A kind of telecommunications network service performance monitoring system
Technical field
The present invention relates to a kind of system that monitors telecommunications network service performance.
Background technology
Because it is various that communication network relates to business model, has the magnanimity public users, require high to service feature.In recent years, it is found that live network is different from the regular network and the random network of broad research in the past, have the statistical nature of worldlet effect (small-world effect) and scale free characteristic (scale-free property), this network is named as complex network.Complex network is the topology basis of the topological abstract of a large amount of real complex system and existence, and its research is helped correct understanding and description " network in the real world ".
The characteristics of the power-law distribution that presents at data service performance in the communication network actual motion, the robustness that the analysis communication network has chance failure reaches the fragility to collecting and distributing node calculated attack, design a communication network data service performance self-adapting monitoring model based on power law, by a small amount of collecting and distributing node in the key monitoring network, guarantee safety, performance and the quality of service of communication network data service.
OSS (OSS) is data-centered to Network Management, operator is in operation process for many years, expansion along with growth, network complexity and the scale of time, these data messages present explosive increase, how from the data of these magnanimity, to obtain O﹠M personnel and information that operator was concerned about, and therefrom to obtain information knowledge be the urgent problem of being concerned about of operator.
We notice, although in these huge report systems, relate to many-sided network operation data, but wherein can influence network performance and quality, be key performance indicators KPI (the Key PerformanceIndicator of a series of reflection service operation states, hereinafter to be referred as critical index or KPI), KPI is meant and is used for critical technical parameter index that the service operation situation is weighed, generally comprise key business figureofmerit, crucial quality of service index, Key Performance Indicator etc., for example at the short message service amount of short message service.Between these KPI, close inner link is arranged, carry out Modeling Research, just can carry out multianalysis network performance according to the internal relation between the KPI.
Along with the continuous maturation of data warehouse technology and the reinforcement of the application study in the telecommunication service system, data warehouse technology more and more is subject to people's attention.Data warehouse is a subject-oriented, integrated, stable, the data acquisition system that comprises historical data, the aid decision formulation process in being suitable for very much supporting to manage.In recent years, the application study of data warehouse has covered various aspects such as the service, decision-making, performance analysis, charging of telecommunication service, becomes one of focus in the research of telecommunications correlation technique.
Present both at home and abroad each big operator of mobile communication in the work in early stage, in decision-making and macro-management aspect, has formed based on data warehouse and DSS, is the network service and the performance evaluating system of core with group's business datum analysis; In operation and network operation aspect, formed based on telecom operation flow process and database technology, run automation and monitoring system based on the Network of telecom operation back-up environment NGOSS of future generation.
But these systems owing to be used in different data processing techniques, serve different service layers, analyze data from different equipment and channel, and professional service can only be provided for the user of certain service layer.
The invention provides a KPI and analyze surveillance based on OSS, the self-defined KPI of system, towards full-service, application data REPOSITORY TECHNOLOGY, realize analysis, the real time monitoring of various dimensions, multi-level telecommunication service data target, thereby provide comprehensive, accurate, real-time analysis and decision data for the business and the administrative staff of telecom operation.
Summary of the invention
Purpose of the present invention is for providing on the basis based on the framework of traditional telecom business support system, utilize data warehouse and data mining technology, thus the comprehensive real time monitoring monitoring system that realizes the key performance indicators KPI of reflection service operation state in the telecom operation process.
The technical scheme that realizes the object of the invention is:
A kind of system that monitors telecommunications network service performance, this system comprises:
Data Layer: data pick-up conversion loads engine and utilizes the basic data in the predetermined Service Database in the operational indicator system model to carry out unified processed, and operational indicator is extracted conversion, finishes the comprehensive modeling of operational indicator;
Analyze the business logic processing layer: by analysis path engine and intelligent data analysis engine, utilize the operational indicator analytical model, from basic database, obtain qualified data set;
Business diagnosis represents layer: represent form and represent operational means by multiple, the reflection user is towards multi-angle, multi-threaded domain analysis knowledge, and the display analysis result data helps the user to obtain to analyze conclusion flexibly.
The comprehensive modeling of operational indicator comprises:
Indicator layer: extract from original business datum, the achievement data that polymerize provides Data Source for analyzing dimension;
Analyze the dimension layer: the desired indicator of combination forms analyzes dimension, and the user can realize the configuration of dimension by setting up model;
Analyze granularity layers: the presenting of business datum scheduled time granularity;
Analytical model layer: by setting up analytical model, set up the analyze relationship between data between predetermined first theme, realize the redirect between predetermined first theme;
Operation layer: the operation layer under the dimension.
The data pick-up conversion loads the course of work of engine:
From a plurality of Service Databases and OMP resources bank, utilize the rule of data pick-up conversion modeling, with data centralization in the comprehensive modeling of operational indicator storehouse.
This unified data framework basis has solved the inconsistent problem of data in using, the data basis of having prepared enough degree of depth and range for later analysis.
The course of work of intelligent data analysis engine and analysis path engine:
The user represents the user interface of layer by business, determines the business diagnosis theme, simultaneously defined analysis represent means, drill through etc. as form, multi-angle, send analysis request to the analysis path engine; Analysis path engine analysis request, and a path data business of selecting of returning to represents layer; Business represents layer relevant analysis request is sent to the intelligent data analysis engine, and the intelligent data analysis engine is used corresponding analytic function by computing, obtains qualified data slicer; The intelligent data analysis engine returns analysis result and represents layer to business.
