CN105207812B - A kind of cloud computing resources Forecasting Methodology and system based on business model - Google Patents
A kind of cloud computing resources Forecasting Methodology and system based on business model Download PDFInfo
- Publication number
- CN105207812B CN105207812B CN201510537328.1A CN201510537328A CN105207812B CN 105207812 B CN105207812 B CN 105207812B CN 201510537328 A CN201510537328 A CN 201510537328A CN 105207812 B CN105207812 B CN 105207812B
- Authority
- CN
- China
- Prior art keywords
- demand
- resources
- virtual machine
- requirement model
- resources requirement
- 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.)
- Active
Links
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of cloud computing resources Forecasting Methodology based on business model, including：The demand information of the target service system corresponding to user is obtained, the demand information includes the application time of the type of service of the affiliated operation system of virtual machine of user's application, virtual machine purposes of the virtual machine in operation system of user's application, the business scale of the affiliated operation system of virtual machine of user's application and the virtual machine of user's application；Resources requirement model storehouse is matched according to the demand information, obtains the resource quantity that target service system should apply.The invention also discloses a kind of cloud computing resources forecasting system based on business model.Using the present invention, by carrying out rational resource allocation according to Resources requirement model in application, improve accuracy during first sub-distribution resource, so as to reduce client cost, business continuance is improved, prior art distribution is avoided and adjusts the shortcomings that wasting of resources brought is with influencing business continuance again.
Description
Technical field
The present invention relates to field of cloud calculation, more particularly to a kind of cloud computing resources Forecasting Methodology and one based on business model
Cloud computing resources forecasting system of the kind based on business model.
Background technology
Cloud computing is allowed users to as using water, electricity, enjoys information resources service on demand.In IAAS
（Infrastructure as a Service are based on framework and service）Layer, cloud computing resources are embodied in CPU, interior in distribution
Deposit, the division of the resource such as network bandwidth, memory capacity.
The first sub-distribution of cloud computing resources needs user to file an application at present, administrative staff's examination ＆ verification and manual assignment.For the first time
After being assigned, the cloud management platform of mainstream generally provides the resource re-allocation based on monitoring.Such as the DRS of VMware
（Distributed Resource Scheduler distributed resource schedulings）, by the utilization for continuously monitoring resource pool
Rate, distributes suitable resource automatically in virtual machine as needed, in this way dynamically distributes and balance resource, makes the money of virtual machine
Source is distributed realizes opposite match with resource requirement.
But since user lacks accurately understanding to itself required information resources quantity, often application is remote super real
The resource that border needs, administrator is in the case where lacking data supporting, it is also difficult to effective suggestion is provided, only by application quantity point
Match somebody with somebody.Therefore, the resource requirement of distortion at initial stage can make capacity planning have relatively large deviation, cause to waste.
In addition, optimal way has two kinds after distribution, a kind of is the reality that administrator manually monitors customer service by cloud platform
Border resource requirement, adjustment user resources distribution；A kind of is the resource re-allocation function based on monitoring by cloud platform, by system
Adjust automatically user resources are distributed.But due to the limitation of virtualization technology and operating system, it is most of to reduce resource allocation behaviour
Make and the increase resource allocation operations of a part need to carry out offline, this can cause service disconnection.
The content of the invention
The technical problems to be solved by the invention are, there is provided a kind of cloud computing resources Forecasting Methodology based on business model
And system, accuracy during first sub-distribution resource can be improved, reduces client cost, improves business continuance.
In order to solve the above technical problem, the present invention provides a kind of cloud computing resources prediction side based on business model
Method, including：The demand information of the target service system corresponding to user is obtained, the demand information includes the virtual of user's application
The type of service of the affiliated operation system of machine, the virtual machine of user's application virtual machine purposes in operation system, user's application
The application time of the business scale of the affiliated operation system of virtual machine and the virtual machine of user's application；Matched according to the demand information
Resources requirement model storehouse, obtains the resource quantity that target service system should apply.
As the improvement of such scheme, Resources requirement model is stored with the Resources requirement model storehouse；The resource needs
Element information in modulus type includes type of service, business scale, virtual machine purposes, cpu demand, memory requirements, memory space
Demand, storage IO demands, network I/O demand, time to peak and peak demand growth ratio.
