CN107894944A - A kind of intelligent control method and system based under big data and cloud calculation service - Google Patents
A kind of intelligent control method and system based under big data and cloud calculation service Download PDFInfo
- Publication number
- CN107894944A CN107894944A CN201711241870.8A CN201711241870A CN107894944A CN 107894944 A CN107894944 A CN 107894944A CN 201711241870 A CN201711241870 A CN 201711241870A CN 107894944 A CN107894944 A CN 107894944A
- Authority
- CN
- China
- Prior art keywords
- warning information
- operation system
- center
- monitoring strategies
- alarm
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/32—Monitoring with visual or acoustical indication of the functioning of the machine
- G06F11/324—Display of status information
- G06F11/327—Alarm or error message display
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Alarm Systems (AREA)
Abstract
The invention discloses a kind of intelligent control method based under big data and cloud calculation service, including：Surveillance center sets the monitoring strategies of differentiation in units of operation system；When operation system triggers monitoring strategies, Surveillance center's generation warning information simultaneously sends warning information to Intelligent treatment center and intellectual analysis center；Intelligent treatment center is handled operation system according to the warning information and processing strategy, and the processing strategy includes expanding policy, take-back strategy and cooling strategy；Intellectual analysis central collection and statistical analysis warning information, and result is fed back into Surveillance center.The invention also discloses based on the intelligent monitor system under big data and cloud calculation service., can be according to the differentiation monitoring strategies that different business systems can customize using the present invention, and Intelligent treatment center and analysis center are provided, the alarm to appearance carries out intelligent processing method.
Description
Technical field
The present invention relates to areas of information technology, more particularly to a kind of intelligent monitoring based under big data and cloud calculation service
Method and based on the intelligent monitor system under big data and cloud calculation service.
Background technology
After VMware in 1998 propose the virtualization technology of X86-based, the development of virtualization technology tide is swift and violent, and short
After short a few years, the concept of the cloud computing based on virtualization technology is suggested, and user can take on demand in cloud computing environment
With resource, resource is used based on high in the clouds whenever and wherever possible under conditions of network is reachable.Pass through the development of more than 10 years, virtualization technology
It is gradually ripe with cloud computing technology, and be most of user admissions, increasing business will be all run on beyond the clouds.And according to
《State Council cultivates the decision with development strategy new industry on accelerating》In the expected future of strategic new industry that refers to,
At least more than 30% growth level will be reached in the five-year of Chinese cloud computing market.
On the other hand, under the fast-developing overall background of global IT application, big data has turned into important basic of country
Strategic resource, just leading new round scientific and technical innovation.Issued according to Chinese information Communication Studies institute《Chinese big data hair in 2015
Open up survey report》It has been shown that, Chinese big data market scale in 2015 are up to 115.9 hundred million yuan, and speedup is up to 38%.Furthermore it is contemplated that
2016 to 2018 years Chinese big data market scales are also by the rapid growth of maintenance 40% or so.
Cloud computing and big data technology are the main trends of Information Technology Development so far at the beginning of 21 century, and due to cloud computing
The flexible characteristic easily extended, it turns into the base node framework of big data data analysis and process, started in 2013, two kinds
Technology has shown highly effective and close marriage relation, and this relation will be more close in future development.
Based on above technology trends, big data business can be not only supported on cloud computing platform, also has substantial amounts of biography
Service operation unite above.Based under this pattern, it would be desirable to a kind of intelligent monitoring system is provided, to big data business and
Cloud calculation service is monitored management, helps user to grasp service operation situation in real time.
Monitoring management mode on cloud computing platform common at present is mainly following several：
(1) basis monitoring：The CPU of physical server and virtual machine operations, internal memory operation, storage running, IO are read and write
Monitored in real time, there is provided availability data, there is provided alarming mechanism, alerted according to fixed threshold triggers.
(2) advanced monitoring：The object of monitoring includes underlying resource, application layer, safety etc..Monitored object is comprehensive, that is, includes
Underlying resource, also including the application monitoring above resource, while also the safety under virtualized environment is monitored.
Two above-mentioned technical schemes, it is more stiff, not enough intelligently.In scheme (1), simply to the resource of cloud computing environment
Aspect is monitored management, and user can only be helped to find the bottleneck that resource uses in time, can prevent in certain degree big
Data service and traditional operation system go wrong.And in scheme (2), although the scope of monitoring is wider, can monitor
The underlying resource of service operation, the application running situation on upper strata can be also monitored, triggering warning information can also be in time provided and helped
User solves the problems, such as service operation, but scheme (2) is not still optimal monitoring scheme, because each operation system is the bottom of to
The sensitivity of layer resource is that different, i.e., different operation system is not using reaction caused by bottleneck to same resource
Equally.For example in big data business, the collection of data and processing procedure have bigger requirement to CPU computings, IO read-writes, such as
Fruit network goes wrong, and read-write postpones, then the data collection service of whole big data will receive serious influence.It is another
Aspect, in scheme (1) and scheme (2), all without the monitoring and treatment mechanism for providing the slack resources to become more meticulous.
