CN105871957A

CN105871957A - Monitoring framework design method, monitoring server, proxy unit and center control server

Info

Publication number: CN105871957A
Application number: CN201510031593.2A
Authority: CN
Inventors: 徐振佳; 张丹枫; 陈杰; 冯亮
Original assignee: Shenzhen Tencent Computer Systems Co Ltd
Current assignee: Shenzhen Tencent Computer Systems Co Ltd
Priority date: 2015-01-21
Filing date: 2015-01-21
Publication date: 2016-08-17
Anticipated expiration: 2035-01-21
Also published as: CN105871957B

Abstract

The invention discloses a monitoring framework design method, a monitoring server, a proxy unit and a center control server which are used for realizing a three-dimensional all-round monitoring system. The monitoring framework design method provided by the invention includes the following steps that: the monitoring server obtains monitoring data reported by proxy units in virtual machines and obtains monitoring data exist when the center control server controls the virtual machines; the monitoring server stores the obtained monitoring data in a database; and the monitoring server reads the monitoring data from the database, performs abnormality analysis on the monitoring data, performs warning against monitoring data where abnormalities exist and stores the monitoring data where abnormalities exist to the database.

Description

Monitoring framework method for designing and monitoring server, agent unit, control server

Technical field

The present invention relates to field of computer technology, particularly relate to a kind of Monitoring framework method for designing and monitoring clothes Business device, agent unit, control server.

Background technology

Cloud computing Infrastructure platform is a complicated service platform, has variation, isomerism and moves The feature of state change.The support of the properly functioning too busy to get away cloud monitoring system of cloud computing system, cloud monitoring is System can reflect the operation conditions of cloud platform in real time, it is possible to finds and process own of cloud computing platform in time Raw and potential problem, this serves critical effect for management and scheduling cloud computing system resource. Therefore, how to design Monitoring framework and conclusive effect is played for the normal O&M of cloud computing system.Existing Have in technology and the demand how designing Monitoring framework and could meeting cloud computing system is the most definitely specified.

Summary of the invention

Embodiments provide a kind of Monitoring framework method for designing and monitoring server, agent unit, Control server, is used for realizing the omnibearing monitoring system of three-dimensional.

For solving above-mentioned technical problem, embodiment of the present invention offer techniques below scheme:

First aspect, the embodiment of the present invention provides a kind of Monitoring framework method for designing, including:

Monitoring server obtains the monitoring data that in each virtual machine, agent unit reports respectively, therefrom controls clothes Business device obtains the monitoring data existed when each virtual machine described is controlled by described control server；

Described monitoring server is by the supervising data storage that gets to data base；

Described monitoring server reads monitoring data from described data base and carries out anomaly analysis, different to existing Normal monitoring data alert, and by supervising data storage abnormal for described existence to described data base.

Second aspect, the embodiment of the present invention also provides for a kind of Monitoring framework method for designing, including:

Agent unit monitoring virtual machine, produces monitoring data according to the service data of described virtual machine, described Agent unit is deployed in described virtual machine；

Described agent unit reports described monitoring data to monitoring server.

The third aspect, the embodiment of the present invention also provides for a kind of Monitoring framework method for designing, including:

Control server produces the monitoring data when being controlled each virtual machine；

Described control server sends described monitoring according to the request of monitoring server to described monitoring server Data.

Fourth aspect, the embodiment of the present invention also provides for a kind of monitoring server, including:

Collector unit, for obtaining the monitoring data that in each virtual machine, agent unit reports respectively, therefrom Control server obtains the monitoring data existed when each virtual machine described is controlled by described control server；

Memory element, by the supervising data storage that gets to data base；

Alarm Unit, carries out anomaly analysis for reading monitoring data from described data base, different to existing Normal monitoring data alert；

Described memory element, is additionally operable in supervising data storage abnormal for described existence to described data base.

5th aspect, the embodiment of the present invention also provides for a kind of agent unit, and described agent unit is deployed in institute State in virtual machine, including: monitoring subelement and transmission subelement, wherein,

Described monitoring subelement, is used for monitoring virtual machine, produces prison according to the service data of described virtual machine Control data,

Described transmission subelement, for reporting described monitoring data to monitoring server.

6th aspect, the embodiment of the present invention also provides for a kind of control server, including: data generating unit And transmitting element, wherein,

Described data generating unit, for producing the monitoring data when being controlled each virtual machine；

Described transmitting element, sends described for the request according to monitoring server to described monitoring server Monitoring data.

As can be seen from the above technical solutions, the embodiment of the present invention has the advantage that

In embodiments of the present invention, during monitoring server obtains each virtual machine, agent unit reports respectively Monitoring data, obtain, from control server, the monitoring number existed when each virtual machine is controlled by control server According to, monitoring server by the supervising data storage that gets to data base, monitoring server is from data base The middle monitoring data that read carry out anomaly analysis, alert there are abnormal monitoring data, and will exist Abnormal supervising data storage is in data base.Source due to the monitoring data that monitoring server gets Including each virtual machine and control server that virtual machine is controlled, the monitoring face of monitoring server Wider, cover virtual machine self and the control server that each virtual machine is controlled, based on The anomaly analysis that the monitoring data collected from virtual machine and control server are carried out can determine product accurately Raw abnormal monitoring data, and abnormal monitoring data are alerted, it is achieved that three-dimensional is omnibearing Monitoring system.

Accompanying drawing explanation

For the technical scheme being illustrated more clearly that in the embodiment of the present invention, in embodiment being described below The required accompanying drawing used is briefly described, it should be apparent that, the accompanying drawing in describing below is only this Some embodiments of invention, to those skilled in the art, it is also possible to obtain according to these accompanying drawings Other accompanying drawing.

The process blocks schematic diagram of a kind of Monitoring framework method for designing that Fig. 1 provides for the embodiment of the present invention；

The process blocks schematic diagram of the another kind of Monitoring framework method for designing that Fig. 2 provides for the embodiment of the present invention；

The process blocks schematic diagram of the another kind of Monitoring framework method for designing that Fig. 3 provides for the embodiment of the present invention；

The design structural representation of a kind of three-dimensional Monitoring framework that Fig. 4-a provides for the embodiment of the present invention；

Handling process schematic diagram to monitoring data in the Monitoring framework that Fig. 4-b provides for the embodiment of the present invention；

The composition structural representation of a kind of monitoring server that Fig. 5-a provides for the embodiment of the present invention；

The composition structural representation of the another kind of monitoring server that Fig. 5-b provides for the embodiment of the present invention；

The composition structural representation of a kind of collector unit that Fig. 5-c provides for the embodiment of the present invention；

The composition structural representation of a kind of agent unit that Fig. 6 provides for the embodiment of the present invention；

The composition structural representation of a kind of control server that Fig. 7 provides for the embodiment of the present invention；

Fig. 8 is applied to the composition of monitoring server for the Monitoring framework method for designing that the embodiment of the present invention provides Structural representation；

Fig. 9 is applied to the composition structure of virtual machine for the Monitoring framework method for designing that the embodiment of the present invention provides Schematic diagram；

Figure 10 is applied to the composition of control server for the Monitoring framework method for designing that the embodiment of the present invention provides Structural representation.

