KR102418594B1 - Ict equipment management system and method there of - Google Patents

Abstract

The present invention is a distributed data storage unit for storing device information, an operation log collection unit for storing operation data related to failure and performance of the device, an operation quality learning unit for learning the operation data using machine learning logic, the learning It includes an operation quality prediction unit that calculates the severity and the cause of quality deterioration through , and a quality deterioration equipment extraction unit including a quality deterioration profile management block that stores the inquiry condition of the quality deterioration equipment to be replaced.
The present invention automatically detects signs of performance degradation in advance based on ICT operation information that appears as a mass event, and extracts equipment that may cause problems, thereby blocking factors that cause performance degradation in advance to provide a continuous and stable service to be able to provide

Description

ICT EQUIPMENT MANAGEMENT SYSTEM AND METHOD THERE OF

The present invention relates to a management system and method for ICT equipment, and more specifically, for the stable operation of ICT equipment such as a network, server, and power equipment for providing internal and external services in a network operator or company operating a large-scale network, deterioration of quality It relates to a system and method that provides a service to replace equipment at an optimal time by extracting equipment that may have an impact on the system in advance.

Currently, most companies, including telecommunication service providers that sell networks, use various ICT (Information & Communication Technology) equipment such as leased lines, servers, switches, routers, UPS power supplies, and various controller equipment.

Although some companies lease these equipments, most companies are still building their own equipment through purchase. Therefore, in running a company, ICT purchase and operation costs a lot of money, so companies are researching how to purchase an appropriate size at the right time and operate it stably and efficiently.

The conventional operation management system is a system for monitoring network equipment and lines by having a management function of the FCAPS (Fault, Configuration, Account, Performance, Security) domain for operation and management of networks and servers. and failure management unit to monitor and monitor failures, configuration management unit to monitor physical/logical configuration and operation status information of equipment or lines, collection/storage/control account management unit for billing information, performance/traffic/ Performance management department in charge of quality monitoring and statistical management, security management department monitoring security/safety/confidentiality management, etc. Protocols include standard protocols such as SNMP (Simple Network Management), ICMP, Netconf, Netflow, RMON, etc. and own protocols It consisted of the use of

All of these conventional systems are methods for follow-up, and there was an aspect that could not properly cope with anomalies or pre-measurements through pre-detection. In addition, the method of detecting anomalies by TCA (Threshold Crossing Alert) for traffic utilization, CPU, and memory utilization is a method to detect when the load by a specific value is out of the normal range by turning it into an event. There was a problem that the accuracy was significantly lowered because there were always too many false positives because there could not be a threshold value that worked exactly when there was an abnormality in the equipment that appeared.

Under the conventional operation management system, it is difficult to predict failures and quality deterioration in advance, and the causes thereof cannot be eliminated in advance, and consequently, the stability of the entire service cannot be guaranteed.

In other words, in the conventional resource management system, the contents related to the financial resources of equipment can be managed, but by detecting the deterioration of equipment in advance, the equipment that causes continuous quality deterioration among numerous equipment is extracted in advance to upgrade or replace equipment. You cannot prepare an object.

The present invention is to solve the above problems, and in advance, the management of ICT equipment that extracts equipment to be upgraded and replaced so that a network operator or company operating a large-scale network can stably manage ICT equipment To provide a system and method.

In order to solve the above technical problem, the present invention discloses a quality degradation equipment management apparatus, a distributed data storage unit for storing device information, an operation log collection unit for storing operation data related to failure and performance of the device, and the operation data An operation quality learning unit that learns using machine learning logic, an operation quality prediction unit that calculates the severity and the cause of quality deterioration through the learning, and a quality that stores the inquiry condition of the quality deterioration equipment to be replaced and a degradation equipment extraction unit including a degradation profile management block.

In addition, the operation data stored in the operation log collection unit is characterized in that it includes a device failure event, a threshold value crossing warning (Thresholds Crossing Alert, TCA) event, and system log information.

In addition, the operation quality learning unit is characterized in that it includes a machine learning processing block for performing machine learning, a preprocessing function block for converting a data type so that the machine learning processing block can learn.

