CN107579858A - The alarm method and device of cloud main frame, communication system - Google Patents
The alarm method and device of cloud main frame, communication system Download PDFInfo
- Publication number
- CN107579858A CN107579858A CN201710906291.4A CN201710906291A CN107579858A CN 107579858 A CN107579858 A CN 107579858A CN 201710906291 A CN201710906291 A CN 201710906291A CN 107579858 A CN107579858 A CN 107579858A
- Authority
- CN
- China
- Prior art keywords
- main frame
- performance data
- cloud main
- data
- cloud
- 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
Landscapes
- Computer And Data Communications (AREA)
Abstract
The invention discloses a kind of alarm method of cloud main frame and device, communication system.Wherein, this method includes:The performance data that at least one cloud main frame uploads is obtained, wherein, performance data is used for the running status for reflecting at least one cloud main frame;Performance data is analyzed using preset model, it is determined whether at least one cloud main frame is alerted, wherein, preset model trains to obtain using multi-group data by machine learning, and every group of data in multi-group data include:Alerted corresponding to performance data and the performance data;In the case of it is determined that being alerted, alarm action corresponding with least one cloud main frame is performed.The increase that the present invention solves monitoring data present in correlation technique causes inquiry slowly and alarm measurement is the artificial technical problem for setting and causing alarm by mistake easily occur.
Description
Technical field
The present invention relates to cloud host performance monitor field, alarm method and device in particular to a kind of cloud main frame,
Communication system.
Background technology
In cloud computing platform, Ceilometer is commonly used to realize cloud host monitor and alarm function.The partial function
Realization dependent on RPC message channels, MongoDB data storage mediums and pass through artificial default properties threshold values and realize
Basic alarm function, it is primarily present some following problem:
1:Remote procedure call protocol (RPC) message channel based on Rabbitmq is the important component of cloud computing platform,
Communication between the modules such as nova, cinder depends on RPC message queues, the Message Processing speed of the queue, often shadow
Ring the operational efficiency of cloud computing platform.
2:MongoDB exchanges the data storage medium of read-write speed for internal memory, can disappear along with ever-increasing data volume
The internal memory of a large amount of main controlled nodes is consumed, and then causes the process failures such as haproxy, apache, causes the overall steady of cloud computing platform
It is qualitative to ensure.
3:The formulation of warning strategies depends critically upon subjective behavior, the default constantly consumption mail of malice, SMS notification be present
Situations such as Deng interface.
In existing build environment, prevent and slow down the function of cloud host monitor and alarm to cloud from the level of O&M more
The influence of calculating platform, such as:MongoDB and control node are peeled off, are individually deployed to a single server, reduce by
Largely taken in MongoDB internal memories, the problems such as causing platform response speed rapidly to reduce, it is equally serious for RPC message channels
Occupation problem, also mainly specify specific RPC to carry out special Message Processing by being serviced to Ceilometer, subtract come what is tried one's best
Occupancy of the slow monitoring process to main message channel.And corresponding alarm function, mainly take on the basis of sacrificing flexibility, to each
The performance monitoring threshold values of item cloud main frame carries out default or fixed mode and carried out.
As can be seen here, mainly from the angle of O&M in correlation technique, to alleviate cloud host monitor and alarm function brings
To the performance impact of whole platform, while alarm relies primarily on artificial preset, it is difficult to estimates different types of server and normally transports
Performance threshold values under row situation, occur situation about reporting by mistake unavoidably, increase the cost of system O&M.
For it is above-mentioned the problem of, not yet propose effective solution at present.
The content of the invention
The embodiments of the invention provide a kind of alarm method of cloud main frame and device, communication system, at least to solve correlation
The increase of monitoring data present in technology causes inquiry slowly and alarm measurement is that artificial set causes easily to occur alerting by mistake
Technical problem.
One side according to embodiments of the present invention, there is provided a kind of alarm method of cloud main frame, including:Obtain at least one
The performance data that individual cloud main frame uploads, wherein, performance data is used for the running status for reflecting at least one cloud main frame;Using default
Model is analyzed performance data, it is determined whether at least one cloud main frame alerted, wherein, preset model is more to use
Group data train what is obtained by machine learning, and every group of data in multi-group data include:Performance data and the performance data
Corresponding alarm;In the case of it is determined that being alerted, alarm action corresponding with least one cloud main frame is performed.