Detailed description of the invention

For making the goal of the invention of the present invention, feature, the advantage can be the most obvious and understandable, below will In conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete Ground describes, it is clear that the embodiments described below are only a part of embodiment of the present invention, and not all Embodiment.Based on the embodiment in the present invention, the every other enforcement that those skilled in the art is obtained Example, broadly falls into the scope of protection of the invention.

Term in description and claims of this specification and above-mentioned accompanying drawing " includes " and " having " And their any deformation, it is intended that cover non-exclusive comprising, in order to comprise a series of unit Process, method, system, product or equipment are not necessarily limited to those unit, but can include the most clearly That list or for intrinsic other unit of these processes, method, product or equipment.

It is described in detail individually below.

In Monitoring framework method for designing of the present invention, monitoring server is used for realizing Monitoring framework, monitoring service Device and each virtual machine and the control server being controlled each virtual machine have all set up communication link Connecing, the fault occurred, by the monitoring to virtual machine and control server, can be had by monitoring server Alarm accurately and timely, thus facilitate operation maintenance personnel quickly to position solution problem, it is ensured that QoS of customer. The Monitoring framework method for designing that the embodiment of the present invention provides can realize the omnibearing monitoring system of three-dimensional.Connect Lower real to each equipment in monitoring server, virtual machine, control server execution Monitoring framework design respectively Existing method is illustrated.

One embodiment of Monitoring framework method for designing of the present invention, specifically can apply in monitoring server, Refer to shown in Fig. 1, the control frame design method that one embodiment of the invention provides, can include as follows Step:

101, the monitoring data that during monitoring server obtains each virtual machine, agent unit reports respectively, therefrom Control server obtains the monitoring data existed when each virtual machine is controlled by control server.

In embodiments of the present invention, in order to enable monitoring server determination fault accurately and timely and carry out Alarm, disposes agent unit in each virtual machine respectively, agent unit collect the operation number of virtual machine According to and generate the monitoring data that each virtual machine is monitored, virtual machine is being controlled by control server Also produce monitoring data time processed, the Monitoring framework of the present invention needs to configure in virtual machine agent unit to prison Control server reports monitoring data, it is also desirable to configuration monitoring server therefrom controls server request monitoring data, So monitoring server needs to obtain the monitoring data that in each virtual machine, agent unit reports respectively, therefrom Control server obtains the monitoring data existed when each virtual machine is controlled by control server, due to monitoring clothes The source of the monitoring data that business device gets includes each virtual machine and the middle control being controlled virtual machine Server, the monitoring face of monitoring server is wider, has covered virtual machine self and to each virtual machine The control server being controlled.

Further, in order to realize finer monitoring service, in some embodiments of the invention, The monitoring data that during monitoring server obtains each virtual machine in step 101, agent unit reports respectively, tool Body may include steps of:

A1, monitoring server obtain in each virtual machine agent unit according to each preset monitored item and The monitoring data of collection cycle real-time report.

It is to say, so that monitoring server can more finely provide monitoring service, in addition it is also necessary to right The agent unit being deployed in virtual machine carries out pre-setting configuration file, and agent unit is according to configuration file Requirement virtual machine is monitored, such as arrange in configuration file need monitoring monitored item and set Put the collection cycle of agent unit, then agent unit is accomplished by the monitored item pair arranged according to configuration file Every service data of virtual machine monitors the most in real time, according to configuration file arrange the collection cycle to Monitoring server reports the monitoring data of many groups.Concrete, pre-in the configuration file that agent unit is arranged The monitored item put comprises the steps that system layer and the process level of virtual machine, i.e. agent unit collect this agent unit System layer data in the virtual machine at place and process level data, agent unit reports void to monitoring server The system layer data of plan machine and process level data.

It should be noted that the system layer of virtual machine refers to the bottom hardware aspect of virtual machine, system layer Data refer to the monitoring data that the underlying infrastructure class of virtual machine is monitored obtain, including but not The situation being confined to illustrate as follows broadly falls into system layer data: the network bandwidth takies situation, the load of bag amount, Physical content amount, virutal machine memory capacity, network interface card speed, resolve dmesg information, to various system heaps Stack, input/output port (Input/Output, I/O).The process level of virtual machine refers on virtual machine Critical processes aspect, process level data refer to supervise the running status of the critical processes on virtual machine The monitoring data that control obtains, the situation including, but not limited to following citing broadly falls into process level data: in Central processor (Central Processing Unit, CPU) utilization rate situation, EMS memory occupation situation, disk Service condition.

Further, step 101 obtains control server from control server each virtual machine is controlled Time exist monitoring data, specifically may include steps of:

A2, monitoring server from control server obtain control server according to each preset monitored item with And the monitoring data that the collection cycle sends in real time.

It is to say, so that monitoring server can more finely provide monitoring service, in addition it is also necessary to right Monitoring server carries out pre-setting configuration file, monitoring server according to the requirement of configuration file to middle control Server request monitoring data, such as, arrange in configuration file and need the monitored item of monitoring and arrange receipts The collection cycle, then monitoring server is accomplished by the monitored item according to configuration file setting and the collection cycle is fixed Time to control server request monitoring data, concrete, to monitoring server arrange configuration file in Preset monitored item comprises the steps that the module layer in control server, podium level and client layer, i.e. monitors clothes Business device needs therefrom to control server pull module layer data, podium level data and client layer data, and middle control takes Be engaged in the device request according to monitoring server to monitoring server feedback module layer data, podium level data and use Family layer data.

It should be noted that can dispose virtual platform in control server, virtual platform can realize money Distributing, reclaim, circulate in source, disposes this virtual platform and can realize the concentration to resource in control server Management, additionally control server disposes multiple modules also by virtual platform, and module is to realize virtualization Each ingredient of platform, by each module distribution completing resource alternately etc., in control server Podium level refer to virtual platform aspect, it is flat to virtualization that podium level data refer to control server Platform is monitored the monitoring data obtained, and the situation including, but not limited to following citing broadly falls into podium level Data: the stream compression within virtual platform and application programming interface (Application Programming Interface, API) call.It is flat that module layer in control server refers to virtualization The modules aspect that platform includes, module layer data refer to control server to each in virtual platform Individual module is monitored the monitoring data obtained, and the situation including, but not limited to following citing broadly falls into mould Block layer data: the situations such as the state of the internal correlation module of virtual platform, time-consuming, survival.Middle control service Client layer in device refers to user level when control server uses virtual machine to user, user's number of plies According to referring to the monitoring data that user is used virtual machine to be monitored obtaining by control server, including but not The situation being confined to illustrate as follows broadly falls into client layer data: user uses the CPU usage of virtual machine, User uses the I/O of virtual machine, and user uses the Internet Use of virtual machine.