In addition, the inquiry condition of the degraded equipment includes equipment introduction date, quality event threshold, failure event threshold, performance event threshold, operational Voice of Customer (VoC) event threshold, SYSLOG event threshold, severity, and service level. It is characterized in that it is a condition including any one or more selected from.

In addition, the quality impact analysis block that evaluates the failure and quality correlation of the degraded equipment and related equipment, the performance prediction block that predicts the resource use of the degraded equipment, and the recommended replacement equipment It is characterized in that it further comprises an ICT design module including a substitute simulator block for calculating the quality impact.

In addition, the quality impact analysis block predicts the resource usage status of the equipment, extracts a recommended model of the replacement target equipment, and performs a process of predicting the replacement time.

In addition, it is characterized in that the extraction of the recommended collection of the replacement target equipment is displayed in a graphical user interface (GUI).

According to an embodiment of the present invention, based on ICT operation information that appears as a mass event, automatically detects signs of performance degradation in advance and extracts equipment that may cause problems, thereby blocking factors that cause performance degradation in advance This enables us to provide a continuous and stable service.

In addition, by automatically identifying and providing equipment that is a factor in failure and quality deterioration, it is possible to determine the time and scale of replacement, and through this, it is possible to continuously provide stable service.

In addition, it is possible to prevent obstacles related to ICT services in advance, reduce related network operation costs, and improve work efficiency. This can be helpful not only to companies operating large-scale network-based ICT equipment, but also to network operators who provide services to them and perform operation on behalf of them.

However, the effects that can be achieved by the management system and method of ICT equipment according to the embodiments of the present invention are not limited to those mentioned above, and other effects not mentioned are from the description below in the technical field to which the present invention belongs It will be clearly understood by those of ordinary skill in the art.

BRIEF DESCRIPTION OF THE DRAWINGS The accompanying drawings, which are included as a part of the detailed description to help the understanding of the present invention, provide embodiments of the present invention, and together with the detailed description, explain the technical spirit of the present invention.
1 is an overall configuration diagram of a management system for ICT equipment according to the present invention.
2 is a block diagram illustrating a detailed configuration of the intelligent operating environment prediction module 1000 according to the present invention.
3 is a block diagram illustrating a detailed configuration of the operation log collection unit 200 according to the present invention.
4 is a block diagram illustrating a detailed configuration of the operation quality learning unit 300 according to the present invention.
5 is a block diagram illustrating a detailed configuration of the operation quality prediction unit 400 according to the present invention.
6 is a block diagram showing a detailed configuration of the quality deterioration equipment extraction unit 500 according to the present invention.
7 is a block diagram showing a detailed configuration of the ICT design module 2000 according to the present invention.
8 is a flowchart illustrating a method for automatically extracting a replacement target device according to the present invention.

The present invention can apply various transformations and can have various embodiments. Hereinafter, specific embodiments will be described in detail based on the accompanying drawings.

The following examples are provided to provide a comprehensive understanding of the methods, apparatus and/or systems described herein. However, this is merely an example, and the present invention is not limited thereto.

In describing the embodiments of the present invention, if it is determined that the detailed description of the known technology related to the present invention may unnecessarily obscure the gist of the present invention, the detailed description thereof will be omitted. And, the terms to be described later are terms defined in consideration of functions in the present invention, which may vary according to intentions or customs of users and operators. Therefore, the definition should be made based on the content throughout this specification. The terminology used in the detailed description is for the purpose of describing embodiments of the present invention only, and should not be limiting in any way. Unless explicitly used otherwise, expressions in the singular include the meaning of the plural. In this description, expressions such as “comprising” or “comprising” are intended to indicate certain features, numbers, steps, acts, elements, some or a combination thereof, one or more other than those described. It should not be construed to exclude the presence or possibility of other features, numbers, steps, acts, elements, or any part or combination thereof.

In addition, terms such as first, second, etc. may be used to describe various components, but the components are not limited by the terms, and the terms are for the purpose of distinguishing one component from other components. is used only as

Hereinafter, exemplary embodiments of an ICT equipment management system and method according to the present invention will be described in detail with reference to the accompanying drawings.

