 'sPRValue,,
Wherein, pjIndicate the node for being directed toward node i, L (pj) it is directed to the number of nodes of node i, q is random walk coefficient q ∈ (0,1);
The PR value of each node is ranked up, the PR value of micro services is higher, and intensity of anomaly is bigger.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811114173.0A CN109213616A (en) | 2018-09-25 | 2018-09-25 | A kind of micro services software systems method for detecting abnormality based on calling map analysis |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811114173.0A CN109213616A (en) | 2018-09-25 | 2018-09-25 | A kind of micro services software systems method for detecting abnormality based on calling map analysis |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109213616A true CN109213616A (en) | 2019-01-15 |
Family
ID=64984638
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811114173.0A Pending CN109213616A (en) | 2018-09-25 | 2018-09-25 | A kind of micro services software systems method for detecting abnormality based on calling map analysis |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109213616A (en) |
Cited By (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109933452A (en) * | 2019-03-22 | 2019-06-25 | 中国科学院软件研究所 | A kind of micro services intelligent monitoring method towards anomalous propagation |
CN110175085A (en) * | 2019-04-19 | 2019-08-27 | 同济大学 | A kind of Hadoop system abnormal cause diagnostic method and device using map analysis |
CN110365762A (en) * | 2019-07-10 | 2019-10-22 | 腾讯科技(深圳)有限公司 | Service processing method, device, equipment and storage medium |
CN110825548A (en) * | 2019-10-24 | 2020-02-21 | 新华三信息安全技术有限公司 | Anomaly detection method, model training method and related device |
CN110825589A (en) * | 2019-11-07 | 2020-02-21 | 字节跳动有限公司 | Anomaly detection method and device for micro-service system and electronic equipment |
CN110909015A (en) * | 2019-09-12 | 2020-03-24 | 华为技术有限公司 | Splitting method, device and equipment of microservice and storage medium |
CN111130880A (en) * | 2019-12-24 | 2020-05-08 | 杭州米络星科技(集团)有限公司 | Micro-service system exception mechanism transfer method |
CN111711607A (en) * | 2020-05-20 | 2020-09-25 | 西安理工大学 | Block chain-based flow type micro-service trusted loading and verifying method |
CN111737033A (en) * | 2020-05-26 | 2020-10-02 | 复旦大学 | Micro-service fault positioning method based on runtime map analysis |
CN111752641A (en) * | 2020-06-29 | 2020-10-09 | 深圳壹账通智能科技有限公司 | Method, device, equipment and storage medium for starting service between micro-service containers |
CN113014421A (en) * | 2021-02-08 | 2021-06-22 | 武汉大学 | Micro-service root cause positioning method for cloud native system |
CN114598742A (en) * | 2022-03-04 | 2022-06-07 | 北京北信源软件股份有限公司 | Micro-service importance determination method, device, electronic equipment and storage medium |
CN114780619A (en) * | 2022-06-07 | 2022-07-22 | 国网浙江省电力有限公司金华供电公司 | Abnormity early warning method for automatic engineering audit data |
US11500888B2 (en) | 2020-08-07 | 2022-11-15 | NEC Laboratories Europe GmbH | Methods and systems for detecting anomalies in cloud services based on mining time-evolving graphs |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105183659A (en) * | 2015-10-16 | 2015-12-23 | 上海通创信息技术有限公司 | Software system behavior anomaly detection method based on multi-level mode predication |
CN106330576A (en) * | 2016-11-18 | 2017-01-11 | 北京红马传媒文化发展有限公司 | Automatic scaling and migration scheduling method, system and device for containerization micro-service |
WO2018042002A1 (en) * | 2016-09-02 | 2018-03-08 | Telefonaktiebolaget Lm Ericsson (Publ) | Systems and methods of managing computational resources |
CN108173911A (en) * | 2017-12-18 | 2018-06-15 | 中国科学院声学研究所 | A kind of micro services fault detect processing method and processing device |
CN108400904A (en) * | 2018-01-26 | 2018-08-14 | 北京奇艺世纪科技有限公司 | A kind of health examination method and device based on micro services framework |
CN108512689A (en) * | 2017-12-15 | 2018-09-07 | 中国平安财产保险股份有限公司 | Micro services business monitoring method and server |
-
2018
- 2018-09-25 CN CN201811114173.0A patent/CN109213616A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105183659A (en) * | 2015-10-16 | 2015-12-23 | 上海通创信息技术有限公司 | Software system behavior anomaly detection method based on multi-level mode predication |
WO2018042002A1 (en) * | 2016-09-02 | 2018-03-08 | Telefonaktiebolaget Lm Ericsson (Publ) | Systems and methods of managing computational resources |
CN106330576A (en) * | 2016-11-18 | 2017-01-11 | 北京红马传媒文化发展有限公司 | Automatic scaling and migration scheduling method, system and device for containerization micro-service |
CN108512689A (en) * | 2017-12-15 | 2018-09-07 | 中国平安财产保险股份有限公司 | Micro services business monitoring method and server |
CN108173911A (en) * | 2017-12-18 | 2018-06-15 | 中国科学院声学研究所 | A kind of micro services fault detect processing method and processing device |
CN108400904A (en) * | 2018-01-26 | 2018-08-14 | 北京奇艺世纪科技有限公司 | A kind of health examination method and device based on micro services framework |
Cited By (21)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109933452B (en) * | 2019-03-22 | 2020-06-19 | 中国科学院软件研究所 | Micro-service intelligent monitoring method facing abnormal propagation |
CN109933452A (en) * | 2019-03-22 | 2019-06-25 | 中国科学院软件研究所 | A kind of micro services intelligent monitoring method towards anomalous propagation |