In some embodiments of the invention, need the monitored item reported different for agent unit, step 101 monitoring servers obtain the monitoring data that in each virtual machine, agent unit reports respectively, specifically include as Lower step:

B1, monitoring server obtain system layer data that in each virtual machine, agent unit reports respectively and enter Journey layer data.

Wherein, the configuration file arranged in agent unit requires that the monitored item that agent unit reports includes: be System layer and process level, such as, be set to each monitored item according to different monitoring dimension and arrange globally unique prison Control (Identity, ID), then system layer monitored item and process level monitored item use different monitoring ID respectively, Agent unit uses the monitoring ID of system layer to identify system layer data, uses the monitoring ID mark of process level Process level data, system layer data that monitoring server Receiving Agent unit reports and process level data, root System layer data and process level data are identified according to the monitoring ID of system layer and the monitoring ID of process level.

Further, perform step B1 realize under scene, step B1 monitoring server obtains each System layer data that in virtual machine, agent unit reports respectively and process level data, specifically can include as follows Step:

Monitoring server arranges systematic collection unit and process collector unit, systematic collection unit, is used for Receive the system layer data that in each virtual machine, agent unit reports respectively；Process collector unit, is used for connecing Receive the process level data that in each virtual machine, agent unit reports respectively.

It is to say, implementing aspect, monitoring server is required for the monitored item design of each layer Respective collector unit, arranges systematic collection unit for system layer, arranges process for process level and collects Unit, to ensure motility, it is single that systematic collection unit and process collector unit will receive each agency respectively The monitoring data that unit reports.

In some embodiments of the invention, need the monitored item reported different for agent unit, step 101 obtain, from control server, the monitoring data existed when each virtual machine is controlled by control server, specifically Comprise the steps:

B2, monitoring server obtain control server from control server and control each by virtual platform Module layer data, podium level data and the client layer data existed during virtual machine, virtual platform is disposed Multiple module, client layer data is had to include: the virtual machine service data that user exists when using virtual machine.

Wherein, the configuration file arranged in monitoring server requires that monitoring server obtains from control server Monitored item include: module layer, podium level and client layer, be such as set to often according to different monitoring dimension Individual monitored item arranges globally unique monitoring ID, then module layer monitored item, podium level monitored item and client layer Monitored item uses different monitoring ID respectively, and monitoring server uses the monitoring ID of module layer to identify module layer Data, use the monitoring ID label platform layer data of podium level, use the monitoring ID mark of client layer to use Family layer data, monitoring server is obtained control server from control server and is controlled by virtual platform each Module layer data, podium level data and the client layer data existed during individual virtual machine, according to the prison of module layer The monitoring ID of control ID, the monitoring ID of podium level and client layer identifies module layer data, podium level data With client layer data.

Further, perform step B2 realize under scene, step B2 monitoring server therefrom controls clothes Business device obtain the module layer data of existence when control server controls each virtual machine by virtual platform, Podium level data and client layer data, specifically may include steps of:

Monitoring server arranges module collection unit, platform collector unit and user's collector unit, module Collector unit, controls each for therefrom control server pull control server by virtual platform virtual The module layer data produced during machine；Platform collector unit, for therefrom control server pull virtual platform Podium level data；User's collector unit, produced when therefrom control server pull user uses virtual machine Raw client layer data.

It is to say, implementing aspect, monitoring server is required for the monitored item design of each layer Respective collector unit, arranges module collection unit for module layer, arranges platform for podium level and collects Unit, arranges user's collector unit for client layer, to ensure motility, and module collection unit, platform Collector unit and user's collector unit will receive the monitoring data that control server returns respectively.

102, monitoring server is by the supervising data storage that gets to data base.

In embodiments of the present invention, monitoring server gets monitoring data from agent unit and therefrom controls After server gets monitoring data, the monitoring data that monitoring server gets just include: come from In each virtual machine agent unit send monitoring data and control server send monitoring data, monitor Server is deployed with data base (Data Base, DB), the monitoring data that monitoring server will get All storing in data base, in data base, storage has the monitoring data that monitoring server gets, these prisons Control data can be transferred by monitoring server when needed.

In some embodiments of invention, for performing the application scenarios of abovementioned steps B1, step 102 is supervised Control server, by the supervising data storage that gets to data base, specifically may include steps of:

Arranging system data word bank and process data word bank in monitoring server, monitoring server is by system layer Data store in system data word bank, process level data are stored in process data word bank.

Same, for performing the application scenarios of abovementioned steps B2, step 102 monitoring server will obtain To supervising data storage in data base, including:

Monitoring server arranges module data word bank, platform data word bank and user data word bank, monitoring Module layer data are stored in module data word bank by server, and podium level data are stored platform data In word bank, client layer data are stored in user data word bank.

It is to say, data base can store monitoring number respectively according to the fine degree that monitored item is arranged According to, it is also carried out marking off five memory spaces according to monitored item, for storage monitoring data in data base Monitoring ID be expressed as: system data word bank, process data word bank, module data word bank, platform data Word bank and user data word bank, the monitoring data of each word bank storage this word bank corresponding, convenient monitoring service The device classification storage to monitoring data, if monitoring server needs the monitoring calling each monitored item, can Realize efficiently.Fast.

103, monitoring server reads monitoring data from data base and carries out anomaly analysis, abnormal to existing Monitoring data alert, and by supervising data storage abnormal for existence to data base.

In embodiments of the present invention, monitoring server is from the agent unit of each virtual machine and control server The monitoring data collected all leave in data base, and monitoring server can be real-time from this data base Read out monitoring data, then the monitoring data read are carried out anomaly analysis, so that it is determined that go out to exist Abnormal monitoring data alert, it should be noted that monitoring server determines whether monitoring data produce Life is abnormal to be needed to combine concrete application scenarios and operation maintenance personnel concrete prison available to monitoring server Control service is relevant, under different scenes, the anomaly analysis of monitoring data can be used concrete data analysis Mode, additionally monitoring server carries out alarm to the monitoring data that there is exception can also multiple realization side The information instruction that formula, such as priority are high, or use animation, audio and specific program all can make For the mode of alarm, concrete alarm mode can be indicated by being pre-configured with Alarm ID, not limit Fixed.Additionally monitoring server is after analyzing the monitoring data that there is exception, in addition it is also necessary to exist different by this Normal monitoring data store, and are stored in monitoring server in the data base disposed, Ke Yili Solving, data base states monitoring data that in step 102, monitoring server gets before storing and deposits When being stored in abnormal monitoring data, it is only necessary to use different storage concordance lists, two class monitoring numbers According to being stored separately, in order to the subsequent calls of monitoring server.

In some embodiments of the invention, abnormal supervising data storage will be there is to number in step 103 After in storehouse, the Monitoring framework method for designing that the present invention provides can also comprise the steps:

C1, monitoring server extract from data base and there are abnormal monitoring data, and outwards show with Just accident analysis is carried out.