1 is an overall configuration diagram of a management system for ICT equipment according to the present invention.

The present invention generates statistical information in conjunction with an operation management system (or network management system) (Network Management System, NMS) in a corporate ICT environment, detects the state of the operation environment by using a machine learning algorithm, and each equipment It is related to a system and method that can maintain service quality by predicting the severity and cause of service quality, and automatically extracting and providing equipment that has deteriorated in quality so that it can be upgraded and replaced in advance.

The ICT equipment management system according to the present invention includes an intelligent operating environment prediction module 1000 , an ICT design module 2000 , and an operation management system (NMS) 3000 .

The operation management system (NMS) 3000 monitors the configuration, failure, performance, and traffic of ICT equipment. The intelligent operating environment prediction module 1000 predicts and extracts performance degradation equipment based on the operation information monitored by the operation management system 3000 . The ICT design module 2000 recommends upgrade or replacement through the prediction information predicted by the intelligent operating environment prediction module 1000 . On the other hand, the demonstration of the upgraded or replacement equipment may be demonstrated (or displayed) by a GUI (Graphical User Interface).

The present invention may provide an intelligent operating environment support system through the intelligent operating environment prediction module 1000 interworking with the operation management system (NMS) 3000 . The operating environment resource system is a system that can perform upgrade or replacement after automatically extracting equipment causing quality deterioration, and can prevent failure and quality deterioration in advance.

2 is a block diagram illustrating a detailed configuration of the intelligent operating environment prediction module 1000 according to the present invention.

The intelligent operating environment prediction module 1000 includes a distributed data storage 100, an operation log collection unit 200, an operation quality learning unit 300, an operation quality prediction unit 400, and a quality deterioration equipment extraction unit 500 do.

The distributed data storage 100 serves to store data based on a database or Hadoop Distributed File System. The distributed data storage 100 requires batchability (a method of collecting and processing a certain amount of data at once) or performance capable of processing large amounts of data in real time, so a design is required in consideration of this.

The distributed data storage 100 stores inventory information by equipment, failure information by equipment, performance information by equipment, system log information by equipment, learning information, operation quality prediction information, and the like.

Inventory information for each equipment is facility information for all equipment operated in conjunction with each NMS. The facility information may be information such as a unique ID of the equipment, a date of introduction of the equipment, a place where the equipment is installed, a manager, a contact information, a manufacturer, an IOS version, and the like.

The equipment-specific failure information may be periodically collected failure information in conjunction with each NMS.

The performance information for each device may be performance information periodically collected in conjunction with each NMS.

The system log information may be system log (syslog) information of a device in which a failure occurs by interworking with each NMS.

The learning information may be learning result information analyzed through machine learning in the operation quality learning unit 300 .

The operation quality prediction result information may be result information determined by the system periodically analyzed by the operation quality prediction unit 400 .

Additionally, conditions necessary for learning to be used in the operation quality learning module can be stored. The operation log collection unit 200 , the operation quality learning unit 300 , the operation quality prediction unit 400 , and the quality deterioration equipment extraction unit 500 will be described in detail below with reference to the drawings.

3 is a block diagram illustrating a detailed configuration of the operation log collection unit 200 according to the present invention.

The operation log collection unit 200 collects and buffers data related to numerous failures and performance degradation that the administrator cannot recognize from devices and various management systems, processes them, and stores them in a database.

Operation log collection unit 200 is interlocking block 210, failure event management block 220, performance TCA management block 230, SYS log management block 240, batch data processing 250, statistics management block 260 ) is included.

The interworking block 210 is responsible for interworking with operation management systems such as IPNMS, transmission NMS, and server management system.

Interworking IP devices can be routers, L2/L3/L4 switches, and security devices. Transmission devices are Multi-Service Provisional Platform (MSPP), PTN (Packet Transport Network) equipment, and optical transmission network (Optical). Transport Network, OTN) equipment.

In addition, the server and power may be an authentication server, a billing server, a web server, a cache server, DHCP (Dynamic Host Configuration Protocol), DNS (Domain Name System), UTM (Unified Threat Management), and the like.

