CN108683518A - The fast data processing method and device for ringing the Internet, applications system is cooperateed with to the world - Google Patents
The fast data processing method and device for ringing the Internet, applications system is cooperateed with to the world Download PDFInfo
- Publication number
- CN108683518A CN108683518A CN201810294562.XA CN201810294562A CN108683518A CN 108683518 A CN108683518 A CN 108683518A CN 201810294562 A CN201810294562 A CN 201810294562A CN 108683518 A CN108683518 A CN 108683518A
- Authority
- CN
- China
- Prior art keywords
- data
- kpis
- network
- exceptional value
- threshold value
- 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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/06—Management of faults, events, alarms or notifications
- H04L41/0631—Management of faults, events, alarms or notifications using root cause analysis; using analysis of correlation between notifications, alarms or events based on decision criteria, e.g. hierarchy, tree or time analysis
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/06—Management of faults, events, alarms or notifications
- H04L41/0654—Management of faults, events, alarms or notifications using network fault recovery
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/06—Management of faults, events, alarms or notifications
- H04L41/0681—Configuration of triggering conditions
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/16—Threshold monitoring
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The fast data processing method and device for ringing the Internet, applications system, the present invention is cooperateed with quickly and effectively to make detecting, compensation, reparation by using machine learning method and big data treatment technology in the world the invention discloses a kind of.The problem of cannot fast and accurately identifying that world collaboration rings the failure of the Internet, applications system soon in the prior art to efficiently solve, and quickly it is repaired.
Description
Technical field
The present invention relates to field of computer technology, and fast sound the Internet, applications system is cooperateed with to the world more particularly to a kind of
Data processing method and device.
Background technology
Collaboration fast sound the Internet, applications architectural framework design consideration in the world is to rely on ground tactics internet, performance day
The advantages such as the covering of base means is wide, response is fast, it is main target to provide hot spot region coverage enhancement and tactics collaboration.In terms of space-based with
Micro-nano rings satellite node and carries out United Dispatching as main resource soon, and service cloud platform is deployed in tactics edge customer communication section
Point, anchor point cloud platform are that micro-nano rings satellite node network management, user access administration and tactical information service and alllication soon
Support platform;Basic cloud platform is deployed in management through specialized department department, and service cloud platform is connected to basic cloud by infrastructure network
Platform, and receive the configuration management strategy of basic cloud platform, basic data and service are shared, as shown in Figure 1.
The rapid deployment of business is supported in data center internet, simplifies business configuration flow, and there is flexible network to expand
Exhibition ability reduces device configuration risk, improves network operation efficiency.The technology related generally to has SDN, NFV, big data technology
Deng.In the world cooperates with fast sound the Internet, applications architectural framework, the network at user oriented center and the network towards data center
Middle NE quantity is more, and data traffic is big, only by being difficult manually to carry out in time and accurately failure is diagnosed and repaired.
Invention content
The fast data processing method and device for ringing the Internet, applications system is cooperateed with to the world the present invention provides a kind of, with solution
Certainly the prior art cannot fast and accurately identify the failure of the fast sound the Internet, applications system of world collaboration, and quickly be carried out to it
The problem of reparation.
On the one hand, the fast data processing method for ringing the Internet, applications system, packet are cooperateed with to the world the present invention provides a kind of
It includes:The data in the network towards customer center and the network towards data center are collected by network element and operational administrative system,
The data are pre-processed;KPIs and alarm data are calculated, judges the exceptional value of the data whether higher than preset different
Constant value threshold value if it is, the method by machine learning carries out fault diagnosis, and repaiies failure according to diagnostic result
It is multiple.
Further, this method further includes:The data are put into Hive databases and stores and manages.
Further, this method further includes:After fault restoration, generates report and return to operational administrative system.
Further, KPIs and alarm data are calculated, judges whether the exceptional value of the data is higher than preset exceptional value
Threshold value specifically includes:According to the failure cause of history and KPIs is calculated, it is super to calculate the KPIs when each failure cause occurs
The probability for going out threshold value, trains Bayesian network model;
The KPIs values for inputting current data, judge whether the exceptional value of the data is higher than preset exceptional value threshold value, such as
Fruit is then otherwise to continue to detect by the method for machine learning progress fault diagnosis.