The content of the invention
The technical problems to be solved by the invention are, there is provided a kind of intelligence prison based under big data and cloud calculation service
Method and system are controlled, may be according to the differentiation monitoring strategies that different business systems can customize, and intelligent place is provided
Reason center and analysis center, the alarm to appearance carry out intelligent processing method.
In order to solve the above-mentioned technical problem, the invention provides a kind of intelligence prison based under big data and cloud calculation service
Prosecutor method, including：Surveillance center sets the monitoring strategies of differentiation in units of operation system；When operation system triggers monitoring plan
When slightly, Surveillance center's generation warning information simultaneously sends warning information to Intelligent treatment center and intellectual analysis center；At intelligence
Reason center according to the warning information and processing strategy operation system is handled, it is described processing strategy include expanding policy,
Take-back strategy and cooling strategy；Intellectual analysis central collection and statistical analysis warning information, and result is fed back into Surveillance center.
As the improvement of such scheme, the monitoring strategies include underlying resource basis monitoring strategies, upper-layer service application
Monitoring strategies and the advanced monitoring strategies of slack resources.
As the improvement of such scheme, underlying resource basis monitoring strategies include：Surveillance center obtains virtual in real time
The technical indicator of main frame, the technical indicator include central processor CPU utilization rate, fictitious host computer memory usage and disk I/O
Utilization rate；The threshold value and logical relation of each technical indicator triggering alarm are set respectively.
As the improvement of such scheme, the upper-layer service application monitoring strategies include：Surveillance center obtains operation system
In the application installed, and set and open business monitoring function.
As the improvement of such scheme, the advanced monitoring strategies of slack resources include：Set for the virtual machine of operation system
Determine monitor control index, the monitor control index includes index single-point and left unused threshold value, single-point acquiring cycle, idle measurement period, index spare time
Put ratio and logical relation；The idle result of virtual machine is judged according to monitor control index.
As the improvement of such scheme, the intellectual analysis central collection and statistical analysis warning information simultaneously feed back result
Method to Surveillance center includes：The warning information of Surveillance center is obtained, the warning information includes：Alarm object, virtual machine
Index parameter value and metrics-thresholds when the operation system at place, alarm triggered time, the type of alarm, triggering alert；Division is accused
The alarm type of alert information, the alarm type include underlying resource basis, upper-layer service application, slack resources；According to alarm
Type carries out law-analysing and anticipation to warning information.
Correspondingly, present invention also offers a kind of intelligent monitor system based under big data and cloud calculation service, including：
Surveillance center, for setting the monitoring strategies of differentiation in units of operation system, and when operation system triggers monitoring strategies
Generation warning information simultaneously sends warning information to Intelligent treatment center and intellectual analysis center；Intelligent treatment center, for root
Operation system is handled according to the warning information and processing strategy, the processing strategy includes expanding policy, take-back strategy
And cooling strategy；Intellectual analysis center, for collecting simultaneously statistical analysis warning information, and result is fed back into Surveillance center.
As the improvement of such scheme, the Surveillance center includes：Monitoring strategies setting unit, for using operation system as
Unit sets the monitoring strategies of differentiation；Warning information generation unit, for the generation announcement when operation system triggers monitoring strategies
Alert information；Warning information transmitting element, for warning information to be sent to Intelligent treatment center and intellectual analysis center.
As the improvement of such scheme, the monitoring strategies setting unit includes：Underlying resource basis monitoring strategies are set
Unit, for obtaining the technical indicator of fictitious host computer in real time and setting the threshold value of each technical indicator triggering alarm and logic to close respectively
System；Upper-layer service application monitoring strategies setting unit, for obtaining the application installed in operation system and setting unlatching business to supervise
Control function；The advanced monitoring strategies setting unit of slack resources, for setting monitor control index and basis for the virtual machine of operation system
Monitor control index judges the idle result of virtual machine.
As the improvement of such scheme, the intellectual analysis center includes：Acquiring unit, for obtaining the announcement of Surveillance center
Alert information, the warning information include：Operation system, alarm triggered time, the class of alarm where alarm object, virtual machine
Index parameter value and metrics-thresholds when type, triggering alarm；Division unit, for dividing the alarm type of warning information, the announcement
Alert type includes underlying resource basis, upper-layer service application, slack resources；Analytic unit, for according to alarm type to alarm
Information carries out law-analysing and anticipation；Feedback unit, for result to be fed back into Surveillance center.
Implement the present invention, have the advantages that：
1st, the present invention is the self-defined monitoring system using operation system as granularity, it is allowed to which user is in magnanimity operation system
On cloud computing platform in units of operation system, the monitoring strategies with differentiation are set.
2nd, Surveillance center can link Intelligent treatment center, when operation system triggers the prison set in the process of running
Control strategy, solves the problems, such as resource bottleneck automatically by system.