The failure event management block 220 collects failure events for each analysis period in association with each NMS. The failure event may be a port down in which a physical port is disconnected or a link down in which the link between links is disconnected.

The performance TCA (Thresholds Crossing Alert) management block 230 collects performance TCA events for each analysis period in association with each NMS. TCA events can be counted due to CPU overload, excessive memory usage, disk exhaustion, or excessive major process exceptions.

The SYS log management block 240 interworks with each NMS to collect system log (syslog) information on equipment in which a failure event has occurred for each analysis cycle.

The batch data processing block 250 provides a specific condition and aims to inquire the result of machine learning for this. The batch data processing block 250 includes a function processing unit for learning data and storing a result thereof.

The statistics management block 260 is a system function block and a function block for generating statistical data for an operator inquiry. The statistics management block 260 generates statistical data for each day, week, or month.

The value stored by the operation log collection unit 200 is the time taken to collect the log, ID of each ICT equipment, equipment type, number of failures (count), performance TCA number (count), failure cause code, quality TCA This can be the cause, the number of SYSLOG events, or the SYSLOG value.

4 is a block diagram illustrating a detailed configuration of the operation quality learning unit 300 according to the present invention.

The operation quality learning unit 300 includes a preprocessing function block 310 , a machine learning processing block 320 , and a machine learning environment management block 330 .

The pre-processing function block 310 performs a function of converting a data type based on the collected operational data. Specifically, it is a block that converts the type of data so that it can be used in the machine learning processing block 320 and transmits the data.

As an embodiment, for machine learning, a data format to be stored and processed may be a format as shown in Table 1 below.

[Table 1]

Here, the quality event value is the sum of the number of failure events, the number of performance TCA events, and the number of SYSLOG events.

The 　 value corresponding to the quality degradation cause code 　 is the result obtained by analyzing the 　 data format.

The service severity value is a value that defines the influence of one set of quality events on the service as 1 to 5 levels. As an embodiment, the severity value may be set in a range of a minimum of 1 to a maximum of 5, where 1 may mean Monitoring, 2 may be Warning, 3 may be Minor, 4 may be Major, and 5 may be Critical.

The machine learning processing block 320 is a module for learning data by having machine learning logic such as Convolutional Neural Networks (CNN) and Recurrent Neural Network (RNN). Machine learning logic can be used with any currently used algorithm.

The machine learning processing block 320 is a functional module that actually learns data through machine learning so that the severity can be predicted and automatically processed using the values refined in the preprocessing function block 310. have.

Learning is performed by reading the conditions necessary for learning from the data storage. In this case, a machine learning algorithm (CNN, RNN, epoch) used for learning may be used, etc. The learning rate may be based on 70% of learning data, 30% of test data, etc., and the ratio may be adjusted .

The machine learning environment management block 330 is a function processing unit that communicates with the machine learning processing unit 320 with a result value learned in advance based on real-time operation data, and manages the input and the result value.

The machine learning environment management block 330 receives data periodically from the data collection processing unit operated based on the learned data, automatically processes the severity and stores the value.

In addition, a learning profile that stores the operating environment of the learning algorithm in operation and the definition of the learning data along with the operating time, so that when new machine learning operates, the information is loaded and used or can be referred to when analyzing the result data It performs management functions.

5 is a block diagram illustrating a detailed configuration of the operation quality prediction unit 400 according to the present invention.

The operation quality prediction unit 400 is a module for generating data for regular operation with data stored by the operation log collection unit 200 . The operation quality prediction unit 400 includes a preprocessing function block 410 and a machine learning environment management block 420 .

Reads the operation log data stored periodically (eg, once a day, at 00:00) and processes it in the format shown in Table 1 above through the preprocessing function block 410, and in the operation quality learning unit 300 According to the rules of the algorithm made by machine learning, the severity and the cause of quality deterioration are obtained and stored. The most relevant value based on the similarity is automatically generated in the system by an artificial intelligence algorithm.

In this case, the machine learning operation applies the profile generated as a result of learning in the machine learning processing block 320 of the operation quality learning unit 300 as it is.