On the other hand, the fast data processing equipment for ringing the Internet, applications system is cooperateed with to the world the present invention also provides a kind of,
Including:Collector unit, for by the network of network element and the collection of operational administrative system towards customer center and towards data center
Network in data, the data are pre-processed;Processing unit, for calculating KPIs and alarm data, described in judgement
Whether the exceptional value of data is higher than preset exceptional value threshold value, if it is, the method by machine learning carries out fault diagnosis,
And failure is repaired according to diagnostic result.
Further, which further includes:Storage unit is stored and is managed for the data to be put into Hive databases
Reason.
Further, the processing unit is additionally operable to, and after fault restoration, is generated report and is returned to operational administrative system.
Further, the processing unit is additionally operable to, and according to the failure cause of history and KPIs is calculated, calculates every
KPIs exceeds the probability of threshold value when kind failure cause occurs, and trains Bayesian network model;The KPIs values of current data are inputted,
Judge whether the exceptional value of the data is higher than preset exceptional value threshold value, if it is, the method by machine learning carries out
Otherwise fault diagnosis continues to detect.
The present invention has the beneficial effect that:
The present invention is to rely on double center applications architectural frameworks, network for user oriented center and towards data center
Network in failure problems, by using machine learning method and big data treatment technology quickly and effectively make detecting, mend
It repays, repair.The fast sound the Internet, applications body of world collaboration cannot be fast and accurately identified in the prior art to efficiently solve
The failure of system, and the problem of quickly repaired to it.
Description of the drawings
Fig. 1, which is that world collaboration is fast, rings the Internet, applications architectural framework schematic diagram;
Fig. 2 is a kind of stream cooperateing with the fast data processing method for ringing the Internet, applications system to the world of the embodiment of the present invention
Journey schematic diagram;
Fig. 3 is that the another of the embodiment of the present invention cooperates with the fast data processing method for ringing the Internet, applications system to the world
Flow diagram;
Fig. 4 is a kind of knot cooperateing with the fast data processing equipment for ringing the Internet, applications system to the world of the embodiment of the present invention
Structure schematic diagram.
Specific implementation mode
The fast failure for ringing the Internet, applications system of world collaboration cannot be fast and accurately identified in order to solve the prior art,
And the problem of quickly it is repaired, the world is cooperateed at the fast data for ringing the Internet, applications system the present invention provides a kind of
Reason method, the present invention quickly and effectively make detecting by using machine learning method and big data treatment technology, compensate, repair
It is multiple, cost of labor is saved, the accuracy of fault diagnosis is improved.Below in conjunction with attached drawing and embodiment, the present invention is carried out
It is further described.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, this hair is not limited
It is bright.
The fast data processing method for ringing the Internet, applications system is cooperateed with to the world an embodiment of the present invention provides a kind of, referring to
Fig. 2, this method include:
S201, network towards customer center is collected by network element and operational administrative system and towards the network of data center
In data, the data are pre-processed;
S202, KPIs and alarm data are calculated, judge whether the exceptional value of the data is higher than preset exceptional value threshold value,
If it is, the method by machine learning carries out fault diagnosis, and failure is repaired according to diagnostic result.
The present invention quickly and effectively makes detecting by using machine learning method and big data treatment technology, compensates, repaiies
It is multiple, cost of labor is saved, the accuracy of fault diagnosis is improved.
When it is implemented, the method described in the embodiment of the present invention further includes:The data are put into Hive databases and are deposited
Storage and management.
Also, the embodiment of the present invention generates report and returns to operational administrative system after fault restoration.
When it is implemented, calculating KPIs and alarm data described in the embodiment of the present invention, judge that the exceptional value of the data is
It is no to be higher than preset exceptional value threshold value, it specifically includes:According to the failure cause of history and KPIs is calculated, calculates in each event
Hinder the probability that KPIs when reason occurs exceeds threshold value, trains Bayesian network model;The KPIs values of current data are inputted, are judged
Whether the exceptional value of the data is higher than preset exceptional value threshold value, if it is, the method by machine learning carries out failure
Diagnosis, otherwise, continues to detect.