Alternatively, the performance data that at least one cloud main frame uploads is obtained, including:From corresponding with least one cloud main frame
Performance data is obtained in buffer queue;Method also includes:Whether it is empty every preset time period detection buffer queue, is detecting
When buffer queue reaches predetermined threshold value for empty number, then the queue is removed.
Alternatively, buffer queue and the cloud main frame at least one cloud main frame are one-to-one.
Alternatively, obtaining the performance data that at least one cloud main frame uploads includes:Receive at least one cloud main frame pass through it is negative
The performance data of equalization server forwarding is carried, wherein, load-balanced server is used to pre-process performance data, pre-processes
Including at least one of:Arrange merging, the normalized of data structure, selection forwarding queue.
Alternatively, before receiving the performance data that at least one cloud main frame is forwarded by load-balanced server, method is also
Including:Data cleansing is carried out to performance data.
Alternatively, after obtaining the performance data that at least one cloud main frame uploads, method also includes:Receive at least one cloud
The original performance data that main frame uploads;Original performance data is classified, obtains primary sources and secondary sources;By
A kind of data storage receives the call instruction of human-computer interaction interface into caching, to show the operation of at least one cloud main frame
State;Secondary sources are inputted into preset model, preset model is trained or whether accused according to secondary sources
It is alert.
Alternatively, the identification information of cloud main frame is also carried in performance data.
According to the other side of the embodiment of the present application, there is provided a kind of communication system, the system include:At least one cloud
Main frame, for uploading performance data, wherein, performance data is used for the running status for reflecting at least one cloud main frame;Load balancing
Server, sent for being pre-processed to performance data, and by pretreated performance data into distributed memory system,
Wherein, pretreatment includes at least one of:Arrange merging, the normalized of data structure, selection forwarding queue;It is distributed
Storage system, for storing the pretreated performance data of load-balanced server forwarding;Alarm server, for using pre-
If model is analyzed performance data, it is determined whether at least one cloud main frame is alerted;It is determined that the feelings alerted
Under condition, alarm action corresponding with least one cloud main frame is performed;Wherein, preset model is to pass through engineering using multi-group data
Practise what training obtained, every group of data in multi-group data include:Alerted corresponding to performance data and the performance data;
Alternatively, distributed memory system, for by the pretreated performance data of load-balanced server store to
In HBase (Hadoop Database) database.
According to the another aspect of the embodiment of the present application, there is provided a kind of alarm device of cloud main frame, including:Acquisition module,
The performance data uploaded for obtaining at least one cloud main frame, wherein, performance data is used for the fortune for reflecting at least one cloud main frame
Row state;Determining module, for being analyzed using preset model performance data, it is determined whether enter at least one cloud main frame
Row alarm, wherein, preset model trains to obtain using multi-group data by machine learning, every group of data in multi-group data
Include:Alerted corresponding to performance data and the performance data;Alarm module, in the case of it is determined that being alerted, holding
Row alarm action corresponding with least one cloud main frame.
According to the another aspect of the embodiment of the present application, there is provided a kind of storage medium, storage medium include the program of storage,
Wherein, equipment performs the alarm method of the cloud main frame of the above where controlling storage medium when program is run.
According to the another aspect of the embodiment of the present application, there is provided a kind of processor, processor are used for operation program, wherein,
The alarm method of the cloud main frame of the above is performed when program is run.
In embodiments of the present invention, performance data is divided using the preset model for training to obtain by machine learning
Analysis, determines whether the mode alerted to cloud main frame, realize flexibly to be alerted according to different operation conditions
Technique effect, and then the increase for solving monitoring data present in correlation technique cause inquiry slowly and alarm measurement be
It is artificial that the technical problem for causing alarm by mistake easily occur is set.
Brief description of the drawings
Accompanying drawing described herein is used for providing a further understanding of the present invention, forms the part of the application, this hair
Bright schematic description and description is used to explain the present invention, does not form inappropriate limitation of the present invention.In the accompanying drawings:
Fig. 1 is the schematic flow sheet according to a kind of alarm method of cloud main frame of the embodiment of the present application;
Fig. 2 is a kind of structured flowchart of the alarm device of cloud main frame according to embodiments of the present invention;
Fig. 3 is a kind of structural representation of communication system according to embodiments of the present invention;
Fig. 4 is a kind of structural representation of optional communication system according to embodiments of the present invention.