3rd, slack resources policing algorithm process support it is self-defined, can with user-defined counter single-point leave unused threshold value, collection period,
Measurement period, idle ratio logical relation.
4th, monitoring system includes intellectual analysis center, analysis result reverse correlation Surveillance center, helps user setting prison
Rational monitoring is carried out when controlling tactful according to the data at intellectual analysis center to set.
Brief description of the drawings
Fig. 1 is the flow chart of the invention based on the intelligent control method under big data and cloud calculation service；
Fig. 2 is the structural representation of the invention based on the intelligent monitor system under big data and cloud calculation service；
Fig. 3 is the structural representation of Tu2Zhong Surveillance center；
Fig. 4 is the structural representation at intellectual analysis center in Fig. 2.
Embodiment
To make the object, technical solutions and advantages of the present invention clearer, the present invention is made into one below in conjunction with accompanying drawing
It is described in detail on step ground.Only this is stated, appearance in the text of the invention or the side such as the up, down, left, right, before and after that will appear from, inside and outside
Position word, only on the basis of the accompanying drawing of the present invention, it is not the specific restriction to the present invention.
Referring to Fig. 1, Fig. 1 shows the implementation of the invention based on the intelligent control method under big data and cloud calculation service
Example, it includes：
S101, Surveillance center set the monitoring strategies of differentiation in units of operation system；
It is high that the monitoring strategies include underlying resource basis monitoring strategies, upper-layer service application monitoring strategies and slack resources
Level monitoring strategies.Specifically：
Underlying resource basis monitoring strategies include：Surveillance center obtains the technical indicator of fictitious host computer in real time, and divides
The threshold value and logical relation of each technical indicator triggering alarm are not set.The technical indicator include central processor CPU utilization rate,
Fictitious host computer memory usage and disk I/O utilization rate.
It should be noted that user can be respectively to " central processor CPU utilizes in the monitoring strategies of underlying resource basis
The self-defined threshold value that triggering alarm is set of these three technical indicators of rate, fictitious host computer memory usage and disk I/O utilization rate ", and
And can be with the logical relation of these three self-defined technical indicators.For example A virtual machines, user can be set when cpu busy percentage height
In 80% and memory usage be higher than 90% when trigger virtual machine alarming mechanism at once.
The upper-layer service application monitoring strategies include：Surveillance center obtains the application installed in operation system, and sets
Open business monitoring function.Specifically, service application monitoring is supported：MySQL, Oracle, web are applied.MySQL：Operation system,
Triggered time, connection status, thread connection number, most frequent SQL, slow query SQL；Oracle：Operation system, triggered time, company
Connect state, thread connection number, most frequent SQL, slow query SQL；Web is applied：Operation system, port numbers, connection status.
The advanced monitoring strategies of slack resources include：Monitor control index is set for the virtual machine of operation system and according to monitoring
Index judges the idle result of virtual machine.The monitor control index includes the idle threshold value of index single-point, single-point acquiring cycle, idle system
Count cycle, the idle ratio of index and logical relation；
The idle threshold value of index single-point is set for the virtual machine of operation system, i.e., judges that virtual machine is on certain acquisition time
The no criterion to leave unused, corresponding specific parameter are exactly CPU, internal memory, the idle threshold value of storage.As setting target single-point is not busy
Put threshold value：Cpu busy percentage≤85%, memory usage≤60%, space utilisation≤90%, then what each collection point obtained
Data all can be contrasted and recorded with the idle threshold value of single-point set.
Single-point acquiring cycle, that is, the interval of acquisition time are set for the virtual machine of operation system.It is assumed to be business system
The collection period that the A that unites is set is 30 minutes, then monitoring system can gather 4 void in an operation system A every 30 minutes
The CPU of plan machine, internal memory, space utilisation are simultaneously contrasted with single-point threshold value of leaving unused.
For the idle measurement period of virtual machine setting of operation system, that is, it is arranged in a period of time, the statistics for situation of leaving unused.
The idle measurement period for being assumed to be operation system A settings is 10 days, then monitoring system can be tactful with 10 days by what is set
For a cycle, in this 10 days, all data acquisition and idle monitoring were carried out to operation system A virtual machine every 30 minutes,
Draw idle result.
The idle ratio of index and logical relation are set for the virtual machine of operation system.CPU, internal memory, storage index are set
The idle standard of item, it is assumed that be that the idle ratio that CPU is set is 70%, the idle ratio for being memory setting is 80%, is set for storage
The idle ratio put be 90%, their logical relation be and.Hereinbefore, we are provided with for operation system A virtual machine
The measurement period of 10 days and the collection period of 30 minutes, and it is provided with the idle threshold value of index single-point, respectively CPU for virtual machine
Utilization rate≤85%, memory usage≤60%, space utilisation≤90%.Every virtual machine collects quantity in so 10 days
For：10 × 24 × 60/30=480.The value of i.e. 480 cpu busy percentages, the value of 480 memory usages, 480 storages utilize
The value of rate.Assuming that in this 10 day time, the number of cpu busy percentage≤85% collected accounts for total number and accounts for total number 480
Ratio more than 70%, and the number of memory usage≤60% collected accounts for the ratio that total number accounts for total number 480
Example is more than 80%, and the number of space utilisation≤60% collected accounts for total number and accounts for the ratio of total number 480 and surpasses
Cross 90%, then the virtual machine is may determine that in this 10 days measurement period, for the virtual machine that leaves unused.