It is to say, monitoring server is after determining the monitoring data that there is exception and alerting, for side Just operation maintenance personnel carries out the quick positioning analysis of fault, and monitoring server also needs to extract from data base deposit In abnormal monitoring data, and outwards showing to carry out accident analysis, such as fortune monitoring server is permissible By OSS (Operation Support System, OSS) to operation maintenance personnel output abnormality Monitoring data, are quickly positioned solution problem, it is ensured that QoS of customer by operation maintenance personnel.Concrete, can To dispose display unit in monitoring server, display unit there is abnormal monitoring number with extraneous output According to.Concrete, display unit, may include that system demonstration unit, process display unit, module exhibition Showing unit, platform display unit and user's display unit, corresponding display unit is for realizing corresponding prison The displaying of control item.

In some embodiments of the invention, for performing abovementioned steps A1 and the application scenarios of A2, step In rapid 103 monitoring server read from data base monitoring data carry out anomaly analysis, including:

Monitoring server carries out anomaly analysis according to the many groups of monitoring data got, and determines that existence is abnormal Monitored item.

Further, realizing under scene aforesaid, step 103 is carried out there are abnormal monitoring data Alarm, specifically may include steps of:

Monitoring server is to determining that the multiple abnormal monitoring data that there is abnormal same monitored item are gathered Close, carry out concentrating alarm according to polymerization result.

If it is to say, monitoring server receives the monitoring number of many groups from agent unit or control server According to, the horizontal analysis to many groups of monitoring data in same monitored item, it is also possible to realizing monitoring box The self-monitoring of frame, further, can arrange Alarm Unit in monitoring server, abnormal for existing Monitoring data alert, and illustrate: as a example by virtual machine card of surfing Internet packet loss, the agency on virtual machine Unit is within every 5 minutes, to gather once to report once, and monitoring server will receive agent unit in every 5 minutes The monitoring data reported, can deposit a packet loss information in every 5 minutes in that data base, if Alarm Unit Do not do if being polymerized, an alarm within every 5 minutes, can be sent, so have substantial amounts of information redundancy, if accusing Alert unit carries out polymerization and calculates alarm, by this kind of information fusion is the most disposably sent alarm, Avoid unnecessary interference., Alarm Unit is in order to suppress alarm windstorm, it is also possible to same monitored item Multiple alarms are polymerized, and carry multiple abnormal monitoring data in polymerization result, can realize concentrating alarm. Concrete, Alarm Unit, may include that the alarm of ALM unit, process Alarm Unit, module is single Unit, platform Alarm Unit and user's Alarm Unit, corresponding Alarm Unit is for realizing corresponding monitored item Alarm.

By the above example description to the embodiment of the present invention, it is virtual that monitoring server obtains each The monitoring data that in machine, agent unit reports respectively, obtain control server from control server empty to each The monitoring data existed when plan machine controls, monitoring server by the supervising data storage that gets to data base In, monitoring server reads monitoring data from data base and carries out anomaly analysis, to there is abnormal monitoring Data alert, and by supervising data storage abnormal for existence to data base.Due to monitoring server The source of the monitoring data got includes each virtual machine and the middle control service being controlled virtual machine Device, the monitoring face of monitoring server is wider, has covered virtual machine self and has carried out each virtual machine Control control server, based on the monitoring data collected from virtual machine and control server carry out different Often analyze and can determine the monitoring data producing exception accurately, and abnormal monitoring data are alerted, Achieve the omnibearing monitoring system of three-dimensional.

Illustrated from monitoring server side to Monitoring framework method for designing of the present invention above, connect down Introduce the another kind of Monitoring framework method for designing that the embodiment of the present invention provides, specifically can apply to each In the agent unit disposed in virtual machine, refer to shown in Fig. 2, the control that one embodiment of the invention provides Frame design method, may include steps of:

201, agent unit monitoring virtual machine, produces monitoring data, agency according to the service data of virtual machine Unit is deployed in virtual machine.

202, agent unit reports monitoring data to monitoring server.

In inventive embodiments, in order to enable monitoring server determination fault accurately and timely and accuse Alert, each virtual machine is disposed an agent unit respectively, agent unit collects the fortune of this virtual machine Row data also generate the monitoring data being monitored each virtual machine, and each virtual machine disposes agency Unit, is actively reported to monitoring server by agent unit, and the mode of active reporting avoids single-point to collect prison Control information, say, that the acquisition to all virtual machine informations is by disposing generation on each virtual machine That reason unit realizes rather than obtained all virtual machines by monitoring server program monitoring Data.

In some embodiments of the invention, step 201 agent unit monitoring virtual machine, according to virtual machine Service data produce monitoring data, specifically may include steps of:

Every service data of virtual machine is carried out by monitored item that agent unit is arranged according to configuration file respectively Monitoring in real time, according to the monitoring data of the many groups of cycle of collecting generation that configuration file is arranged.

It should be noted that the system layer of virtual machine refers to the bottom hardware aspect of virtual machine, system layer Data refer to the monitoring data that the underlying infrastructure class of virtual machine is monitored obtain, including but not The situation being confined to illustrate as follows broadly falls into system layer data: the network bandwidth takies situation, the load of bag amount, Physical content amount, virutal machine memory capacity, network interface card speed, resolve dmesg information, to various system heaps Stack, I/O.The process level of virtual machine refers to the critical processes aspect on virtual machine, and process level data refer to The monitoring data that are monitored obtaining of the running status to the critical processes on virtual machine, including but not office The situation being limited to illustrate as follows broadly falls into process level data: CPU usage situation, EMS memory occupation situation, Disk service condition.

Further, if supervision packet includes: system layer data and process level data, step 202 is acted on behalf of Unit reports monitoring data to monitoring server, specifically may include steps of:

The systematic collection unit that agent unit is arranged in monitoring server sends system layer data, to monitoring The process collector unit arranged in server sends process level data.

By the above example description to the embodiment of the present invention, in each virtual machine, agent unit divides The monitoring data not reported to monitoring server, monitoring server also control service from control server obtains The monitoring data existed when each virtual machine is controlled by device, the monitoring data got are deposited by monitoring server Storing up in data base, monitoring server reads monitoring data from data base and carries out anomaly analysis, to existence Abnormal monitoring data alert, and by supervising data storage abnormal for existence to data base.Due to The source of the monitoring data that monitoring server gets includes each virtual machine and is controlled virtual machine Control server, the monitoring face of monitoring server is wider, has covered virtual machine self and to each The control server that virtual machine is controlled, based on the monitoring number collected from virtual machine and control server Can determine accurately according to the anomaly analysis carried out and produce abnormal monitoring data, and to abnormal monitoring number According to alerting, it is achieved that the omnibearing monitoring system of three-dimensional.

Illustrated from monitoring server side to Monitoring framework method for designing of the present invention above, connect down Introduce the another kind of Monitoring framework method for designing that the embodiment of the present invention provides, specifically can apply to middle control In server, refer to shown in Fig. 3, the control frame design method that one embodiment of the invention provides, can To comprise the steps:

301, control server produces the monitoring data when being controlled each virtual machine.