As the improvement of such scheme, the step of Resources requirement model storehouse of information matches according to demand, includes：Providing
Retrieval has the Resources requirement model of identical services type with target service system on the demand model storehouse of source, obtains and matches for the first time
As a result；Retrieval has the Resources requirement model of identical virtual machine purposes with target service system in first time matching result, obtains
Obtain second of matching result；Retrieval is needed with the resource that target service system has identical services scale in second of matching result
Modulus type, obtains third time matching result；The third time matching result is screened, extracts optimal Resources requirement model；
According to the element information corresponding to the optimal Resources requirement model, extraction standard quantity required, the standard requirement quantity bag
Include cpu demand quantity, memory requirements quantity, memory space requirements quantity, storage IO quantity requireds and network I/O quantity required；Sentence
Whether the application time of disconnected target service system in the time to peak corresponding to optimal Resources requirement model, is judged as NO
When, the standard requirement quantity is the resource quantity that should apply, when being judged as YES, according to corresponding to optimal Resources requirement model
Peak demand growth ratio, calculates the resource quantity that should apply.
As the improvement of such scheme, the cloud computing resources Forecasting Methodology based on business model further includes：Building resource needs
Seek model library；The method in the structure Resources requirement model storehouse includes：Gather the basic information of initial service system, the basis
Information is needed including type of service, number of users, virtual machine purposes, cpu demand, memory requirements, memory space requirements, storage IO
Ask, memory space needs during memory requirements, peak value during cpu demand, peak value during network I/O demand, time to peak, peak value
Ask, network I/O demand during IO demands and peak value is stored during peak value；Business scale is divided according to number of users；According to the base
Plinth information calculates peak demand growth ratio；According to the basic information, business scale and peak demand growth ratio, extraction member
Prime information；Resources requirement model storehouse is built according to the element information.
Correspondingly, present invention also offers a kind of cloud computing resources forecasting system based on business model, including cloud computing
Resources device and Resources requirement model storehouse；The cloud computing resources prediction meanss include：Acquiring unit, for obtaining user
The demand information of corresponding target service system, the demand information include the affiliated operation system of virtual machine of user's application
Type of service, virtual machine purposes of the virtual machine in operation system of user's application, the affiliated business system of virtual machine of user's application
The application time of the business scale of system and the virtual machine of user's application；Matching unit, provides for being matched according to the demand information
Source demand model storehouse, obtains the resource quantity that target service system should apply；
As the improvement of such scheme, Resources requirement model is stored with the Resources requirement model storehouse；The resource needs
Element information in modulus type includes type of service, business scale, virtual machine purposes, cpu demand, memory requirements, memory space
Demand, storage IO demands, network I/O demand, time to peak and peak demand growth ratio.
As the improvement of such scheme, the matching unit includes：First matching unit, in Resources requirement model storehouse
Upper retrieval has the Resources requirement model of identical services type with target service system, obtains first time matching result；Second
With unit, for the retrieval in first time matching result and resource requirement mould of the target service system with identical virtual machine purposes
Type, obtains second of matching result；3rd matching unit, for the retrieval in second of matching result and target service system tool
There is the Resources requirement model of identical services scale, obtain third time matching result；Screening unit, for being matched to the third time
As a result screened, extract optimal Resources requirement model；Extraction unit, for according to corresponding to the optimal Resources requirement model
Element information, extraction standard quantity required, the standard requirement quantity include cpu demand quantity, memory requirements quantity, storage
Space requirement quantity, storage IO quantity requireds and network I/O quantity required；Judging unit, for judging the Shen of target service system
Please whether be in the time to peak corresponding to optimal Resources requirement model the time, when being judged as NO, the standard requirement quantity
The resource quantity that should as apply, when being judged as YES, according to peak demand growth ratio corresponding to optimal Resources requirement model, meter
Calculate the resource quantity that should apply.
As the improvement of such scheme, it is single that the cloud computing resources prediction meanss further include Resources requirement model storehouse structure
Member；Resources requirement model storehouse construction unit includes：Collecting unit, for gathering the basic information of initial service system, institute
Stating basic information includes type of service, number of users, virtual machine purposes, cpu demand, memory requirements, memory space requirements, deposits
Stored during storage IO demands, network I/O demand, time to peak, peak value during cpu demand, peak value during memory requirements, peak value
Network I/O demand during IO demands and peak value is stored during space requirement, peak value；Division unit, for being divided according to number of users
Business scale；Ratio computing unit, for calculating peak demand growth ratio according to the basic information；Element extraction unit,
According to the basic information, business scale and peak demand growth ratio, element information is extracted；Construction unit, for according to institute
State element information structure Resources requirement model storehouse.