Embodiment
In order that those skilled in the art more fully understand the present invention program, below in conjunction with the embodiment of the present invention
Accompanying drawing, the technical scheme in the embodiment of the present invention is clearly and completely described, it is clear that described embodiment is only
The embodiment of a part of the invention, rather than whole embodiments.Based on the embodiment in the present invention, ordinary skill people
The every other embodiment that member is obtained under the premise of creative work is not made, it should all belong to the model that the present invention protects
Enclose.
It should be noted that term " first " in description and claims of this specification and above-mentioned accompanying drawing, "
Two " etc. be for distinguishing similar object, without for describing specific order or precedence.It should be appreciated that so use
Data can exchange in the appropriate case, so as to embodiments of the invention described herein can with except illustrating herein or
Order beyond those of description is implemented.In addition, term " comprising " and " having " and their any deformation, it is intended that cover
Cover it is non-exclusive include, be not necessarily limited to for example, containing the process of series of steps or unit, method, system, product or equipment
Those steps or unit clearly listed, but may include not list clearly or for these processes, method, product
Or the intrinsic other steps of equipment or unit.
For ease of understanding the embodiment of the present application, technical term involved in the embodiment of the present application is summarized as follows below:
Cloud main frame, it is important component of the cloud computing in infrastructure application, positioned at cloud computing industrial chain pyramid
Bottom, product atom computing machine platform.Cloud main frame is the part that multiple similar unique host are fictionalized on one group of cluster system, collection
There is a mirror image of cloud main frame in group on each main frame, so as to improve the security and stability of fictitious host computer.Cloud main frame is to integrate
The IT infrastructure ability rental service of Internet resources is calculated, be stored in, the use on demand based on cloud computing mode can be provided
With the rental server service of pay-for-use ability.
In correlation technique, because warning strategies rely primarily on artificial preset, it is difficult to it is normal to estimate different types of server
Performance threshold values under operation conditions, therefore, occur situation about reporting by mistake unavoidably, increase the cost of system O&M.To solve above-mentioned ask
Topic, the embodiment of the present application, using distributed raw information collection mode, report and collect data to Spark distributed computing systems
In storage medium, by data cleansing, the corresponding step such as form conversion, the real-time monitoring and controls of data is realized.Introduce simultaneously
The machine learning algorithms such as random forest, Bayes's classification, realize and alarm judgement is carried out to real-time monitoring data, and timing updates machine
Device learning algorithm.Described in detail below in conjunction with accompanying drawing and related embodiment.
Embodiment 1
According to embodiments of the present invention, there is provided a kind of embodiment of the method for the alarm of cloud main frame is, it is necessary to illustrate, attached
The step of flow of figure illustrates can perform in the computer system of such as one group computer executable instructions, though also,
So logical order is shown in flow charts, but in some cases, can be with different from shown by order execution herein
Or the step of description.
Fig. 1 is the schematic flow sheet according to a kind of alarm method of cloud main frame of the embodiment of the present application.As shown in figure 1, should
Method comprises the following steps:
Step S102, the performance data that at least one cloud main frame uploads is obtained, wherein, performance data is used to reflect at least one
The running status of individual cloud main frame;
In one alternate embodiment, performance data is obtained from buffer queue corresponding with least one cloud main frame;Often
Detect whether buffer queue is sky every preset time period, when detecting that buffer queue reaches predetermined threshold value for empty number, then
Remove the queue.Alternatively, above-mentioned buffer queue be with the cloud main frame at least one cloud main frame can be one-to-one, when
Right or one-to-many relation, flexibly set with specific reference to actual conditions.
Specifically, implementations below, but not limited to this can be shown as:Calculate node (Compute Node) is run
Instance Agent are in real time by cloud host monitor performance information, write cache:a:Real-time queue, Instance_ID teams
Row mark, queue size 4K, the empty queue of the Qing Dynasty of team, data throw to layer proxy;b:In queue discipline interval time, it is continuously
Sky, upward short-selling queue identity, the empty queue mark for stipulated number of persistently dishing out, then remove the queue;c:Monitor present node
Cloud Host List.Newly-increased then create new queue, removal then removes respective queue.Meanwhile to ensure security, can be to original
Data are backed up.
In one alternate embodiment, calculate node agent data (Compute Node Data Agent) is monitored at a high speed
The data of upthrow are cached, the normal data after forward process to load equalizer.Data processing includes:Dealing of abnormal data,
I.e. abnormity point removes and empty data remove etc..Data structure normalized, such as the cloud host CPU utilization rate phase of multi-core CPU
Conversion is closed, reads and writes the normalized of unit between speed.Standard data format, and data compression process etc..