In simple terms, the present invention be with double center applications architectural frameworks be rely on, for user oriented center network and
Failure problems in the network at data-oriented center are quickly and effectively made using machine learning method and big data treatment technology
Detecting, compensation and reparation.
That is, the present invention the world cooperate with it is fast ring the Internet, applications architectural framework on the basis of, for user oriented
The mass data generated in the network at center and network towards data center introduces big data treatment technology and carries out failure and detects
It surveys, diagnosis and reparation, key step include:1) by mass data present in network element and operational administrative system collection network,
Big data is put into database and stores and manages, is simply pre-processed.2) by calculating correlation KPIs and alarm data, sentence
It is disconnected whether to be higher than outlier threshold, if then being diagnosed;Otherwise, continue to monitor.3) failure is carried out using the method for machine learning
Diagnosis, such as Bayesian network, fuzzy set theory, genetic algorithm.4) according to system diagnostics as a result, the action for executing response is used
In fault restoration.5) report is generated after fault restoration return to operational administrative system.
Fig. 3 is that the another of the embodiment of the present invention cooperates with the fast data processing method for ringing the Internet, applications system to the world
Flow diagram, below in conjunction with Fig. 3, by a specific example to method of the present invention carry out detailed explanation and
Explanation:
1. in the network data deposit Hive data warehouses that network element and operational administrative system generate.
2. data prediction.Find out relevant KPI data and alarm data, including (accessibility of the network, stability,
Load, receives signal strength etc.).
3. judging whether related data is more than outlier threshold.
4. if so, judging that the network area is broken down.Utilize the data of the failure cause and correlation KPIs of history, meter
The probability that the correlation KPIs when each failure cause occurs exceeds threshold value is calculated, Bayesian network model (as above figure) is trained.It is defeated
Enter the current related KPIs values that network failure occurs, if it is that covering signal is weak that probability is highest, it is to cover to diagnose the failure cause
Lid signal is weak.Otherwise, continue to detect.
5. calculating optimal solution, corresponding actions are made, such as improve antenna gain.
For the embodiment of the present invention in terms of detecting fault, detecting fault includes identifying the network that abnormal interrupt is serviced in network.
When present networks are run, a large amount of network data will produce.This key point is exactly to find relevant announcement by data processing technique
Alert data and KPIs.The method of detecting is made of the threshold value of some KPIs, so, when some KPI value exceeds abnormal behaviour
When threshold value, for example receive signal strength and be less than -63dBm, whether network is abnormal by the Auto-Sensing region.
For the embodiment of the present invention in terms of diagnosis, diagnosis is that the analysis relevant KPIs of failure finds out corresponding reason.This work(
It can be responsible for identification failure cause, reason may be hardware problem or software issue, it may be possible to the parameter configuration of mistake, it is also possible to
Cover defect etc..Reason will be divided into basic fault and the system failure by we.Basic fault is mainly that the network equipment is (hardware, soft
Part or functionality resources) caused by, the system failure does not have correlation (such as covering, interaction, configuration with the given network equipment
Deng).Currently, the method for many machine learning is applied in diagnosis, such as Bayesian network, fuzzy set, genetic algorithm.
For the embodiment of the present invention in terms of fault restoration, the purpose of this function of fault restoration is provided accordingly to solve failure
Repair action.Action can be divided into simple action and parameter actions.Simple action is exactly not do depth analysis and directly execute
Action.Under normal conditions, each failure has ready-made simple solution, once problem is determined, and it is corresponding to repair
Double action work can be triggered.On the contrary, parameter actions are not made a response not instead of immediately, calculated by running specific algorithm
Optimal parameter solution, the recovery scenario made can be some actions, can also be a series of actions.
The present invention, which applies, to be cooperateed in the world on fast sound the Internet, applications architectural framework, is had at least the following advantages:
1, network complexity handles skill far beyond the range that can be manually controlled by rule, network by big data
Art independently carries out detecting fault, improves the reaction speed of network failure identification;
2, fault diagnosis is independently carried out by machine learning method, saves cost of labor, improve the standard of fault diagnosis
True property;
3, run specific algorithm and calculate optimal parameter solution, eliminate in the prior art different personnel according to
The solution disunity that same fault is made, ensure that the reliability of fault restoration.