302, control server sends monitoring data according to the request of monitoring server to monitoring server.

In inventive embodiments, control server can also be referred to as central control server, and central authorities control Server is for being controlled each virtual machine, in order to enable monitoring server determination accurately and timely Fault also alerts, and control server produces monitoring data, the present invention when being controlled virtual machine Monitoring framework in need configuration monitoring server therefrom control server request monitoring data, so monitoring clothes Business device needs to obtain the monitoring data that in each virtual machine, agent unit reports respectively, obtains from control server Take the monitoring data existed when each virtual machine is controlled by control server.Control server is to monitoring service The monitoring data existed when device provides this control server to control virtual machine so that control server is permissible Coordinating monitoring server to improve the design of Monitoring framework, the monitoring data that control server provides can meet The needs that omnibearing stereoization is monitored by monitoring server are single compared to monitoring server in prior art Monitoring Data Source have more specific aim.

In some embodiments of the invention, step 301 control server produces entering each virtual machine Monitoring data when row controls, specifically may include steps of:

Monitored item and collection cycle that control server is arranged according to configuration file produce the prison organized respectively Control data.

It should be noted that can dispose virtual platform in control server, virtual platform can realize money Distributing, reclaim, circulate in source, disposes this virtual platform and can realize the concentration to resource in control server Management, additionally control server disposes multiple modules also by virtual platform, and module is to realize virtualization Each ingredient of platform, by each module distribution completing resource alternately etc., in control server Podium level refer to virtual platform aspect, it is flat to virtualization that podium level data refer to control server Platform is monitored the monitoring data obtained, and the situation including, but not limited to following citing broadly falls into podium level Data: the stream compression within virtual platform and API.Module layer in control server refers to virtual Changing the modules aspect that platform includes, module layer data refer to control server to virtual platform Middle modules is monitored the monitoring data obtained, and the situation including, but not limited to following citing all belongs to In module layer data: situations such as the state of the internal correlation module of virtual platform, time-consuming, survivals.Middle control Client layer in server refers to user level when control server uses virtual machine to user, user Layer data refers to the monitoring data that user is used virtual machine to be monitored obtaining by control server, including But the situation being not limited to illustrate as follows broadly falls into client layer data: user uses the CPU of virtual machine to use Rate, user uses the I/O of virtual machine, and user uses the Internet Use of virtual machine.

Further, if supervision packet includes: module layer data, podium level data and client layer data, Step 302 control server sends monitoring data, tool according to the request of monitoring server to monitoring server Body can comprise the steps:

The module collection unit sending module layer data that control server is arranged in monitoring server, to prison The platform collector unit arranged in control server sends podium level data, the use arranged in monitoring server Family collector unit sends client layer data.

By the above example description to the embodiment of the present invention, control server is to monitoring server Sending the monitoring data existed when each virtual machine is controlled by control server, monitoring server also obtains respectively The monitoring data that in individual virtual machine, agent unit reports respectively, the monitoring data that monitoring server will get Storing in data base, monitoring server reads monitoring data from data base and carries out anomaly analysis, to depositing Alert in abnormal monitoring data, and by supervising data storage abnormal for existence to data base.By The source of the monitoring data got in monitoring server includes each virtual machine and controls virtual machine The control server of system, the monitoring face of monitoring server is wider, has covered virtual machine self and to respectively The control server that individual virtual machine is controlled, based on the monitoring collected from virtual machine and control server The anomaly analysis that data are carried out can determine the monitoring data producing exception accurately, and to abnormal monitoring Data alert, it is achieved that the omnibearing monitoring system of three-dimensional.

For ease of being better understood from and implement the such scheme of the embodiment of the present invention, citing below accordingly should It is specifically described by scene.

The Monitoring framework method for designing that the present invention provides can realize the omnibearing monitoring system of three-dimensional, such as The cloud computing system of (Infrasturcture-as-a-Service, IaaS) is i.e. serviced, to cloud meter in infrastructure The stability of calculation system has strict requirements, can realize the tight prison to cloud computing system in conjunction with the present invention Control, and the fault occurred is had alarm accurately and timely, thus facilitate operation maintenance personnel quickly to position solution and ask Topic, it is ensured that QoS of customer.

First the three-dimensional monitoring scheme of the present invention is formed structure to illustrate, refer to such as Fig. 4-a institute Show, for the design structural representation of a kind of three-dimensional Monitoring framework that the embodiment of the present invention provides, three-dimensional Monitored item is subdivided into five aspects and realizes the monitor closely to cloud computing system by monitoring scheme, is to use respectively Family layer, podium level, module layer, process level and system layer.Wherein, these five layers of monitored item can realize void The conduct monitoring at all levels of planization.

Next the agent unit from monitoring server, control server, virtual machine from Monitoring framework The handling process of monitoring data is illustrated, refers to shown in Fig. 4-b, provide for the embodiment of the present invention To the handling process schematic diagram monitoring data in Monitoring framework.Monitoring server is divided into four primary layer Secondary: collector unit, data base, Alarm Unit and display unit.

1) agent unit, in virtual machine is mainly responsible for collecting local system layer and the key message of process level, And reporting corresponding Centroid, such as system layer data report systematic collection unit, the process number of plies According to reporting process collector unit.

Such as, disposing monitoring programme (i.e. agent unit) on all virtual machines, agent unit registers each prison The monitoring ID of control item, and the collecting and reporting frequency of each monitoring ID is determined according to configuration file.

2), systematic collection unit and process collector unit have been responsible for the system reported of Receiving Agent unit Layer data and process level data, module collection unit, platform collector unit and user's collector unit are therefrom controlled Server pull module layer data, podium level data and client layer data.

Concrete, the different monitoring item of five layers of monitoring dimension is arranged globally unique monitoring ID, it is right to accelerate The fault location of system.Agent unit and collector unit may utilize APScheduler and realize timed task, APScheduler is a python framework based on Quartz, it is provided that based on the date, fixing time Between interval and the task of crontab type, and with persistence task, thus acquisition monitoring number can be realized According to motility.

For data base, two kinds of data bases can be used: memory database redis and mysql, collector unit The monitoring Refresh Data in real time agent unit reported in redis, redis be one increase income based on The NoSQL data base of key-value, has high-performance and high availability, can tackle height and concurrently access. Redis data are regularly imported in mysql by collector unit, and mysql deposits all of metadata and prison Control data, and ensure that monitoring data is continuous print, utilizes the horizontal cluster expansion of HAProxy simultaneously, breaks through The mysql of unit reads and writes bottleneck, it is ensured that the stability of system.