The operation quality prediction unit 400 performs the same functional operation as the operation quality learning unit 300 except for the machine learning environment management block 330 . However, if the operation quality learning unit 300 previously used data to make a rule through learning, the operation quality prediction unit 400 performs a role of processing the data actually operated.

6 is a block diagram showing a detailed configuration of the quality deterioration equipment extraction unit 500 according to the present invention.

The quality deterioration equipment extraction unit 500 performs a function of automatically extracting replacement target equipment, which is a main operation of the system. The quality deterioration equipment extraction unit 500 includes a quality deterioration profile management block 510 and a quality deterioration equipment extraction function block 520 .

The quality degradation profile management block 510 is a functional block for storing and managing inquiry conditions of the quality degradation equipment, which is a replacement target.

For example, among the server equipment of a company with a medium network size, among the equipment that is at least three years old, it is a server that provides major services with service class 3, and equipment that has been introduced for five years with 1000 users / 50 concurrent users or more There is a serious problem in operation for equipment that has 200 failures per month, 500 performance TCAs, 2000 related Syslogs, 20 operational VoCs and 100 Severity 3 or higher quality degradation alarms. It can be evaluated and extracted that the system needs to be upgraded or replaced.

In this case, the relevant profile information serving as a reference may be as shown in Table 2 below.

[Table 2]

The above profile items can be used in various combinations, such as applying all at once or by applying each column. The application to the whole can be considered as the default application, and it is also possible to operate the function by separately setting the inquiry and application threshold items for an arbitrary period on the operation GUI screen.

Equipment introduction date refers to the date of introduction of equipment that is recognized as obsolete equipment. In case of inquiry, calculate 3 or 5 years from the current date and input it as a date.

The quality event daily threshold means a default threshold for the number of quality events per day.

The failure event threshold means the failure event threshold that occurred during the corresponding period in the equipment-specific operation data collected by cycle (eg, default 1 hour).

The performance event threshold means the threshold for the performance event (CPU, MM, DISK utilization rate, etc.) of the equipment for each collection period that is judged to be a quality deterioration equipment.

Operational VoC (Voice of Customer) number threshold means the threshold for customer complaints received by the equipment.

The SYSLOG event threshold means the threshold for the count of “syslogs determined in advance to be quality-related items” in SYSLOG data for each equipment that is interlocked and collected by cycle.

The severity of the applied quality is the failure and performance of each equipment. It means a severity threshold that is a standard in raw data of operational statistics information such as SYSLOG.

The service class means the threshold of the service class related to the equipment. By setting this, the equipment can exclude equipment with low service class, such as test/standby equipment, or a small number of users.

The quality deterioration continuity threshold means a threshold value for the number of times per day (eg, 24 times a day) of quality deterioration determined by the quality deterioration equipment.

The quality deterioration equipment extraction function block 520 is a module that performs a function of representing the result value as a probability based on data stored in the learning process through prediction of the equipment deterioration quality by machine learning.

For the equipment presented in Table 1, the result can be derived as (2015-05-DaeJeon-SVR100, 90%). For example, the user selects the equipment extraction standard i) Severity 3 or higher, ii) the service Grade 3 or higher, iii) Quality degradation continuity threshold of 20 or higher, iv) Failure event threshold of 10 or higher When the same criteria are set, if the equipment satisfies 90% of the above setting criteria, it is calculated as 90%, and if all criteria are satisfied, it is calculated as 100 It is calculated in %.

In the quality deterioration equipment extraction function block 520, a list of equipment to be upgraded and replaced is inquired from the database (DB), and related equipment is extracted based on the “quality deterioration profile”.

When searching by month/year/specific period, the searched result may be as follows.

1. Number of equipment with poor service quality / Total number of equipment

2. Inquiry period, equipment name, equipment type, severity, equipment introduction date, number of quality deterioration events during period, cumulative number of events, number of registrations subject to annual replacement, cause of quality deterioration, date of introduction, etc. can be inquired.

Additionally, the number of ports of the device, the service provided, the service importance, the number of clients, and the continuous count of quality degradation may be inquired.