Implement the present invention, have the advantages that：
Cloud computing resources Forecasting Methodology of the invention based on business model, user is in application, i.e., according to resource requirement mould
Type carries out rational resource allocation, improves accuracy during first sub-distribution resource, so as to reduce client cost, improves business and connects
Continuous property, avoids prior art distribution and adjusts the shortcomings that wasting of resources brought is with influencing business continuance again.
In addition, the Resources requirement model of the invention by building industry universal, provides perfect for cloud computing resources prediction
Matching basis.
Brief description of the drawings
Fig. 1 is the first example flow chart of the cloud computing resources Forecasting Methodology of the invention based on business model；
Fig. 2 is the second example flow chart of the cloud computing resources Forecasting Methodology of the invention based on business model；
Fig. 3 is the flow that Resources requirement model storehouse is built in the cloud computing resources Forecasting Methodology of the invention based on business model
Figure；
Fig. 4 is the structure diagram of the cloud computing resources forecasting system of the invention based on business model；
Fig. 5 is the structure diagram of matching unit in Fig. 4；
Fig. 6 is another structure diagram of the cloud computing resources forecasting system of the invention based on business model；
Fig. 7 is the structure diagram of Resources requirement model storehouse construction unit in Fig. 6.
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 attached 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 attached drawing of the present invention, it is not that the specific of the present invention is limited.
Fig. 1 is the first example flow chart of the cloud computing resources Forecasting Methodology of the invention based on business model, including：
S101, obtains the demand information of the target service system corresponding to user.
The type of service of virtual machine affiliated operation system of the demand information including user's application, user are applied virtual
Virtual machine purposes of the machine in operation system, the business scale of the affiliated operation system of virtual machine of user's application and user's application
The application time of virtual machine.
It should be noted that when user is target service system application resource, industry belonging to the virtual machine of application need to be uploaded
The type of service of business system, business scale, virtual machine purposes of the virtual machine in operation system, and application time.
S102, Resources requirement model storehouse is matched according to the demand information, obtains the resource that target service system should apply
Quantity.
Resources requirement model is stored with the Resources requirement model storehouse.Element information bag in the Resources requirement model
Include type of service, business scale, virtual machine purposes, cpu demand, memory requirements, memory space requirements, storage IO demands, network
IO demands, time to peak and peak demand growth ratio.Wherein, type of service includes：School business management class, teaching and scientific research class, recruit
Adult Students ' Employment class, integrated service class, financial management, asset management, personnel management, logistics management, student education work management, student
Physical health data management, file administration, Party affairs, office and issued transaction, official document and information exchange, reform in education pipe
Reason, subject, specialized management, educational administration's teaching management, teaching resource management, Teaching Quality Assessment and guarantee, scientific research project management, section
Grind information management, unification of the motherland recruit management, ordinary higher learning school's enrollment enrolment on internet management, school's autonomous enrolment management, student just
Industry management, portal website, forum, community class website, digital library, Email, Video service, safety monitoring, campus one
Cartoon, Intranet door and authentication, public database, operation management etc., but it is not limited system.Virtual machine purposes includes：
HTTP server, application server and database server, but it is not limited system.Cpu demand refers to that user is actually using
The CPU abilities of middle requirement, such as 32 threads, 64 threads etc..Business scale refers to the range intervals of number of users.
It should be noted that the element information in demand model storehouse is to be investigated to obtain with continuing to monitor according to long-term data
, different types of service, different virtual machine purposes, different business scale, corresponding different resource requirement, meanwhile, it is every kind of
Resource requirement growth ratio when type of service all has corresponding traffic peak time and peak value.
The prior art is for the first time after the completion of resource allocation, and by monitoring the utilization rate of resource, operation conditions carries out again
Distribution.Compared with prior art, the present invention in advance build industry universal Resources requirement model, by application when according to money
Source demand model carries out rational resource allocation, without sub-distribution again, accuracy during first sub-distribution resource is improved, so as to drop
Low client cost, improves business continuance, meanwhile, avoid prior art distribution and adjust the wasting of resources and influence brought again
The shortcomings that business continuance.
Fig. 2 is the second example flow chart of the cloud computing resources Forecasting Methodology of the invention based on business model, including：
S201, obtains the demand information of the target service system corresponding to user.