3), Alarm Unit from each data word bank, extract monitoring data, determine and there is abnormal prison Control data, and carry out being polymerized calculating to monitoring data, realize flexible, various ways according to alarm item configuration Alarm.Such as, Alarm Unit increases the analysis result of dmesg in warning content, and fast and easy is fixed Position and analysis fault.Wherein, there are abnormal monitoring data can be client layer data, process level data With system layer data, it is unavailable that the exception of these monitoring data is likely to result in virtual machine, if the platform number of plies Extremely user can be made cannot virtual machine to be operated according to existing with module layer data, such as shutdown, restart. Such as, operation and maintenance tools use tornado web server framework to realize, and tornado has non-obstruction, speed Fast feature, in that context it may be convenient to increase and delete function, thus before and after realizing OSS, end separates, it is ensured that be The extensibility of system and motility.

4), display unit obtains monitoring data from the data word bank of each monitored item in real time, it is provided that to cloud meter The condition monitoring that calculation system is real-time, and multiple operation and maintenance tools are provided, facilitate operation maintenance personnel quickly to operate, Repair fault, analyze reason.Wherein, operation and maintenance tools can be to troubleshooting stream inside cloud computing system The instrument of journey, such as, create a Trouble ticket, record fault and follow the tracks of the solution situation of fault.

By above illustration, the monitoring of Monitoring framework method for designing of the present invention is wide general, from The bottom hardware of system, critical processes, console module waits until that top application covers comprehensively, monitors number According in continuum, make the analysis to monitoring data very convenient, and there is single-point, expansible, from I monitors, alarm feature promptly and accurately.Globally unique monitoring ID is set, can be square by monitoring ID Just earth fault type, increases the dmesg information that resolves, acceleration disturbance location and analysis in warning content.Generation Reason unit is deployed on all virtual machines, and the mode of active reporting avoids single-point collection monitoring information.Each prison Control layer connects respective collector unit, makes the function of collector unit simplify and conveniently removes, adds and upgrade, Favorable expandability.Real-time continuous print monitoring data being analyzed, according to the monitoring of monitored item of Alarm Unit ID is polymerized warning information, suppresses alarm windstorm.It addition, Monitoring framework design is upper requires that monitoring data are even Continuing, the information that no matter monitors is required for reporting monitored results the most extremely, monitors number accordingly if lacked According to, it may be determined that the virtual machine at agent unit place exists abnormal, thus realizes self-monitoring.

It should be noted that for aforesaid each method embodiment, in order to be briefly described, therefore by its all table Stating as a series of combination of actions, but those skilled in the art should know, the present invention is by being retouched The restriction of the sequence of movement stated because according to the present invention, some step can use other orders or Carry out simultaneously.Secondly, those skilled in the art also should know, embodiment described in this description Belong to preferred embodiment, necessary to involved action and the module not necessarily present invention.

For ease of preferably implementing the such scheme of the embodiment of the present invention, it is also provided below for implementing State the relevant apparatus of scheme.

Refer to shown in Fig. 5-a, a kind of monitoring server 500 that the embodiment of the present invention provides, may include that Collector unit 501, memory element 502, Alarm Unit 503, wherein,

Collector unit 501, for obtaining the monitoring data that in each virtual machine, agent unit reports respectively, from Control server obtains the monitoring data existed when each virtual machine described is controlled by described control server；

Memory element 502, by the supervising data storage that gets to data base；

Alarm Unit 503, carries out anomaly analysis, to existence for reading monitoring data from described data base Abnormal monitoring data alert；

Described memory element 502, is additionally operable to there is abnormal supervising data storage to described data base by described In.

In some embodiments of the invention, refer to as shown in Fig. 5-b, relative to as shown in Fig. 5-a Monitoring server, monitoring server 500, also include: display unit 504, for described memory element 502 After in supervising data storage abnormal for described existence to described data base, extract from described data base Go out and there are abnormal monitoring data, and outwards show to carry out accident analysis.

In some embodiments of the invention, described collector unit 501, specifically for obtaining each virtual machine Middle agent unit is according to each preset monitored item and the monitoring data of collection cycle real-time report；Therefrom Control server obtains described control server and sends in real time according to each preset monitored item and collection cycle Monitoring data.

In some embodiments of the invention, described Alarm Unit 503, specifically for many according to getting Group monitoring data carry out anomaly analysis, determine and there is abnormal monitored item；The same of exception is there is to determining Multiple abnormal monitoring data of one monitored item are polymerized, and carry out concentrating alarm according to polymerization result.

In some embodiments of the invention, refer to as shown in Fig. 5-c, described collector unit 501, bag Include: systematic collection unit 5011, process collector unit 5012, module collection unit 5013, platform are collected Unit 5014 and user's collector unit 5015, wherein,

Described systematic collection unit 5011, is used for receiving agent unit in each virtual machine described and reports respectively System layer data；

Described process collector unit 5012, is used for receiving agent unit in each virtual machine described and reports respectively Process level data.

Described module collection unit 5013, leads to for pulling described control server from described control server Cross the module layer data that virtual platform controls to produce during each virtual machine described；

Described platform collector unit 5014, for pulling described virtual platform from described control server Podium level data；

Described user's collector unit 5015, for from described control server pull user use virtual machine time The client layer data produced；Multiple module, described client layer packet it is deployed with on described virtual platform Include: the virtual machine service data that user exists when using virtual machine.

In some embodiments of the invention, described monitoring server arranges system data word bank and process Data word bank, described memory module 502, specifically for storing described system number by described system layer data According in word bank, described process level data are stored in described process data word bank；

Described monitoring server arranges module data word bank, platform data word bank and user data word bank, Described memory module 502, specifically for described module layer data are stored in described module data word bank, Described podium level data are stored in described platform data word bank, described client layer data are stored institute State in user data word bank.

Refer to shown in Fig. 6, a kind of agent unit 600 that the embodiment of the present invention provides, described agent unit It is deployed in described virtual machine, may include that monitoring subelement 601 and send subelement 602, wherein,

Described monitoring subelement 601, is used for monitoring virtual machine, produces according to the service data of described virtual machine Monitoring data,

Described transmission subelement 602, for reporting described monitoring data to monitoring server.

In some embodiments of the invention, described monitoring subelement 601, specifically for according to configuration file Every service data of described virtual machine is monitored by monitored item the most in real time that arrange, joins according to described Put the monitoring data of the many groups of collection cycle generation that file is arranged.

In some embodiments of the invention, described monitoring data, including: system layer data and process level Data；Described transmission subelement 602, specifically for the systematic collection list arranged in described monitoring server Unit sends described system layer data, and the process collector unit arranged in described monitoring server sends described Process level data.

Refer to shown in Fig. 7, a kind of control server 700 that the embodiment of the present invention provides, may include that Data generating unit 701 and transmitting element 702, wherein,

Described data generating unit 701, for producing the monitoring number when being controlled each virtual machine According to；

Described transmitting element 702, sends institute for the request according to monitoring server to described monitoring server State monitoring data.

In some embodiments of the invention, described data generating unit 701, specifically for according to configuration literary composition Monitored item and collection cycle that part is arranged produce the monitoring data organized respectively.