The inquired result is determined to be equipment that has an effect on service quality degradation due to periodic failures and performance problems, and is automatically extracted as outdated equipment. This is determined in consideration of the equipment utilization rate and the importance of the service.

7 is a block diagram showing a detailed configuration of the ICT design module 2000 according to the present invention.

The ICT design module 2000 is a block for determining when and with which equipment to replace the extracted quality-degrading equipment. That is, the ICT design module 2000 is a module that recommends the replacement timing and size of equipment, and includes a quality impact analysis block 600 , a performance prediction block 700 , a replacement simulator block 800 , and an upgrade simulator block ( 900).

The quality impact analysis block 600 calculates the quality correlation with the failure of the related equipment for the quality degradation equipment (generally target equipment) automatically extracted from the intelligent operating environment prediction unit. Equipment with a high quality correlation is equipment that affects a large number of equipment, and it is judged that the quality deterioration of such equipment can have a serious impact on network operation and service provision.

In addition, the quality impact analysis block 600 predicts the resource usage status of the corresponding equipment, predicts the time when the corresponding resource is completely depleted through a machine learning algorithm, and extracts the equipment that is a general recommendation model.

Equipment with high performance TCA has no problem with the current service itself, but if this phenomenon continues, it implies that service failure may occur in the short and long term. Overload and resource problems in these performance index items may lead to quality degradation, and subsequent depletion of the corresponding resource may lead to failure.

The quality impact analysis block 600 predicts the resource usage status of these items, predicts the time when the corresponding resource is completely depleted with a machine learning algorithm, and can designate the time of equipment replacement.

On the other hand, when extracting replacement recommended equipment for equipment with reduced quality, the model and service provided (web, DB, daemon server, development server, etc.) Telecommunication services, manufacturing, ICT, food, textiles, wood, forestry, automobiles, steel, etc.), and the number of customers using the service (total and simultaneous use) are taken into account.

The performance prediction block 700 is a block for performing prediction on resource usage trends, such as CPU, MM, DISK, etc. of the quality degradation equipment which is the target of the general replacement.

The replacement simulator block 800 is a block for calculating the expected quality impact when resource performance is supplemented with the recommended replacement equipment.

The upgrade simulator block 900 is a block that calculates an expected quality impact when the resource performance is supplemented by performing most of the recommended upgrade equipment.

8 is a flowchart illustrating a method for automatically extracting a replacement target device according to the present invention.

Step S10 is a step in which the manager sets and updates the learning profile of all ICT equipment in the operation quality learning unit 300 if necessary. It determines the learning-related conditions used in the operational quality learning module, that is, the process of setting or updating various parameters, for example, the machine learning learning cycle, related data capacity, and the ratio of learning and verification.

Step S20 is a step of setting and updating an operation profile, and the administrator sets or updates various profiles used in the operation log collection unit 200 if necessary. For example, it is possible to set or update the period for interworking data in each NMS, the time and period for batch work that generates statistical data, and the like.

Step S30 is a step in which the administrator sets or updates various profiles used in the quality deterioration equipment extraction unit 500 .

Step S40 is a step of collecting failure, performance, and syslog data, and the operation log collection unit 200 collects various operation logs by interworking with each NMS.

Step S50 is a step of loading a learning profile, and the set value related to learning used in the operation quality learning unit 300 in the system is read from the database and loaded into the memory.

Step S51 is a step of loading and updating the quality deterioration equipment extraction profile, and the system reads the profile used by the quality deterioration equipment extraction unit 500 from the database and loads it into the memory.

Step S60 is a step in which the operation log collection unit 200 temporarily stores data obtained by interworking from each NMS 3000 in the distributed data storage unit 100 such as Hadoop.

Step S70 is a step of reading the learning data from the distributed data storage unit 100 according to a cycle set in the operation quality learning unit 300 .

Step S80 is a step of loading the learned data, and is a process of loading the learning data read from the distributed data storage unit 100 into the machine learning algorithm execution process.

Step S90 is a step of pre-processing the collected data.

Step S100 is a step of transferring the learning result of machine learning to the operation quality prediction unit 400 . The operation quality prediction unit 400 can be said to be a machine learning operation model that applies various parameters and machine learning algorithms delivered as a learning result as it is.