S102, when operation system trigger monitoring strategies when, Surveillance center generation warning information and by warning information send to
Intelligent treatment center and intellectual analysis center；
S103, Intelligent treatment center are handled operation system according to the warning information and processing strategy；
It should be noted that Intelligent treatment center linkage surveillance center, virtual machine underlying resource using once there is bottleneck,
Resource can be laterally added, and inherit automatically into affiliated operation system, one according to pre-set Intelligent treatment mode
The virtual machine that last time laterally adds will be removed out operation system, and release by denier resource automatically using the condition for reaching release
With the relation of other virtual machines in operation system.
Specifically, the processing strategy includes expanding policy, take-back strategy and cooling strategy.
Expanding policy：The underlying resource basis monitoring at linkage surveillance center, obtaining the resource of virtual machine under operation system makes
Use situation.The quantity and resource specification for the virtual machine for needing laterally to add are set, and the virtual machine laterally added is based on former empty
Plan machine is what masterplate was carried out.
Take-back strategy：Obtain the resource service condition of virtual machine under operation system.The recovery threshold of same index parameter is set
Value, activation threshold value is run, by the virtual machine of automatic recovery expanding policy addition.
Cool time：Start to calculate cool time after addition virtual machine success, within cool time, do not perform recovery plan
Slightly.
Way of recycling：Delete manually, automatic shutdown.
Using operation system A as example, set when virtual machine cpu busy percentage is higher than 80% and memory usage is higher than 90%
When trigger virtual machine alarming mechanism at once, linkage Intelligent treatment center is handled.Assuming that virtual machine ACPU utilization rates
It has been higher than 90%, then triggering alarm, and by Intelligent treatment center, using virtual machine A as masterplate, has laterally added into operation system A
Add 2 virtual machines.In the process of running, virtual machine A and the average CPU utilization by it for two virtual machines of masterplate addition
Less than threshold value (such as：20%), and cool time has been had been subjected to, then the resource of the two days virtual machines newly added can be reclaimed, returned
The mode of receipts supports to delete manually and automatic shutdown.
S104, intellectual analysis central collection and statistical analysis warning information, and result is fed back into Surveillance center.
Intellectual analysis center is counted the warning information of all triggerings.Then according to certain rule carry out classification and
Obtained conclusion reverse feedback, is referred to analysis center by analysis to Surveillance center, user when monitoring strategies are set
The trend of offer is configured.
Specifically, the intellectual analysis central collection and statistical analysis warning information and result is fed back into Surveillance center
Method includes：
A, the warning information of Surveillance center is obtained, the warning information includes：Business system where alarm object, virtual machine
Index parameter value and metrics-thresholds when system, alarm triggered time, the type of alarm, triggering alarm；
B, the alarm type of warning information is divided according to the criterion of Surveillance center, the alarm type includes underlying resource base
Plinth, upper-layer service application, slack resources, are specifically classified as follows shown in table：
C, law-analysing and anticipation are carried out to warning information according to alarm type.
Underlying resource basis：CPU, internal memory are analyzed respectively, store three dimensions and time, the relation of threshold value, and according to
There is the tendency of bottleneck in resource in rule one week future of anticipation in past one month.
Upper-layer service application：The rule of MySQL, Oracle query statement is analyzed respectively, counts the TOP most frequencies in one month
Numerous SQL and TOP are inquired about slowly.And according to the most frequent SQL occurred in the rule in past one month one week future of anticipation and slowly
Query statement.
Slack resources：It is most to analyze virtual machine quantity of being left unused existing for the operation system of which type, and in time
The idle virtual machine number change trend of statistics, and walking for quantity of being left unused in following one week is prejudged according to the rule in past one month
Gesture.
From the foregoing, it will be observed that the present invention allows user on the cloud computing platform of magnanimity operation system in units of operation system,
Monitoring strategies with differentiation are set；And when these operation systems trigger the monitoring strategies set in the process of running,
Can then link Intelligent treatment center, by Command Center process problem (for example, when the alarm of triggering meets that Intelligent treatment center is triggered
During condition, it will it is laterally automatic in the operation system to add resource, dynamically to ensure demand of the business to resource)；Intelligence simultaneously
Energy analysis center can collect and warning information caused by statistical analysis, so as to which reverse feedback is to user, helps user setting prison
Rational monitoring is carried out when controlling tactful according to the data at intellectual analysis center to set.