In some embodiments of the invention, described monitoring data, including: module layer data, podium level Data and client layer data；Described transmitting element 702, specifically for arrange in described monitoring server Module collection unit sends described module layer data, and the platform arranged in described monitoring server is collected single Unit sends described podium level data, and the user's collector unit arranged in described monitoring server sends described Client layer data.

Fig. 8 is a kind of monitoring server structural representation that the embodiment of the present invention provides, this monitoring server 800 can produce bigger difference because of configuration or performance difference, can include one or more central authorities Processor (central processing units, CPU) 822 (such as, one or more processors) With memorizer 832, one or more storage application program 842 or storage mediums 830 of data 844 (such as one or more mass memory units).Wherein, memorizer 832 and storage medium 830 can To be of short duration storage or persistently to store.The program being stored in storage medium 830 can include one or one With upper module (diagram does not marks), each module can include a series of command operatings in server. Further, central processing unit 822 could be arranged to communicate with storage medium 830, in monitoring service The a series of command operatings in storage medium 830 are performed on device 800.

Monitoring server 800 can also include one or more power supplys 826, and one or more have Line or radio network interface 850, one or more input/output interfaces 858, and/or, one or one Individual above operating system 841, such as Windows ServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM etc..

Can be based on this Fig. 5-a, Fig. 5-b, figure by the step performed by monitoring server in above-described embodiment Monitoring server structure shown in 5-c.

Fig. 9 is a kind of virtual Machine Architecture schematic diagram that the embodiment of the present invention provides, and this virtual machine 900 can be because of Configuration or performance are different and produce bigger difference, can include one or more central processing units (central processing units, CPU) 922 (such as, one or more processors) and storage Device 932, the storage medium 930 of one or more storage application programs 942 or data 944 is (such as One or more mass memory units).Wherein, memorizer 932 and storage medium 930 can be short Keep in storage or persistently store.The program being stored in storage medium 930 can include one or more moulds Block (diagram does not marks), each module can include a series of command operatings in server.More enter one Step ground, central processing unit 922 could be arranged to communicate with storage medium 930, holds on virtual machine 900 A series of command operatings in row storage medium 930.

Virtual machine 900 can also include one or more power supplys 926, one or more wired or Radio network interface 950, one or more input/output interfaces 958, and/or, one or one with Upper operating system 941, such as Windows ServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM etc..

Above-described embodiment can be tied based on the agent unit shown in this Fig. 6 by the step performed by virtual machine Structure.

Figure 10 is a kind of control server structural representation that the embodiment of the present invention provides, this control server 1000 can produce bigger difference because of configuration or performance difference, can include in one or more (such as, one or more process central processor (central processing units, CPU) 1022 Device) and memorizer 1032, one or more storage application program 1042 or storages of data 1044 Medium 1030 (such as one or more mass memory units).Wherein, memorizer 1032 and storage Medium 1030 can be of short duration storage or persistently store.The program being stored in storage medium 1030 can be wrapped Include one or more modules (diagram do not mark), each module can include in server be Row command operating.Further, central processing unit 1022 could be arranged to lead to storage medium 1030 Letter, performs a series of command operatings in storage medium 1030 in control server 1000.

Control server 1000 can also include one or more power supplys 1026, one or more Wired or wireless network interface 1050, one or more input/output interfaces 1058, and/or, one Or more than one operating system 1041, such as Windows ServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM etc..

Can be based on the middle control clothes shown in this Fig. 7 by the step performed by control server in above-described embodiment Business device structure.

Additionally it should be noted that, device embodiment described above is only schematically, wherein said The unit illustrated as separating component can be or may not be physically separate, shows as unit The parts shown can be or may not be physical location, i.e. may be located at a place, or also may be used To be distributed on multiple NE.Some or all of mould therein can be selected according to the actual needs Block realizes the purpose of the present embodiment scheme.It addition, in the device embodiment accompanying drawing of present invention offer, mould Annexation between block represents have communication connection between them, specifically can be implemented as one or more Communication bus or holding wire.Those of ordinary skill in the art are not in the case of paying creative work, i.e. It is appreciated that and implements.

Through the above description of the embodiments, those skilled in the art is it can be understood that arrive this Invention can add the mode of required common hardware by software and realize, naturally it is also possible to pass through specialized hardware Realize including special IC, dedicated cpu, private memory, special components and parts etc..General feelings Under condition, all functions completed by computer program can realize with corresponding hardware easily, and And, the particular hardware structure being used for realizing same function can also be diversified, such as analog circuit, Digital circuit or special circuit etc..But, the most more in the case of software program realize be more Good embodiment.Based on such understanding, technical scheme is the most in other words to existing skill The part that art contributes can embody with the form of software product, and this computer software product stores In the storage medium that can read, such as the floppy disk of computer, USB flash disk, portable hard drive, read only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic Dish or CD etc., including some instructions with so that computer equipment (can be personal computer, Server, or the network equipment etc.) perform the method described in each embodiment of the present invention.

In sum, above example only in order to technical scheme to be described, is not intended to limit； Although being described in detail the present invention with reference to above-described embodiment, those of ordinary skill in the art should Work as understanding: the technical scheme described in the various embodiments described above still can be modified by it, or to it Middle part technical characteristic carries out equivalent；And these amendments or replacement, do not make appropriate technical solution Essence depart from various embodiments of the present invention technical scheme spirit and scope.

Claims

1. a Monitoring framework method for designing, it is characterised in that including:

Method the most according to claim 1, it is characterised in that described by the described prison that there is exception After control data store in described data base, described method also includes:

Described monitoring server extracts from described data base and there are abnormal monitoring data, and to abduction Show to carry out accident analysis.

Method the most according to claim 1, it is characterised in that described monitoring server obtains each The monitoring data that in virtual machine, agent unit reports respectively, including:

Described monitoring server obtain in each virtual machine agent unit according to each preset monitored item and The monitoring data of collection cycle real-time report；

Described from control server obtain described control server each virtual machine described is controlled time existence Monitoring data, including:

Described monitoring server obtains described control server according to each preset monitoring from control server The monitoring data that item and collection cycle send in real time.

Method the most according to claim 3, it is characterised in that described monitoring server is from described number Anomaly analysis is carried out according to storehouse is read monitoring data, including:

Described monitoring server carries out anomaly analysis according to the many groups of monitoring data got, and determines existence Abnormal monitored item；

The described monitoring data to there is exception alert, including:

Described monitoring server is to determining that the multiple abnormal monitoring data that there is abnormal same monitored item are entered Row polymerization, carries out concentrating alarm according to polymerization result.

The most according to the method in any one of claims 1 to 3, it is characterised in that described monitoring clothes Business device obtains the monitoring data that in each virtual machine, agent unit reports respectively, including:

Described monitoring server obtains system layer data that in each virtual machine, agent unit reports respectively and enters Journey layer data.