As the alternative embodiment of the application, obtaining the performance data that at least one cloud main frame uploads includes:Receive
The performance data that at least one cloud main frame is forwarded by load-balanced server, wherein, load-balanced server is used for performance
Data are pre-processed, and pretreatment includes at least one of:Arrange merging, the normalized of data structure, selection forwarding
Queue.Specifically, implemented below process, but not limited to this can be shown as:Load equalizer receives Compute Node Data
The data that Agent is forwarded, arrangement merging is carried out, is selected through overload, selected a suitable forwarding queue, be transmitted to
Spark data-storage systems.Spark data-storage systems are stored data into HBase by way of Dstream.
It should be noted that in order to distinguish the cloud main frame belonging to performance data, cloud can be also carried in performance data
The identification information of main frame.
Alternatively, after the performance data that at least one cloud main frame uploads is obtained, above-mentioned at least one cloud main frame is received
The original performance data of upload;Above-mentioned original performance data is classified, obtains primary sources and secondary sources;Will be upper
State primary sources to store into caching, and receive the call instruction of human-computer interaction interface, to show above-mentioned at least one cloud master
The running status of machine;Above-mentioned secondary sources are inputted into above-mentioned preset model, above-mentioned preset model are trained or foundation
Whether above-mentioned secondary sources are alerted.Specifically, said process can show as implementations below:Perform cloud main frame
Energy data analysis program, by data-analyzing machine, will be dispersed into two bursts of data flows:
A:Real-time stream, intermediate data buffer of the real-time stream Jing Guo 4M sizes are dynamic in real time for Web page
Check current slot or each cloud host computer performance monitoring running status in time point;
B:Sample data stream, fixed intervals data put down weary sampling, and peculiar discrete point is peeled off, batch write Mysql or
MongoDB data storage mediums.Sample data stream checks history monitoring data for echarts and transfers to warning system to call
To analyze and produce alarm.
Step S104, above-mentioned performance data is analyzed using preset model, it is determined whether to above-mentioned at least one cloud
Main frame is alerted, wherein, above-mentioned preset model trains to obtain using multi-group data by machine learning, above-mentioned multigroup number
Every group of data in include:Alerted corresponding to performance data and the performance data;
To ensure precision of analysis, in the property that at least one cloud main frame of reception is forwarded by load-balanced server
Before energy data, data cleansing can also be carried out to performance data.
Above-mentioned preset model a:The cloud host performance data sample under normal condition is obtained, using random forest or pattra leaves
This sorting algorithm etc., train normal data sample, obtain analyzer (equivalent to above-mentioned preset model).b:It is analysed to data biography
Enter correspondence analysis device, show that alarm acts, perform corresponding alarm action, and correspondence database is arrived into warning information storage.
Step S106, in the case of it is determined that being alerted, perform alarm action corresponding with least one cloud main frame.Should
Alarm action includes but is not limited at least one of:Mail reminder, short message are reminded, WEB interface is reminded, API is reminded.
The scheme that the embodiment of the present application provides utilizes distributed memory system, and it is hundreds of to solve cloud computing platform scale
In the case of, the process problem of caused a large amount of monitoring datas, realizes monitoring, while realize monitoring data in real time in short time period
Intelligent alarm, continuous self study updates cluster policy sample space, to realize the monitoring alarm of no manual intervention.
Embodiment 2
Fig. 2 is a kind of structured flowchart of the alarm device of cloud main frame according to embodiments of the present invention.As shown in Fig. 2 the dress
Put including:
Acquisition module 20, the performance data uploaded for obtaining at least one cloud main frame, wherein, above-mentioned performance data is used for
Reflect the running status of above-mentioned at least one cloud main frame;
Determining module 22, acquisition module 20 is coupled to, for being analyzed using preset model above-mentioned performance data, really
It is fixed whether above-mentioned at least one cloud main frame to be alerted, wherein, above-mentioned preset model is to pass through engineering using multi-group data
Practise what training obtained, every group of data in above-mentioned multi-group data include:Alerted corresponding to performance data and the performance data;
Alarm module 24, be coupled to determining module 22, in the case of it is determined that being alerted, perform with it is above-mentioned at least
Alarm corresponding to one cloud main frame acts.
It should be noted that above-mentioned each pattern can be realized by software or hardware, for the latter, Ke Yitong
In the following manner is crossed to realize:Above-mentioned modules are located in same processor;Or the side of above-mentioned modules in any combination
Formula is located in different processors.