The embodiment of the present invention additionally provides a kind of data processing equipment cooperateing with fast sound the Internet, applications system to the world, ginseng
See Fig. 4, which includes:Collector unit, for collected by network element and operational administrative system network towards customer center and
Data in the network at data-oriented center pre-process the data;Processing unit, for calculating KPIs and alarm number
According to, judge whether the exceptional value of the data is higher than preset exceptional value threshold value, if it is, by the method for machine learning into
Row fault diagnosis, and failure is repaired according to diagnostic result.
The device further includes:The data are put into Hive databases and store and manage by storage unit.
When it is implemented, processing unit described in the embodiment of the present invention is additionally operable to, after fault restoration, generates report and return to operation
Management system.
Further, processing unit described in the embodiment of the present invention is additionally operable to, and according to the failure cause of history and is calculated
KPIs calculates the probability that the KPIs when each failure cause occurs exceeds threshold value, trains Bayesian network model;Input is current
The KPIs values of data, judge whether the exceptional value of the data is higher than preset exceptional value threshold value, if it is, passing through engineering
The method of habit carries out fault diagnosis and otherwise continues to detect.
The relevant portion of apparatus of the present invention embodiment can refer to embodiment of the method part and be understood, no longer carry out herein detailed
Carefully repeat.
Although being example purpose, the preferred embodiment of the present invention is had been disclosed for, those skilled in the art will recognize
Various improvement, increase and substitution are also possible, and therefore, the scope of the present invention should be not limited to the above embodiments.
Claims (8)
1. a kind of cooperateing with the fast data processing method for ringing the Internet, applications system to the world, which is characterized in that including:
The data in the network towards customer center and the network towards data center are collected by network element and operational administrative system,
The data are pre-processed;
KPIs and alarm data are calculated, judges whether the exceptional value of the data is higher than preset exceptional value threshold value, if it is,
Fault diagnosis is carried out by the method for machine learning, and failure is repaired according to diagnostic result.
2. according to the method described in claim 1, it is characterized in that, further including:
The data are put into Hive databases and stores and manages.
3. according to the method described in claim 1, it is characterized in that, further including:
After fault restoration, generates report and return to operational administrative system.
4. according to the method described in any one of claim 1-3, which is characterized in that calculate KPIs and alarm data, judge
Whether the exceptional value of the data is higher than preset exceptional value threshold value, specifically includes:
According to the failure cause of history and KPIs is calculated, it is general beyond threshold value to calculate the KPIs when each failure cause occurs
Rate trains Bayesian network model;
The KPIs values for inputting current data, judge whether the exceptional value of the data is higher than preset exceptional value threshold value, if so,
Then otherwise continue to detect by the method for machine learning progress fault diagnosis.
5. a kind of cooperateing with the fast data processing equipment for ringing the Internet, applications system to the world, which is characterized in that including:
Collector unit, for by the network of network element and the collection of operational administrative system towards customer center and towards data center
Data in network pre-process the data;
Processing unit judges whether the exceptional value of the data is higher than preset exceptional value for calculating KPIs and alarm data
Threshold value if it is, the method by machine learning carries out fault diagnosis, and repairs failure according to diagnostic result.
6. device according to claim 5, which is characterized in that further include:
Storage unit is stored and is managed for the data to be put into Hive databases.
7. device according to claim 5, which is characterized in that
The processing unit is additionally operable to, and after fault restoration, is generated report and is returned to operational administrative system.