The type of service of virtual machine affiliated operation system of the demand information including user's application, user are applied virtual
Virtual machine purposes of the machine in operation system, the business scale of the affiliated operation system of virtual machine of user's application and user's application
The application time of virtual machine.
S202, retrieval and resource requirement mould of the target service system with identical services type on Resources requirement model storehouse
Type, obtains first time matching result.
It should be noted that it is stored with Resources requirement model in the Resources requirement model storehouse.The Resources requirement model
In element information include type of service, business scale, virtual machine purposes, cpu demand, memory requirements, memory space requirements, deposit
Store up IO demands, network I/O demand, time to peak and peak demand growth ratio.
S203, retrieval and resource requirement of the target service system with identical virtual machine purposes in first time matching result
Model, obtains second of matching result.
S204, retrieval and resource requirement mould of the target service system with identical services scale in second of matching result
Type, obtains third time matching result.
S205, screens the third time matching result, extracts optimal Resources requirement model.
It should be noted that if after step S202, step 203 and step 204, comprising more in third time matching result
During a Resources requirement model, then advantage distillation resource Resources requirement model the most sufficient, unique optimal money is obtained with guarantee
Source demand model.
S206, according to the element information corresponding to the optimal Resources requirement model, extraction standard quantity required.
When extracting optimal Resources requirement model, can the optimal Resources requirement model as referring to standard, from optimal
In element information corresponding to Resources requirement model, standard requirement quantity is extracted.The standard requirement quantity includes cpu demand
Quantity, memory requirements quantity, memory space requirements quantity, storage IO quantity requireds and network I/O quantity required.
S207, when judging whether the application time of target service system is in the peak value corresponding to optimal Resources requirement model
In.When being judged as NO, the standard requirement quantity is the resource quantity that should apply；When being judged as YES, according to optimal resource
Peak demand growth ratio corresponding to demand model, calculates the resource quantity that should apply.
The application time of target service system is compared with the time to peak corresponding to optimal Resources requirement model, is sentenced
Whether the application time of disconnected target service system is in the time to peak corresponding to optimal Resources requirement model.If during application
Between be not in time to peak, then standard requirement quantity without correct, standard requirement quantity is the resource quantity that should apply；
If the application time is in time to peak, the peak demand growth ratio according to corresponding to optimal Resources requirement model is needed, is counted
The resource quantity that should apply is calculated, wherein, the resource quantity that should apply=standard requirement quantity ×（1+ peak demands increase ratio
Example）.
It should be noted that peak demand growth ratio includes cpu demand growth ratio, memory requirements growth ratio, deposits
Store up space requirement growth ratio, storage IO demand growth ratio and network I/O demand growth ratio.
Below in conjunction with specific example, the present invention is described in further detail.
Example：User needs application one to be used for the virtual machine for disposing " all-in-one campus card ".
A1, the demand information for obtaining user, wherein, type of service is " all-in-one campus card ", and virtual machine purposes is " application clothes
Business device ", business scale are " 5000 ~ 10000 people ", and the application time is " 8:00”.
A2, retrieves the Resources requirement model that type of service is " all-in-one campus card ", obtains first time matching result；
A3, the Resources requirement model that virtual machine purposes is " application server " is retrieved in first time matching result, is obtained
Obtain second of matching result；
A4, the Resources requirement model that business scale is " 5000 ~ 10000 people " is retrieved in second of matching result, is obtained
Obtain third time matching result；
A5, screens the third time matching result, extracts optimal Resources requirement model;
A6, according to the element information corresponding to the optimal Resources requirement model, extraction standard quantity required;
A7, by the application time " 8:00 " is compared with the time to peak corresponding to optimal Resources requirement model, judges
Application time is not in time to peak, and standard requirement quantity is the resource quantity that should apply at this time.
Therefore, the present invention in advance build industry universal Resources requirement model, by application when according to resource requirement
Model carries out rational resource allocation, without sub-distribution again, accuracy during first sub-distribution resource is improved, so as to reduce client
Cost, improves business continuance, meanwhile, avoid prior art distribution and adjust the wasting of resources brought and influence business company again
The shortcomings that continuous property.
It should be noted that also need to build Resources requirement model storehouse in advance before step S201.
Specifically, the method in the structure Resources requirement model storehouse includes：
S301, the basic information for gathering initial service system.