Also comprise the early warning regulation engine in the system of monitoring telecommunications network service performance of the present invention, be used to carry out professional early warning and traffic supervision work.
The course of work of early warning regulation engine:
Utilize the related data of collection alarm of FMS equipment and testing alarm; FMS sends to the early warning regulation engine with alarm data; The early warning regulation engine utilizes achievement data in the comprehensive modeling of alarm data, operational indicator of FMS and the resources relationship in the OMP resources bank, carry out professional early warning and supervision, when producing alarm, the alarm data loopback is handled to FMS, and the early warning regulation engine is realized KPI index real time monitoring.
The wherein professional business diagnosis theme that represents layer includes but not limited to following theme:
Traffic analysis: according to predetermined business process model, realize the analysis of the relevant KPI of professional overall traffic and traffic carrying capacity composition, predict, analyze, compare professional operation conditions.
Quality of service is analyzed: according to predetermined business process model, realization is at the analysis of the relevant KPI of reaction service operation situation quality, by each stage of service operation and professional total service quality are analyzed, in time find influence weak link in the running of the key element of service operation and business.
Utilance is analyzed:, realize to capacity utilization that the problem that may exist in service is in time found in the analysis of relevant KPI such as resources occupation rate, capacity utilance, traffic load, prediction at the utilization rate of resources such as key business equipment, business configuration.
User behavior analysis: at dissimilar customer service usage behaviors, realize the post analysis of relevant KPI, digging user uses professional preference and professional welcome degree.
Analyze festivals or holidays: according to the model of being scheduled to festivals or holidays, realize real time monitoring and the post analysis of relevant KPI, predict, analyze, compare the situation of service operation, for the basic means that provide are provided festivals or holidays.
Analyze when busy: model when hurrying according to predetermined business, realize the analysis of relevant KPI such as traffic carrying capacity, quality of service, utilance.
Specific activity is analyzed: according to special requirement, realize the analysis of the relevant KPI of the business that the special time period carries out.
Beneficial effect of the present invention is:
1, telecommunications network service performance monitoring system of the present invention is based on the basis of framework of traditional telecom business support system, utilize data warehouse and data mining technology, thus comprehensive real time monitoring monitoring and the early warning that realizes the key performance indicators KPI of reflection service operation state in the whole operation process of telecommunications.
2, the present invention is that a KPI based on OSS analyzes surveillance, the self-defined KPI of system, towards full-service, application data REPOSITORY TECHNOLOGY, realize analysis, the real time monitoring of various dimensions, multi-level telecommunication service data target, thereby provide comprehensive, accurate, real-time analysis and decision data for the business and the administrative staff of telecom operation.
Description of drawings
Fig. 1 is the system configuration schematic diagram of monitoring telecommunications network service performance
Fig. 2 loads engine work schematic diagram for the data pick-up conversion
Fig. 3 is intelligent data analysis engine and analysis path engine work schematic diagram
Fig. 4 is an early warning regulation engine work schematic diagram
Fig. 5 is the comprehensive modeling structure schematic diagram of operational indicator
Fig. 6 is for analyzing the topic model structural representation
Embodiment
Embodiment 1
The system of monitoring telecommunications network service performance comprises:
Data Layer: data pick-up conversion loads engine and utilizes the basic data in the predetermined Service Database in the operational indicator system model to carry out unified processed, and operational indicator is extracted conversion, finishes the comprehensive modeling of operational indicator;
The comprehensive modeling of operational indicator comprises:
Indicator layer: extract from original business datum, the achievement data that polymerize provides Data Source for analyzing dimension;
Analyze the dimension layer: the desired indicator of combination forms analyzes dimension, and the user can realize the configuration of dimension by setting up model;
Analyze granularity layers: the presenting of business datum scheduled time granularity;
Analytical model layer: by setting up analytical model, set up the analyze relationship between data between predetermined first theme, realize the redirect between predetermined first theme;
Operation layer: the operation layer under the dimension.
Analyze the business logic processing layer: by analysis path engine and intelligent data analysis engine, utilize the operational indicator analytical model, from basic database, obtain qualified data set;
Business diagnosis represents layer: represent form and represent operational means by multiple, the reflection user is towards multi-angle, multi-threaded domain analysis knowledge, and the display analysis result data helps the user to obtain to analyze conclusion flexibly.
Its course of work is: data pick-up conversion loads engine from a plurality of short message service datas storehouse and OMP resources bank, utilizes the rule of data pick-up conversion modeling, with data centralization in the comprehensive modeling of operational indicator storehouse; The user represents the user interface of layer by business, determines the business diagnosis theme: the trend analysis of short message service amount, simultaneously defined analysis represent means: form sends analysis request to the analysis path engine; Analysis path engine analysis request, and a path data business of selecting of returning to represents layer; Business represents layer relevant analysis request is sent to the intelligent data analysis engine, and the intelligent data analysis engine is used corresponding analytic function by computing, obtains qualified data slicer; The intelligent data analysis engine returns analysis result and represents layer to business.
Native system also comprises the early warning regulation engine: be used to carry out professional early warning and traffic supervision work;
Its course of work is: the related data of utilizing collection alarm of FMS equipment and testing alarm; FMS sends to the early warning regulation engine with alarm data; The early warning regulation engine utilizes achievement data in the comprehensive modeling of alarm data, operational indicator of FMS and the resources relationship in the OMP resources bank, carry out professional early warning and supervision, when producing alarm, the alarm data loopback is handled to FMS, and the early warning regulation engine is realized KPI index real time monitoring.