It should be noted that the preferred embodiment of the present embodiment may refer to the associated description in embodiment 1, herein not
Repeat again.
Embodiment 3
Fig. 3 is a kind of structural representation of communication system according to embodiments of the present invention.As shown in figure 3, the system includes:
At least one cloud main frame 30, for uploading performance data, wherein, above-mentioned performance data is used to reflect above-mentioned at least one
The running status of individual cloud main frame;
Load-balanced server 32, for being pre-processed to above-mentioned performance data, and by pretreated above-mentioned performance
Data are sent into distributed memory system, wherein, above-mentioned pretreatment includes at least one of:Arrange merging, data structure
Normalized, selection forwarding queue;
Distributed memory system 34, for storing the pretreated above-mentioned performance number of above-mentioned load-balanced server forwarding
According to;
Alarm server 36, for being analyzed using preset model above-mentioned performance data, it is determined whether to it is above-mentioned extremely
A few cloud main frame is alerted;In the case of it is determined that being alerted, accuse corresponding with above-mentioned at least one cloud main frame is performed
Alert action;Wherein, above-mentioned preset model trains to obtain using multi-group data by machine learning, in above-mentioned multi-group data
Every group of data include:Alerted corresponding to performance data and the performance data.
Above-mentioned distributed memory system 34, for by the pretreated performance data of above-mentioned load-balanced server store to
In HBase databases.
For ease of understanding above-described embodiment, the structure of above-mentioned communication system is described in detail below in conjunction with Fig. 4.Fig. 4 is basis
A kind of structural representation of optional communication system of the embodiment of the present invention.As shown in figure 4, the communication system includes:
Multiple cloud main frames 40, load equalizer 42, Mongo database brokers module 44, Mongo databases 46, Spark is deposited
Storage system 48, Spark performance analysers 50, alarm module 52.Wherein, cloud platform
Wherein, the running example of cloud main frame 40 agency (Instance Agent) writes in real time by cloud host monitor performance information
Enter cache;
Load equalizer 42, the number forwarded for receiving the agent data run on cloud main frame (Data Agent)
According to, carry out arrangement merging, selected through overload, select a suitable forwarding queue, be transmitted to Spark data-storage systems
48。
Mongo database brokers module 44, for the performance data of reception to be stored to Mongo databases 46;
Mongo databases 46, for storage performance data;
Spark data-storage systems 48, by way of Dstream, store data into HBase.
Spark performance analysers 50, cloud host performance data analysis program is performed, by data-analyzing machine, will be dispersed into
Two bursts of data flows.a:Real-time stream, intermediate data buffer of the real-time stream Jing Guo 4M sizes, is moved in real time for Web page
State checks current slot or each cloud host computer performance monitoring running status in time point;b:Sample data stream, it is fixed
Interval data puts down weary sampling, and peculiar discrete point is peeled off, and batch writes Mysql or MongoDB data storage mediums.Data from the sample survey
Stream checks history monitoring data for echarts and transfers to warning system to call to analyze and produce alarm.
Alarm module 52, perform alert analysis function:a:The cloud host performance data sample under normal condition is obtained, is used
Random forest or Bayesian Classification Arithmetic etc., normal data sample is trained, obtains model;b:Data to be analyzed, it is passed to corresponding
Model, show that alarm acts, perform corresponding alarm action, and correspondence database is arrived into warning information storage.
The scheme provided using the present embodiment, it is possible to achieve following effect:1:Give up the occupancy to RPC message channels, drop
The Message Processing pressure of low RPC message channels, more message handling abilities are reserved to components such as nova, cinder;2:Prison
Control data query and write-in no longer depend on MongoDB, reduce the occupancy that MongoDB is used internal memory, and it is flat to improve management
The stability of platform;3:Monitoring alarm realizes the intelligent monitoring of self study, reduces the man-machine interactively process during alarm creates,
Maloperation wrong report is reduced, reduces the pressure of system O&M;4:The scalability of cloud computing platform monitoring alarm part is added, is dropped
The coupling of the low module and whole platform;5:On the basis of original function, the quick prison in real time of cloud host data is realized
Control function;6:The standardization of data format is realized, reduces the difficulty of Reports module exploitation;7:Support thousands of cloud masters
Monitoring alarm under machine, support the data quick-processing under a large amount of monitoring datas.