CN110175085A (en) * | 2019-04-19 | 2019-08-27 | 同济大学 | A kind of Hadoop system abnormal cause diagnostic method and device using map analysis |
CN110175085B (en) * | 2019-04-19 | 2020-11-27 | 同济大学 | Hadoop system abnormity reason diagnosis method and device by utilizing graph analysis |
CN110365762A (en) * | 2019-07-10 | 2019-10-22 | 腾讯科技(深圳)有限公司 | Service processing method, device, equipment and storage medium |
CN110909015B (en) * | 2019-09-12 | 2022-07-22 | 华为云计算技术有限公司 | Splitting method, device and equipment of microservice and storage medium |
CN110909015A (en) * | 2019-09-12 | 2020-03-24 | 华为技术有限公司 | Splitting method, device and equipment of microservice and storage medium |
CN110825548A (en) * | 2019-10-24 | 2020-02-21 | 新华三信息安全技术有限公司 | Anomaly detection method, model training method and related device |
CN110825548B (en) * | 2019-10-24 | 2022-11-08 | 新华三信息安全技术有限公司 | Anomaly detection method, model training method and related device |
CN110825589B (en) * | 2019-11-07 | 2024-01-05 | 字节跳动有限公司 | Abnormality detection method and device for micro-service system and electronic equipment |
CN110825589A (en) * | 2019-11-07 | 2020-02-21 | 字节跳动有限公司 | Anomaly detection method and device for micro-service system and electronic equipment |
CN111130880A (en) * | 2019-12-24 | 2020-05-08 | 杭州米络星科技(集团)有限公司 | Micro-service system exception mechanism transfer method |
CN111711607A (en) * | 2020-05-20 | 2020-09-25 | 西安理工大学 | Block chain-based flow type micro-service trusted loading and verifying method |
CN111711607B (en) * | 2020-05-20 | 2022-04-22 | 西安理工大学 | Block chain-based flow type micro-service trusted loading and verifying method |
CN111737033A (en) * | 2020-05-26 | 2020-10-02 | 复旦大学 | Micro-service fault positioning method based on runtime map analysis |
CN111737033B (en) * | 2020-05-26 | 2024-03-08 | 复旦大学 | Microservice fault positioning method based on runtime pattern analysis |
CN111752641A (en) * | 2020-06-29 | 2020-10-09 | 深圳壹账通智能科技有限公司 | Method, device, equipment and storage medium for starting service between micro-service containers |
US11500888B2 (en) | 2020-08-07 | 2022-11-15 | NEC Laboratories Europe GmbH | Methods and systems for detecting anomalies in cloud services based on mining time-evolving graphs |
CN113014421A (en) * | 2021-02-08 | 2021-06-22 | 武汉大学 | Micro-service root cause positioning method for cloud native system |
CN114598742A (en) * | 2022-03-04 | 2022-06-07 | 北京北信源软件股份有限公司 | Micro-service importance determination method, device, electronic equipment and storage medium |
CN114780619A (en) * | 2022-06-07 | 2022-07-22 | 国网浙江省电力有限公司金华供电公司 | Abnormity early warning method for automatic engineering audit data |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109213616A (en) | A kind of micro services software systems method for detecting abnormality based on calling map analysis | |
US10187328B2 (en) | Anomaly detection and classification using telemetry data | |
US9672085B2 (en) | Adaptive fault diagnosis | |
Yang et al. | A time efficient approach for detecting errors in big sensor data on cloud | |
CN109933452B (en) | Micro-service intelligent monitoring method facing abnormal propagation | |
Sauvanaud et al. | Anomaly detection and diagnosis for cloud services: Practical experiments and lessons learned | |
US20220330050A1 (en) | Proactive customer care in a communication system | |
US20220078188A1 (en) | Change Monitoring and Detection for a Cloud Computing Environment | |
Cotroneo et al. | A fault correlation approach to detect performance anomalies in virtual network function chains | |
US20160055044A1 (en) | Fault analysis method, fault analysis system, and storage medium | |
Bendriss et al. | AI for SLA management in programmable networks | |
US8170965B2 (en) | Method for governing the operation of a generalist agent within a complex multi-agent adaptive system | |
Raj et al. | Cloud infrastructure fault monitoring and prediction system using LSTM based predictive maintenance | |
Wang et al. | A methodology for root-cause analysis in component based systems | |
Wang et al. | Online reliability time series prediction for service-oriented system of systems | |
Gan et al. | Sage: Using unsupervised learning for scalable performance debugging in microservices | |
Shi et al. | Towards automatic troubleshooting for user-level performance degradation in cellular services | |
AU2014200806B1 (en) | Adaptive fault diagnosis | |
CN115150289A (en) | Exception handling method and system based on composite monitoring | |
Murphy et al. | Fault prediction for heterogeneous telecommunication networks using machine learning: a survey | |
Ciccotelli et al. | Nirvana: A non-intrusive black-box monitoring framework for rack-level fault detection | |
Murphy et al. | Fault Prediction for Heterogeneous Networks using Machine Learning: a Survey | |
Ahirwar et al. | Anomaly detection in the services provided by multi cloud architectures: a survey | |
Matsuo et al. | DeepSIP: A system for predicting service impact of network failure by temporal multimodal CNN | |
Bendechache et al. | Analysing dependability and performance of a real-world Elastic Search application |
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 | ||
WD01 | Invention patent application deemed withdrawn after publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190115 |