Referring to Fig. 2, Fig. 2 shows the knot of the invention based on the intelligent monitor system 100 under big data and cloud calculation service
Structure schematic diagram, it includes：
Surveillance center 1, for setting the monitoring strategies of differentiation in units of operation system, and when operation system triggering prison
Warning information is generated when controlling tactful and sends warning information to Intelligent treatment center 2 and intellectual analysis center 3；
Intelligent treatment center 2, for being handled according to the warning information and processing strategy operation system, the place
Reason strategy includes expanding policy, take-back strategy and cooling strategy；It should be noted that Intelligent treatment center linkage surveillance center 1,
Virtual machine underlying resource can laterally add resource using once there is bottleneck according to pre-set Intelligent treatment mode,
And inherit automatically into affiliated operation system, once resource will laterally be added last automatically using the condition for reaching release
The virtual machine added removes out operation system, and releases the relation with other virtual machines in operation system.Specifically, the processing plan
Slightly include expanding policy, take-back strategy and cooling strategy.Expanding policy：The underlying resource basis monitoring at linkage surveillance center 1, is obtained
Take the resource service condition of virtual machine under operation system.The quantity and resource specification for the virtual machine for needing laterally to add are set,
The virtual machine laterally added is to be carried out based on former virtual machine for masterplate.Take-back strategy：Obtain the money of virtual machine under operation system
Source service condition.The recovery threshold value of same index parameter is set, runs activation threshold value, by the void of automatic recovery expanding policy addition
Plan machine.Cool time：Start to calculate cool time after addition virtual machine success, within cool time, do not perform take-back strategy.Return
Debit's formula：Delete manually, automatic shutdown.Using operation system A as example, set when virtual machine cpu busy percentage higher than 80% and
Memory usage triggers virtual machine alarming mechanism at once when being higher than 90%, and linkage Intelligent treatment center 2 is handled.It is false
If virtual machine ACPU utilization rates have been higher than 90%, then triggering alarm, and by Intelligent treatment center 2, using virtual machine A as masterplate,
2 virtual machines are laterally added into operation system A.In the process of running, virtual machine A and two void by it for masterplate addition
The average CPU utilization of plan machine is less than threshold value (such as：20%), and cool time has been had been subjected to, then can reclaim and newly add
The resource of two days virtual machines, the mode of recovery supports to delete manually and automatic shutdown.
Intellectual analysis center 3, for collecting simultaneously statistical analysis warning information, and result is fed back into Surveillance center 1.Intelligence
Analysis center 3 is counted the warning information of all triggerings, is then classified and is analyzed according to certain rule, obtaining
Conclusion reverse feedback to Surveillance center 1, user be referred to when monitoring strategies are set intellectual analysis center 3 offer
Trend is configured.
As shown in figure 3, the Surveillance center 1 includes：
Monitoring strategies setting unit 11, for setting the monitoring strategies of differentiation in units of operation system；The monitoring
Strategy includes underlying resource basis monitoring strategies, upper-layer service application monitoring strategies and the advanced monitoring strategies of slack resources.
Warning information generation unit 12, for generating warning information when operation system triggers monitoring strategies；
Warning information transmitting element 13, for warning information to be sent to Intelligent treatment center 2 and intellectual analysis center 3.
Further, the monitoring strategies setting unit 11 includes：
Underlying resource basis monitoring strategies setting unit 11, for obtaining the technical indicator of fictitious host computer in real time and setting respectively
Put the threshold value and logical relation of each technical indicator triggering alarm；The technical indicator includes central processor CPU utilization rate, virtual
Host memory utilization rate and disk I/O utilization rate.It should be noted that user can distinguish in the monitoring strategies of underlying resource basis
To " central processor CPU utilization rate, fictitious host computer memory usage and disk I/O utilization rate ", these three technical indicators are self-defined
The threshold value of triggering alarm is set, and can be with the logical relation of these three self-defined technical indicators.Such as to A virtual machines, user
It can set and trigger virtual machine alarm machine at once when cpu busy percentage is higher than 80% and memory usage is higher than 90%
System.
Upper-layer service application monitoring strategies setting unit 112, for obtaining the application installed in operation system and setting is opened
Open business monitoring function；Specifically, service application monitoring is supported：MySQL, Oracle, web are applied.MySQL：Operation system, touch
Send out time, connection status, thread connection number, most frequent SQL, slow query SQL；Oracle：Operation system, triggered time, connection
State, thread connection number, most frequent SQL, slow query SQL；Web is applied：Operation system, port numbers, connection status.
The advanced monitoring strategies setting unit 113 of slack resources, for setting monitor control index simultaneously for the virtual machine of operation system
The idle result of virtual machine is judged according to monitor control index.The monitor control index includes index single-point and left unused threshold value, single-point acquiring week
Phase, idle measurement period, the idle ratio of index and logical relation.