Embodiment 3
A kind of storage medium is present embodiments provided, storage medium includes the program of storage, wherein, run time control in program
Equipment performs the alarm method of the cloud main frame in embodiment 1 where storage medium processed.
Embodiment 4
A kind of processor is present embodiments provided, the processor is used for operation program, wherein, program performs implementation when running
The alarm method of cloud main frame in example 1.
The embodiments of the present invention are for illustration only, do not represent the quality of embodiment.
In the above embodiment of the present invention, the description to each embodiment all emphasizes particularly on different fields, and does not have in some embodiment
The part of detailed description, it may refer to the associated description of other embodiment.
In several embodiments provided herein, it should be understood that disclosed technology contents, others can be passed through
Mode is realized.Wherein, device embodiment described above is only schematical, such as the division of the unit, Ke Yiwei
A kind of division of logic function, can there is an other dividing mode when actually realizing, for example, multiple units or component can combine or
Person is desirably integrated into another system, or some features can be ignored, or does not perform.Another, shown or discussed is mutual
Between coupling or direct-coupling or communication connection can be INDIRECT COUPLING or communication link by some interfaces, unit or module
Connect, can be electrical or other forms.
The unit illustrated as separating component can be or may not be physically separate, show as unit
The part shown can be or may not be physical location, you can with positioned at a place, or can also be distributed to multiple
On unit.Some or all of unit therein can be selected to realize the purpose of this embodiment scheme according to the actual needs.
In addition, each functional unit in each embodiment of the present invention can be integrated in a processing unit, can also
That unit is individually physically present, can also two or more units it is integrated in a unit.Above-mentioned integrated list
Member can both be realized in the form of hardware, can also be realized in the form of SFU software functional unit.
If the integrated unit is realized in the form of SFU software functional unit and is used as independent production marketing or use
When, it can be stored in a computer read/write memory medium.Based on such understanding, technical scheme is substantially
The part to be contributed in other words to prior art or all or part of the technical scheme can be in the form of software products
Embody, the computer software product is stored in a storage medium, including some instructions are causing a computer
Equipment (can be personal computer, server or network equipment etc.) perform each embodiment methods described of the present invention whole or
Part steps.And foregoing storage medium includes:USB flash disk, read-only storage (ROM, Read-Only Memory), arbitrary access are deposited
Reservoir (RAM, Random Access Memory), mobile hard disk, magnetic disc or CD etc. are various can be with store program codes
Medium.
Described above is only the preferred embodiment of the present invention, it is noted that for the ordinary skill people of the art
For member, under the premise without departing from the principles of the invention, some improvements and modifications can also be made, these improvements and modifications also should
It is considered as protection scope of the present invention.
Claims (12)
- A kind of 1. alarm method of cloud main frame, it is characterised in that including:The performance data that at least one cloud main frame uploads is obtained, wherein, the performance data is used to reflect at least one cloud The running status of main frame;The performance data is analyzed using preset model, it is determined whether at least one cloud main frame is alerted, Wherein, the preset model trains to obtain using multi-group data by machine learning, every group of number in the multi-group data According to including:Alerted corresponding to performance data and the performance data;In the case of it is determined that being alerted, alarm action corresponding with least one cloud main frame is performed.
- 2. according to the method for claim 1, it is characterised in thatThe performance data that at least one cloud main frame uploads is obtained, including:Team is cached from corresponding with least one cloud main frame The performance data is obtained in row;Methods described also includes:Detect whether the buffer queue is empty every preset time period, detecting the caching team When being classified as the number arrival predetermined threshold value of sky, then the queue is removed.
- 3. according to the method for claim 2, it is characterised in that in the buffer queue and at least one cloud main frame Cloud main frame is one-to-one.
- 4. according to the method for claim 1, it is characterised in that obtain the performance data bag that at least one cloud main frame uploads Include:The performance data that at least one cloud main frame is forwarded by load-balanced server is received, wherein, the load Equalization server is used to pre-process the performance data, and the pretreatment includes at least one of:Arrange and merge, count According to the normalized of structure, selection forwarding queue.
- 5. according to the method for claim 4, it is characterised in that receive at least one cloud main frame and taken by load balancing It is engaged in before the performance data of device forwarding, methods described also includes:Data cleansing is carried out to the performance data.