8. according to the device described in any one of claim 5-7, which is characterized in that
The processing unit is additionally operable to, and according to the failure cause of history and KPIs is calculated, calculates and is sent out in each failure cause
KPIs exceeds the probability of threshold value when raw, trains Bayesian network model;The KPIs values for inputting current data, judge the data
Exceptional value whether be higher than preset exceptional value threshold value, it is no if it is, carry out fault diagnosis by the method for machine learning
Then, continue to detect.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810294562.XA CN108683518A (en) | 2018-03-30 | 2018-03-30 | The fast data processing method and device for ringing the Internet, applications system is cooperateed with to the world |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810294562.XA CN108683518A (en) | 2018-03-30 | 2018-03-30 | The fast data processing method and device for ringing the Internet, applications system is cooperateed with to the world |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108683518A true CN108683518A (en) | 2018-10-19 |
Family
ID=63799721
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810294562.XA Pending CN108683518A (en) | 2018-03-30 | 2018-03-30 | The fast data processing method and device for ringing the Internet, applications system is cooperateed with to the world |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108683518A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110062410A (en) * | 2019-03-28 | 2019-07-26 | 东南大学 | A kind of cell outage detection localization method based on adaptive resonance theory |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103731221A (en) * | 2014-01-09 | 2014-04-16 | 中国航天标准化研究所 | Space and ground integrated network system availability determining method |
US20170139816A1 (en) * | 2015-11-17 | 2017-05-18 | Alexey Sapozhnikov | Computerized method and end-to-end "pilot as a service" system for controlling start-up/enterprise interactions |
CN107171819A (en) * | 2016-03-07 | 2017-09-15 | 北京华为数字技术有限公司 | A kind of network fault diagnosis method and device |
CN107787023A (en) * | 2017-11-01 | 2018-03-09 | 北京邮电大学 | The low orbit satellite route generating method and device of a kind of Incorporate network |
-
2018
- 2018-03-30 CN CN201810294562.XA patent/CN108683518A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103731221A (en) * | 2014-01-09 | 2014-04-16 | 中国航天标准化研究所 | Space and ground integrated network system availability determining method |
US20170139816A1 (en) * | 2015-11-17 | 2017-05-18 | Alexey Sapozhnikov | Computerized method and end-to-end "pilot as a service" system for controlling start-up/enterprise interactions |
CN107171819A (en) * | 2016-03-07 | 2017-09-15 | 北京华为数字技术有限公司 | A kind of network fault diagnosis method and device |
CN107787023A (en) * | 2017-11-01 | 2018-03-09 | 北京邮电大学 | The low orbit satellite route generating method and device of a kind of Incorporate network |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110062410A (en) * | 2019-03-28 | 2019-07-26 | 东南大学 | A kind of cell outage detection localization method based on adaptive resonance theory |
CN110062410B (en) * | 2019-03-28 | 2021-09-28 | 东南大学 | Cell interruption detection positioning method based on self-adaptive resonance theory |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106257872B (en) | WI-FI access point performance management system and method | |
EP1279211B1 (en) | Topology-based reasoning apparatus for root-cause analysis of network faults | |
CN106165345B (en) | Promote method, system and the computer readable storage medium of network failure in neutralizing data center | |
EP3131234B1 (en) | Core network analytics system | |
US9413597B2 (en) | Method and system for providing aggregated network alarms | |
CN109992440A (en) | A kind of IT root accident analysis recognition methods of knowledge based map and machine learning | |
KR20180108446A (en) | System and method for management of ict infra | |
CN107196780A (en) | Method and apparatus for the failure of management equipment | |
JPH08307524A (en) | Method and equipment for discriminating risk in abnormal conditions of constitutional element of communication network | |
CN106888106A (en) | The extensive detecting system of IT assets in intelligent grid | |
CN105719056A (en) | Small and medium-sized enterprise event management process | |
CN107679576A (en) | The fault monitoring method and device of vehicle | |
CN113516244B (en) | Intelligent operation and maintenance method and device, electronic equipment and storage medium | |
CN105471595A (en) | Fault processing management device and fault processing management method | |
CN109884473A (en) | A kind of electric power overhaul system and method | |
CN109345076A (en) | A kind of whole process engineering consulting project risk management method | |
CN106330588A (en) | BFD detection method and device | |
CN109039795A (en) | A kind of Cloud Server resource monitoring method and system | |
Pernici et al. | Automatic learning of repair strategies for web services | |
CN109995558A (en) | Failure information processing method, device, equipment and storage medium | |
CN106487597A (en) | A kind of service monitoring system and method based on Zookeeper | |
CN108683518A (en) | The fast data processing method and device for ringing the Internet, applications system is cooperateed with to the world | |
Velasco | A Bayesian Network approach to diagnosing the root cause of failure from Trouble Tickets. | |
US11337087B2 (en) | Methods, systems and computer readable media for predicting risk in network elements using machine learning | |
JP3470683B2 (en) | Network operation management system and device failure probability management method |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181019 |