The idle threshold value of index single-point is set for the virtual machine of operation system, i.e., judges that virtual machine is on certain acquisition time
The no criterion to leave unused, corresponding specific parameter are exactly CPU, internal memory, the idle threshold value of storage.As setting target single-point is not busy
Put threshold value：Cpu busy percentage≤85%, memory usage≤60%, space utilisation≤90%, then what each collection point obtained
Data all can be contrasted and recorded with the idle threshold value of single-point set.
Single-point acquiring cycle, that is, the interval of acquisition time are set for the virtual machine of operation system.It is assumed to be business system
The collection period that the A that unites is set is 30 minutes, then monitoring system can gather 4 void in an operation system A every 30 minutes
The CPU of plan machine, internal memory, space utilisation are simultaneously contrasted with single-point threshold value of leaving unused.
For the idle measurement period of virtual machine setting of operation system, that is, it is arranged in a period of time, the statistics for situation of leaving unused.
The idle measurement period for being assumed to be operation system A settings is 10 days, then monitoring system can be tactful with 10 days by what is set
For a cycle, in this 10 days, all data acquisition and idle monitoring were carried out to operation system A virtual machine every 30 minutes,
Draw idle result.
The idle ratio of index and logical relation are set for the virtual machine of operation system.CPU, internal memory, storage index are set
The idle standard of item, it is assumed that be that the idle ratio that CPU is set is 70%, the idle ratio for being memory setting is 80%, is set for storage
The idle ratio put be 90%, their logical relation be and.Hereinbefore, we are provided with for operation system A virtual machine
The measurement period of 10 days and the collection period of 30 minutes, and it is provided with the idle threshold value of index single-point, respectively CPU for virtual machine
Utilization rate≤85%, memory usage≤60%, space utilisation≤90%.Every virtual machine collects quantity in so 10 days
For：10 × 24 × 60/30=480.The value of i.e. 480 cpu busy percentages, the value of 480 memory usages, 480 storages utilize
The value of rate.Assuming that in this 10 day time, the number of cpu busy percentage≤85% collected accounts for total number and accounts for total number 480
Ratio more than 70%, and the number of memory usage≤60% collected accounts for the ratio that total number accounts for total number 480
Example is more than 80%, and the number of space utilisation≤60% collected accounts for total number and accounts for the ratio of total number 480 and surpasses
Cross 90%, then the virtual machine is may determine that in this 10 days measurement period, for the virtual machine that leaves unused.
As shown in figure 4, the intellectual analysis center 3 includes：
Acquiring unit 31, for obtaining the warning information of Surveillance center 1, the warning information includes：It is alarm object, virtual
Index parameter value and metrics-thresholds when the type of operation system, alarm triggered time, alarm where machine, triggering alert；
Division unit 32, for dividing the alarm type of warning information, the alarm type include underlying resource basis, on
Layer service application, slack resources；
Analytic unit 33, for carrying out law-analysing and anticipation to warning information according to alarm type；Wherein, underlying resource
Basis：CPU, internal memory, three dimensions of storage and time, the relation of threshold value are analyzed respectively, and according to the rule in past one month
There is the tendency of bottleneck in resource in anticipation is following one week.Upper-layer service application：MySQL, Oracle query statement are analyzed respectively
Rule, the TOP most frequent SQL and TOP counted in one month are inquired about slowly.And future one is prejudged according to the rule in past one month
The most frequent SQL and slow query statement occurred in week.Slack resources：Analyze void of being left unused existing for the operation system of which type
Plan machine quantity is most, and counts idle virtual machine number change trend in time, and according to the rule in past one month
The tendency for quantity of being left unused in anticipation is following one week.
Feedback unit 34, for result to be fed back into Surveillance center 1.
From the foregoing, it will be observed that the invention has the advantages that：
1st, the present invention is the self-defined monitoring system using operation system as granularity, it is allowed to which user is in magnanimity operation system
On cloud computing platform in units of operation system, the monitoring strategies with differentiation are set.
2nd, Surveillance center can link Intelligent treatment center, solve the problems, such as resource bottleneck automatically by system.
3rd, slack resources policing algorithm process support it is self-defined, can with user-defined counter single-point leave unused threshold value, collection period,
Measurement period, idle ratio logical relation.
4th, monitoring system includes intellectual analysis center, analysis result reverse correlation Surveillance center, there is provided decision data.
Described above is the preferred embodiment of the present invention, it is noted that for those skilled in the art
For, under the premise without departing from the principles of the invention, some improvements and modifications can also be made, these improvements and modifications are also considered as
Protection scope of the present invention.
Claims (10)
- A kind of 1. intelligent control method based under big data and cloud calculation service, it is characterised in that including：Surveillance center sets the monitoring strategies of differentiation in units of operation system；When operation system triggers monitoring strategies, Surveillance center's generation warning information simultaneously sends warning information into Intelligent treatment The heart and intellectual analysis center；Intelligent treatment center is handled operation system according to the warning information and processing strategy, and the processing strategy includes Expanding policy, take-back strategy and cooling strategy；Intellectual analysis central collection and statistical analysis warning information, and result is fed back into Surveillance center.