- 6. according to the method for claim 1, it is characterised in that obtain performance data that at least one cloud main frame uploads it Afterwards, methods described also includes:Receive the original performance data that at least one cloud main frame uploads;The original performance data is classified, obtains primary sources and secondary sources;The primary sources are stored into caching, and receive the call instruction of human-computer interaction interface, with display described at least The running status of one cloud main frame;The secondary sources are inputted into the preset model, the preset model is trained Or whether alerted according to the secondary sources.
- 7. according to the method for claim 1, it is characterised in that the mark of the cloud main frame is also carried in the performance data Know information.
- A kind of 8. communication system, it is characterised in that including:At least one cloud main frame, for uploading performance data, wherein, the performance data is used to reflect at least one cloud master The running status of machine;Load-balanced server, sent out for being pre-processed to the performance data, and by the pretreated performance data Deliver in distributed memory system, wherein, the pretreatment includes at least one of:Arrange merging, the normalizing of data structure Change processing, selection forwarding queue;Distributed memory system, for storing the pretreated performance data of the load-balanced server forwarding;Alarm server, for being analyzed using preset model the performance data, it is determined whether to described at least one Cloud main frame is alerted;In the case of it is determined that being alerted, alarm action corresponding with least one cloud main frame is performed; Wherein, the preset model trains to obtain using multi-group data by machine learning, every group of number in the multi-group data According to including:Alerted corresponding to performance data and the performance data.
- 9. system according to claim 8, it is characterised in that the distributed memory system, for the load is equal The weighing apparatus pretreated performance data of server is stored into HBase databases.
- A kind of 10. alarm device of cloud main frame, it is characterised in that including:Acquisition module, the performance data uploaded for obtaining at least one cloud main frame, wherein, the performance data is used to reflect institute State the running status of at least one cloud main frame;Determining module, for being analyzed using preset model the performance data, it is determined whether at least one cloud Main frame is alerted, wherein, the preset model trains to obtain using multi-group data by machine learning, multigroup number Every group of data in include:Alerted corresponding to performance data and the performance data;Alarm module, moved in the case of it is determined that being alerted, performing alert corresponding with least one cloud main frame Make.
- A kind of 11. storage medium, it is characterised in that the storage medium includes the program of storage, wherein, run in described program When control the storage medium where cloud main frame in equipment perform claim requirement 1 to 7 described in any one alarm method.
- A kind of 12. processor, it is characterised in that the processor is used for operation program, wherein, right of execution when described program is run Profit requires the alarm method of the cloud main frame described in any one in 1 to 7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710906291.4A CN107579858A (en) | 2017-09-28 | 2017-09-28 | The alarm method and device of cloud main frame, communication system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710906291.4A CN107579858A (en) | 2017-09-28 | 2017-09-28 | The alarm method and device of cloud main frame, communication system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107579858A true CN107579858A (en) | 2018-01-12 |
Family
ID=61040391
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710906291.4A Pending CN107579858A (en) | 2017-09-28 | 2017-09-28 | The alarm method and device of cloud main frame, communication system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107579858A (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108376103A (en) * | 2018-02-08 | 2018-08-07 | 厦门集微科技有限公司 | A kind of the equilibrium of stock control method and server of cloud platform |
CN109522095A (en) * | 2018-11-27 | 2019-03-26 | 无锡华云数据技术服务有限公司 | Cloud host abnormal failure detects recovery system, method and cloud platform |
CN109756395A (en) * | 2018-12-28 | 2019-05-14 | 易票联支付有限公司 | A kind of business datum monitoring method and system |
CN112052287A (en) * | 2020-09-02 | 2020-12-08 | 北京世纪互联宽带数据中心有限公司 | Centralized management method, device and system for data center cluster |
CN117290144A (en) * | 2023-10-12 | 2023-12-26 | 北京首都在线科技股份有限公司 | Fault processing method, device, electronic equipment and storage medium |