Method the most according to claim 5, it is characterised in that described monitoring server obtains each System layer data that in virtual machine, agent unit reports respectively and process level data, including:

Described monitoring server arranges systematic collection unit and process collector unit, described systematic collection list Unit, for receiving the system layer data that in each virtual machine described, agent unit reports respectively；Described process Collector unit, for receiving the process level data that in each virtual machine described, agent unit reports respectively.

7. according to the method described in claim 5 or 6, it is characterised in that described monitoring server will obtain The supervising data storage got in data base, including:

Described monitoring server arranges system data word bank and process data word bank, described monitoring server Described system layer data are stored in described system data word bank, described process level data are stored institute State in process data word bank.

The most according to the method in any one of claims 1 to 3, it is characterised in that described therefrom control Server obtains the monitoring data existed when each virtual machine described is controlled by described control server, including:

Described monitoring server obtains described control server from described control server and passes through virtual platform Control module layer data, podium level data and the client layer data existed during each virtual machine described, described Being deployed with multiple module on virtual platform, described client layer data include: user deposits when using virtual machine Virtual machine service data.

Method the most according to claim 8, it is characterised in that described monitoring server is from described Control server exists when obtaining described control server by virtual platform control each virtual machine described Module layer data, podium level data and client layer data, including:

Described monitoring server arranges module collection unit, platform collector unit and user's collector unit, Described module collection unit, for pulling described control server by virtualization from described control server The module layer data produced during each virtual machine described in platform courses；Described platform collector unit, for from Described control server pulls the podium level data of described virtual platform；Described user's collector unit, uses In pulling, from described control server, the client layer data produced when user uses virtual machine.

Method the most according to claim 8 or claim 9, it is characterised in that described monitoring server will The supervising data storage got in data base, including:

Described monitoring server arranges module data word bank, platform data word bank and user data word bank, Described module layer data are stored in described module data word bank by described monitoring server, by described platform Layer data stores in described platform data word bank, and described client layer data are stored described user data In word bank.

11. 1 kinds of Monitoring framework methods for designing, it is characterised in that including:

Described agent unit reports described monitoring data to monitoring server.

12. methods according to claim 11, it is characterised in that the monitoring of described agent unit is virtual Machine, produces monitoring data according to the service data of described virtual machine, including:

The monitored item that described agent unit is arranged according to configuration file every service data to described virtual machine Monitoring the most in real time, the collection cycle arranged according to described configuration file produces the monitoring data of many groups.

13. according to the method described in claim 11 or 12, it is characterised in that described monitoring data, Including: system layer data and process level data；

Described agent unit reports described monitoring data to monitoring server, including:

The systematic collection unit that described agent unit is arranged in described monitoring server sends described system layer Data, the process collector unit arranged in described monitoring server sends described process level data.

14. 1 kinds of Monitoring framework methods for designing, it is characterised in that including:

15. methods according to claim 14, it is characterised in that described control server produces Monitoring data when each virtual machine is controlled, including:

Monitored item and collection cycle that described control server is arranged according to configuration file produce many groups respectively Monitoring data.

16. according to the method described in claims 14 or 15, it is characterised in that described monitoring data, Including: module layer data, podium level data and client layer data；

Described control server sends described monitoring according to the request of monitoring server to described monitoring server Data, including:

The module collection unit that described control server is arranged in described monitoring server sends described module Layer data, the platform collector unit arranged in described monitoring server sends described podium level data, to The user's collector unit arranged in described monitoring server sends described client layer data.

17. 1 kinds of monitoring servers, it is characterised in that including:

Memory element, by the supervising data storage that gets to data base；

18. monitoring servers according to claim 17, it is characterised in that described monitoring server Also include: display unit, there is abnormal supervising data storage to institute for described memory element by described After stating in data base, extract from described data base and there are abnormal monitoring data, and outwards show To carry out accident analysis.

19. monitoring servers according to claim 17, it is characterised in that described collector unit, Real according to each preset monitored item and collection cycle specifically for obtaining agent unit in each virtual machine Time the monitoring data that report；Described control server is obtained according to each preset monitoring from control server The monitoring data that item and collection cycle send in real time.

20. monitoring servers according to claim 19, it is characterised in that described Alarm Unit, Specifically for carrying out anomaly analysis according to the many groups of monitoring data got, determine and there is abnormal monitoring ?；To determining that the multiple abnormal monitoring data that there is abnormal same monitored item are polymerized, according to poly- Close result to carry out concentrating alarm.

21. according to the monitoring server according to any one of claim 17 to 19, it is characterised in that Described collector unit, including: systematic collection unit and process collector unit, described systematic collection unit, For receiving the system layer data that in each virtual machine described, agent unit reports respectively；Described process is collected Unit, for receiving the process level data that in each virtual machine described, agent unit reports respectively.

22. according to the monitoring server according to any one of claim 17 to 19, it is characterised in that Described collector unit, also includes: module collection unit, platform collector unit and user's collector unit, institute State module collection unit, flat by virtualization for pulling described control server from described control server Platform controls the module layer data produced during each virtual machine described；Described platform collector unit, for from institute State control server and pull the podium level data of described virtual platform；Described user's collector unit, is used for The client layer data produced when user uses virtual machine are pulled from described control server；Described virtualization is put down Being deployed with multiple module on platform, described client layer data include: it is virtual that user exists when using virtual machine Machine service data.

23. according to the monitoring server described in claim 21 or 22, it is characterised in that described monitoring Arranging system data word bank and process data word bank in server, described memory module, specifically for by institute State system layer data and store in described system data word bank, described process level data are stored described in enter Number of passes is according in word bank；

Described monitoring server arranges module data word bank, platform data word bank and user data word bank, Described memory module, specifically for described module layer data are stored in described module data word bank, will Described podium level data store in described platform data word bank, store described by described client layer data In user data word bank.

24. 1 kinds of agent units, it is characterised in that described agent unit is deployed in described virtual machine, Including: monitoring subelement and transmission subelement, wherein,

25. agent units according to claim 24, it is characterised in that described monitoring subelement, Every service data of described virtual machine is carried out by monitored item specifically for arranging according to configuration file respectively Monitoring in real time, according to the monitoring data of the many groups of collection cycle generation that described configuration file is arranged.

26. according to the agent unit described in claim 24 or 25, it is characterised in that described monitoring number According to, including: system layer data and process level data；

Described transmission subelement, sends out specifically for the systematic collection unit arranged in described monitoring server Sending described system layer data, the process collector unit arranged in described monitoring server sends described process Layer data.

27. 1 kinds of control servers, it is characterised in that including: data generating unit and transmitting element, Wherein,

28. control servers according to claim 27, it is characterised in that described data genaration list Unit, produces the monitoring of many groups respectively specifically for the monitored item arranged according to configuration file and collection cycle Data.

29. according to the method described in claim 27 or 28, it is characterised in that described monitoring data, Including: module layer data, podium level data and client layer data；

Described transmitting element, sends specifically for the module collection unit arranged in described monitoring server Described module layer data, the platform collector unit arranged in described monitoring server sends described podium level Data, the user's collector unit arranged in described monitoring server sends described client layer data.