Step S110 is a step of setting the operation quality prediction module, and the result of learning is reflected in the operation quality prediction unit 400 .

Step S120 is a step of reading data from the distributed data storage unit 100 for data required for operation quality prediction in the operation quality prediction unit 400 . That is, the necessary data is periodically loaded according to the operation profile.

Step S130 is a step of performing pre-processing of the data read in step S120. The read data is pre-processed in the same way as the "operation quality learning unit 300".

Step S140 is a step of storing the result in the distributed data storage unit 100 again after the operation quality prediction process is performed.

Step S150 is a step of generating various statistical data according to set conditions. Statistics can be created by hour/day/week/month.

Step S160 is a step of periodically loading data used by the quality deterioration equipment extraction unit 500 according to the operation profile.

Step S170 is a step of extracting the quality deterioration equipment by operating the quality deterioration equipment extraction algorithm.

Step S180 is a step of storing the extraction result.

Step S190 is a step of inquiring a list of quality deterioration equipment and detailed details according to the inquiry condition.

The above description is merely illustrative of the technical spirit of the present invention, and various modifications and variations will be possible without departing from the essential characteristics of the present invention by those skilled in the art to which the present invention pertains. Accordingly, the embodiments described in the present invention are not intended to limit the technical spirit of the present invention, but to explain, and are not limited to these embodiments. The protection scope of the present invention should be interpreted by the following claims, and all technical ideas within the scope equivalent thereto should be construed as being included in the scope of the present invention.

1000: intelligent operating environment prediction module
100: distributed data storage unit
200: operation log collection unit
210: interlocking block 220: failure event management block
230: performance TCA management block 240: SYS log management block
250: batch data processing block
260: statistics management block
300: operation quality learning unit
310; preprocessing function block 320: machine learning processing block
330: machine learning environment management block
400: operation quality prediction unit
410: preprocessing function block 420: machine learning processing block
500: quality deterioration equipment extraction unit
2000: ICT design module
600: quality impact analysis module
700: performance prediction module
800: Most Simulator Module
900 : Upgrade Simulator Module
3000: operation management system (NMS)

Claims (7)

Distributed data storage for storing device information;
an operation log collection unit that collects operation data related to device failure and performance;
an operation quality learning unit for learning the operation data based on machine learning logic;
an operation quality prediction unit for calculating operation quality prediction information including a severity through the learning; and
A quality degradation profile management block for managing a quality degradation profile including a query condition of a quality degradation device to be replaced, and a quality degradation device for detecting the quality degradation equipment based on the operation quality prediction information and the quality degradation profile Degraded equipment management device comprising a; quality deterioration equipment extraction unit including an extraction function block.
According to claim 1,
Operation data stored in the operation log collection unit is a device failure event, a threshold value crossing warning (Thresholds Crossing Alert, TCA) event and system log information, characterized in that it includes a quality degradation equipment management apparatus.
According to claim 1,
The operational quality learning unit, a machine learning processing block for performing machine learning, a quality degradation equipment management device, characterized in that it comprises a pre-processing function block for converting a data type so that the machine learning processing block can learn.
According to claim 1,
The inquiry condition of the degraded equipment is the equipment introduction date, quality event threshold, failure event threshold, performance event threshold, operational Voice of Customer (VoC) event threshold, system log (SYSLOG) event threshold, severity, and service level. Equipment management device for quality degradation, characterized in that the condition including any one or more selected.
According to claim 1,
a quality impact analysis block for evaluating the quality degradation equipment and related equipment failure and quality correlation;
a performance prediction block for predicting resource use of the quality degradation equipment; and
a replacement simulator block that calculates the expected quality impact when supplementing with the recommended replacement equipment;
Quality degradation equipment management device, characterized in that it further comprises an ICT design module comprising a.
6. The method of claim 5,
The quality impact analysis block predicts the resource usage status of the equipment, extracts a recommended model of the replacement target equipment, and performs the process of predicting the replacement time.
7. The method of claim 6,
The information on the recommended model of the replacement target equipment is displayed as a graphical user interface (GUI).

Images