- 2. as claimed in claim 1 based on the intelligent control method under big data and cloud calculation service, it is characterised in that described Monitoring strategies include underlying resource basis monitoring strategies, upper-layer service application monitoring strategies and the advanced monitoring strategies of slack resources.
- 3. as claimed in claim 2 based on the intelligent control method under big data and cloud calculation service, it is characterised in that described Underlying resource basis monitoring strategies include：Surveillance center obtains the technical indicator of fictitious host computer in real time, and the technical indicator includes central processor CPU utilization rate, void Intend host memory utilization rate and disk I/O utilization rate；The threshold value and logical relation of each technical indicator triggering alarm are set respectively.
- 4. as claimed in claim 2 based on the intelligent control method under big data and cloud calculation service, it is characterised in that described Upper-layer service application monitoring strategies include：Surveillance center obtains the application installed in operation system, and sets unlatching business monitoring Function.
- 5. as claimed in claim 2 based on the intelligent control method under big data and cloud calculation service, it is characterised in that described The advanced monitoring strategies of slack resources include：Monitor control index is set for the virtual machine of operation system, the monitor control index includes the idle threshold value of index single-point, single-point acquiring Cycle, idle measurement period, the idle ratio of index and logical relation；The idle result of virtual machine is judged according to monitor control index.
- 6. as claimed in claim 1 based on the intelligent control method under big data and cloud calculation service, it is characterised in that described Result is simultaneously fed back to the method for Surveillance center and included by intellectual analysis central collection and statistical analysis warning information：The warning information of Surveillance center is obtained, the warning information includes：Operation system, announcement where alarm object, virtual machine Index parameter value and metrics-thresholds when alert triggered time, the type of alarm, triggering alarm；The alarm type of warning information is divided, the alarm type includes underlying resource basis, upper-layer service application, idle money Source；Law-analysing and anticipation are carried out to warning information according to alarm type.
- A kind of 7. intelligent monitor system based under big data and cloud calculation service, it is characterised in that including：Surveillance center, for setting the monitoring strategies of differentiation in units of operation system, and when operation system triggering monitoring plan Warning information is generated when slightly and sends warning information to Intelligent treatment center and intellectual analysis center；Intelligent treatment center, for being handled according to the warning information and processing strategy operation system, the processing plan Slightly include expanding policy, take-back strategy and cooling strategy；Intellectual analysis center, for collecting simultaneously statistical analysis warning information, and result is fed back into Surveillance center.
- 8. as claimed in claim 7 based on the intelligent monitor system under big data and cloud calculation service, it is characterised in that described Surveillance center includes：Monitoring strategies setting unit, for setting the monitoring strategies of differentiation in units of operation system；Warning information generation unit, for generating warning information when operation system triggers monitoring strategies；Warning information transmitting element, for warning information to be sent to Intelligent treatment center and intellectual analysis center.
- 9. as claimed in claim 8 based on the intelligent monitor system under big data and cloud calculation service, it is characterised in that described Monitoring strategies setting unit includes：Underlying resource basis monitoring strategies setting unit, for obtaining the technical indicator of fictitious host computer in real time and setting each skill respectively The threshold value and logical relation of art index triggering alarm；Upper-layer service application monitoring strategies setting unit, for obtaining the application installed in operation system and setting unlatching business to supervise Control function；The advanced monitoring strategies setting unit of slack resources, for setting monitor control index and according to monitoring for the virtual machine of operation system Index judges the idle result of virtual machine.