CN118012718A (en) * | 2024-04-02 | 2024-05-10 | 北京大道云行科技有限公司 | Real-time monitoring method for distributed storage system |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103023695A (en) * | 2012-11-28 | 2013-04-03 | 绍兴电力局 | Master station system monitoring model based on power dispatching automation |
CN104978612A (en) * | 2015-01-27 | 2015-10-14 | 厦门大学 | Distributed big data system risk predicating method based on AHP-RBF |
WO2016115280A1 (en) * | 2015-01-14 | 2016-07-21 | Virta Laboratories, Inc. | Anomaly and malware detection using side channel analysis |
CN106445639A (en) * | 2016-09-30 | 2017-02-22 | 北京奇虎科技有限公司 | Methods and devices for monitoring virtual machines |
CN106775929A (en) * | 2016-11-25 | 2017-05-31 | 中国科学院信息工程研究所 | A kind of virtual platform safety monitoring method and system |
CN106992994A (en) * | 2017-05-24 | 2017-07-28 | 腾讯科技(深圳)有限公司 | A kind of automatically-monitored method and system of cloud service |
-
2017
- 2017-09-28 CN CN201710906291.4A patent/CN107579858A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103023695A (en) * | 2012-11-28 | 2013-04-03 | 绍兴电力局 | Master station system monitoring model based on power dispatching automation |
WO2016115280A1 (en) * | 2015-01-14 | 2016-07-21 | Virta Laboratories, Inc. | Anomaly and malware detection using side channel analysis |
CN104978612A (en) * | 2015-01-27 | 2015-10-14 | 厦门大学 | Distributed big data system risk predicating method based on AHP-RBF |
CN106445639A (en) * | 2016-09-30 | 2017-02-22 | 北京奇虎科技有限公司 | Methods and devices for monitoring virtual machines |
CN106775929A (en) * | 2016-11-25 | 2017-05-31 | 中国科学院信息工程研究所 | A kind of virtual platform safety monitoring method and system |
CN106992994A (en) * | 2017-05-24 | 2017-07-28 | 腾讯科技(深圳)有限公司 | A kind of automatically-monitored method and system of cloud service |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108376103A (en) * | 2018-02-08 | 2018-08-07 | 厦门集微科技有限公司 | A kind of the equilibrium of stock control method and server of cloud platform |
CN109522095A (en) * | 2018-11-27 | 2019-03-26 | 无锡华云数据技术服务有限公司 | Cloud host abnormal failure detects recovery system, method and cloud platform |
CN109756395A (en) * | 2018-12-28 | 2019-05-14 | 易票联支付有限公司 | A kind of business datum monitoring method and system |
CN112052287A (en) * | 2020-09-02 | 2020-12-08 | 北京世纪互联宽带数据中心有限公司 | Centralized management method, device and system for data center cluster |
CN117290144A (en) * | 2023-10-12 | 2023-12-26 | 北京首都在线科技股份有限公司 | Fault processing method, device, electronic equipment and storage medium |
CN118012718A (en) * | 2024-04-02 | 2024-05-10 | 北京大道云行科技有限公司 | Real-time monitoring method for distributed storage system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107579858A (en) | The alarm method and device of cloud main frame, communication system | |
CN109347668B (en) | Training method and device for service quality assessment model | |
CN111178456A (en) | Abnormal index detection method and device, computer equipment and storage medium | |
CN109586999A (en) | A kind of container cloud platform condition monitoring early warning system, method and electronic equipment | |
CN104011719B (en) | The method and system that message is tracked and checked | |
CN106844138A (en) | O&M warning system and method | |
CN107872457B (en) | Method and system for network operation based on network flow prediction | |
CN106940677A (en) | One kind application daily record data alarm method and device | |
CN110740061B (en) | Fault early warning method and device and computer storage medium | |
CN109034580B (en) | Information system overall health degree evaluation method based on big data analysis | |
CN110471821A (en) | Abnormal alteration detection method, server and computer readable storage medium | |
CN105871957A (en) | Monitoring framework design method, monitoring server, proxy unit and center control server | |
CN113313280B (en) | Cloud platform inspection method, electronic equipment and nonvolatile storage medium | |
CN108809720A (en) | The management method and device of alarming assignment in cloud data system | |
CN106034047B (en) | Data processing method and device | |
CN111339052A (en) | Unstructured log data processing method and device | |
CN105430052B (en) | A kind of referee method and device for group system | |
CN112769620B (en) | Network deployment method, equipment and computer readable storage medium | |
CN110414591A (en) | A kind of data processing method and equipment | |
CN116761194A (en) | Police affair cooperative communication optimization system and method in wireless communication network | |
CN112001622A (en) | Health degree evaluation method, system, equipment and storage medium of cloud virtual gateway | |
CN106161058B (en) | Alarm classification method and device | |
CN108681499B (en) | O&M monitoring method, device and computer readable storage medium | |
CN110287256A (en) | A kind of electric network data parallel processing system (PPS) and its processing method based on cloud computing | |
CN115545452A (en) | Operation and maintenance method, operation and maintenance system, equipment and storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180112 |
|
RJ01 | Rejection of invention patent application after publication |