- 10. as claimed in claim 7 based on the intelligent monitor system under big data and cloud calculation service, it is characterised in that institute Stating intellectual analysis center includes：Acquiring unit, for obtaining the warning information of Surveillance center, the warning information includes：Where alarm object, virtual machine Operation system, the alarm triggered time, the type of alarm, triggering alarm when index parameter value and metrics-thresholds；Division unit, for dividing the alarm type of warning information, the alarm type includes underlying resource basis, upper-layer service Using, slack resources；Analytic unit, for carrying out law-analysing and anticipation to warning information according to alarm type；Feedback unit, for result to be fed back into Surveillance center.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711241870.8A CN107894944A (en) | 2017-11-30 | 2017-11-30 | A kind of intelligent control method and system based under big data and cloud calculation service |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711241870.8A CN107894944A (en) | 2017-11-30 | 2017-11-30 | A kind of intelligent control method and system based under big data and cloud calculation service |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107894944A true CN107894944A (en) | 2018-04-10 |
Family
ID=61807365
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711241870.8A Pending CN107894944A (en) | 2017-11-30 | 2017-11-30 | A kind of intelligent control method and system based under big data and cloud calculation service |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107894944A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111884878A (en) * | 2020-07-24 | 2020-11-03 | 樊馨 | Data monitoring method based on block chain |
| CN112291112A (en) * | 2020-10-27 | 2021-01-29 | 中国光大银行股份有限公司 | Cloud computing service monitoring system and method |
| CN117527523A (en) * | 2023-11-23 | 2024-02-06 | 广东堡塔安全技术有限公司 | Cloud computing-based server security monitoring system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100102982A1 (en) * | 2008-10-28 | 2010-04-29 | Open Systems International, Inc. | Holistic alarm monitoring |
| CN102096461A (en) * | 2011-01-13 | 2011-06-15 | 浙江大学 | Energy-saving method of cloud data center based on virtual machine migration and load perception integration |
| CN104378262A (en) * | 2013-12-13 | 2015-02-25 | 国家计算机网络与信息安全管理中心 | Intelligent monitoring analyzing method and system under cloud computing |
| CN104410535A (en) * | 2014-12-23 | 2015-03-11 | 浪潮电子信息产业股份有限公司 | Intelligent cloud resource monitoring and warning method |
| CN105260235A (en) * | 2015-09-23 | 2016-01-20 | 浪潮集团有限公司 | Method and device for scheduling resources on basis of application scenarios in cloud platform |
-
2017
- 2017-11-30 CN CN201711241870.8A patent/CN107894944A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100102982A1 (en) * | 2008-10-28 | 2010-04-29 | Open Systems International, Inc. | Holistic alarm monitoring |
| CN102096461A (en) * | 2011-01-13 | 2011-06-15 | 浙江大学 | Energy-saving method of cloud data center based on virtual machine migration and load perception integration |
| CN104378262A (en) * | 2013-12-13 | 2015-02-25 | 国家计算机网络与信息安全管理中心 | Intelligent monitoring analyzing method and system under cloud computing |
| CN104410535A (en) * | 2014-12-23 | 2015-03-11 | 浪潮电子信息产业股份有限公司 | Intelligent cloud resource monitoring and warning method |
| CN105260235A (en) * | 2015-09-23 | 2016-01-20 | 浪潮集团有限公司 | Method and device for scheduling resources on basis of application scenarios in cloud platform |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111884878A (en) * | 2020-07-24 | 2020-11-03 | 樊馨 | Data monitoring method based on block chain |
| CN112291112A (en) * | 2020-10-27 | 2021-01-29 | 中国光大银行股份有限公司 | Cloud computing service monitoring system and method |
| CN112291112B (en) * | 2020-10-27 | 2022-11-15 | 中国光大银行股份有限公司 | Cloud computing service monitoring system and method |
| CN117527523A (en) * | 2023-11-23 | 2024-02-06 | 广东堡塔安全技术有限公司 | Cloud computing-based server security monitoring system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP5584780B2 (en) | Data collection method, data collection apparatus, and network management device | |
| CN107894944A (en) | A kind of intelligent control method and system based under big data and cloud calculation service | |
| CN100549982C (en) | Memory management apparatus, method and system based on the dynamic property monitoring | |
| CN104239183B (en) | The method and device of monitoring system performance | |
| CN106844161A (en) | Abnormal monitoring and Forecasting Methodology and system in a kind of carrier state stream calculation system | |
| CN107528870B (en) | A kind of collecting method and its equipment | |
| CN109491850A (en) | A kind of disk failure prediction technique and device | |
| US20130226922A1 (en) | Identification of Complementary Data Objects | |
| CN103684916A (en) | Method and system for intelligent monitoring and analyzing under cloud computing | |
| CN103384206A (en) | Concurrent processing method and system for mass data | |
| CN102081622A (en) | Method and device for evaluating system health degree | |
| CN104965861A (en) | Monitoring device for data access | |
| CN107786616A (en) | Main frame intelligent monitor system based on high in the clouds | |
| CN104572401A (en) | Alarming method and alarming system | |
| CN103914363B (en) | A kind of internal memory monitoring method and relevant apparatus | |
| US10089236B2 (en) | Apparatus and method of performing agentless remote IO catching analysis, prediction, automation, and recommendation in a computer environment | |
| US20220300345A1 (en) | Entity and Method Performed therein for Handling Computational Resources | |
| CN109446017A (en) | A kind of alarm algorithm generation method, monitoring system and terminal device | |
| CN104477776B (en) | The crane remote hierarchical monitoring system of based role | |
| CN104375954A (en) | Implementing dynamic cache enabling and disabling based upon workload | |
| CN113297046A (en) | Early warning method and device for memory fault | |
| CN109800130A (en) | A kind of apparatus monitoring method, device, equipment and medium | |
| CN110515938B (en) | Data aggregation storage method, equipment and storage medium based on KAFKA message bus | |
| CN104702654A (en) | Video cloud storage system based storage and extraction performance balancing method and device | |
| CN107277143A (en) | A kind of resource matched management method and device |
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: 20180410 |
|
| RJ01 | Rejection of invention patent application after publication |