The basic information includes type of service, number of users, virtual machine purposes, cpu demand, memory requirements, storage sky
Between memory requirements, peak during cpu demand, peak value during demand, storage IO demands, network I/O demand, time to peak, peak value
Network I/O demand during IO demands and peak value is stored during value during memory space requirements, peak value.
Basic information in Resources requirement model storehouse comes from a large amount of investigations to each profession system, different industry
Service type, different virtual machine purposes, different business scale, corresponding different resource requirement.
S302, according to number of users divide business scale.
Business scale refers to the range intervals of number of users.Preferably, every kind of type of service can be divided into 7 according to number of users
Kind scale, so that it is determined that business scale, but it is not limited system.
S303, according to the basic information calculate peak demand growth ratio.
It should be noted that peak demand growth ratio can be used for when system enters peak period, automatically to virtual machine
Various resources be automatically adjusted, to meet the resource requirement of peak period.Wherein, peak demand growth ratio is needed including CPU
Growth ratio, memory requirements growth ratio, memory space requirements growth ratio, storage IO demand growth ratio and network I/O is asked to need
Seek growth ratio.
Specifically, subtract cpu demand again with cpu demand during peak value divided by cpu demand can obtain cpu demand and increase ratio
Example；Memory requirements is subtracted with memory requirements during peak value again divided by memory requirements can obtain memory requirements growth ratio；Use peak
Memory space requirements subtract memory space requirements again during value divided by memory space requirements can obtain memory space requirements growth
Ratio；With being stored during peak value, IO demands subtract storage IO demands again divided by storage IO demands can obtain storage IO demand growth
Ratio；Network I/O demand is subtracted with network I/O demand during peak value again divided by network I/O demand can obtain network I/O demand growth
Ratio.
S304, according to the basic information, business scale and peak demand growth ratio, extract element information.
Element information in the Resources requirement model include type of service, business scale, virtual machine purposes, cpu demand,
Memory requirements, memory space requirements, storage IO demands, network I/O demand, time to peak and peak demand growth ratio.
S305, according to the element information build Resources requirement model storehouse.
Therefore, multiple independent Resources requirement models can be built by step S301 ~ 305.Wherein, different service class
Type, different virtual machine purposes, different business scale correspond to different resource requirements respectively, meanwhile, every kind of type of service has
Resource requirement growth ratio when the traffic peak time of oneself and peak value.
Resources requirement model is stored in the form of database table can form Resources requirement model storehouse.
From the foregoing, it will be observed that the Resources requirement model of the invention by building industry universal in advance, carries for cloud computing resources prediction
For perfect matching basis.Meanwhile by carrying out rational resource allocation according to Resources requirement model in application, improve
Accuracy during first sub-distribution resource, so as to reduce client cost, improves business continuance, avoids prior art distribution again
Adjust the shortcomings that wasting of resources brought is with influencing business continuance.
Fig. 4 is the structure diagram of the cloud computing resources forecasting system 100 of the invention based on business model, described to be based on industry
The cloud computing resources forecasting system 100 of business model includes：Cloud computing resources prediction meanss 1 and Resources requirement model storehouse 2.
Wherein：
Resources requirement model is stored with the Resources requirement model storehouse.Element information bag in the Resources requirement model
Include type of service, business scale, virtual machine purposes, cpu demand, memory requirements, memory space requirements, storage IO demands, network
IO demands, time to peak and peak demand growth ratio.Element information in demand model storehouse is according to long-term data investigation
Obtained with continuing to monitor, different types of service, different virtual machine purposes, different business scale, corresponding different resource needs
Ask, meanwhile, resource requirement growth ratio when every kind of type of service all has corresponding traffic peak time and peak value.
The cloud computing resources prediction meanss 1 include：Acquiring unit 11 and matching unit 12.
Acquiring unit 11, for obtaining the demand information of the target service system corresponding to user.
The type of service of virtual machine affiliated operation system of the demand information including user's application, user are applied virtual
Virtual machine purposes of the machine in operation system, the business scale of the affiliated operation system of virtual machine of user's application and user's application
The application time of virtual machine.When user is target service system application resource, business system belonging to the virtual machine of application need to be uploaded
The type of service of system, business scale, virtual machine purposes of the virtual machine in operation system, and application time.
Matching unit 12, for matching Resources requirement model storehouse according to the demand information, obtaining target service system should
The resource quantity of application.
In the present invention, the Resources requirement model in Resources requirement model storehouse is matched by matching unit 12, so that
Reasonably realize resource allocation, without sub-distribution again, improve accuracy during first sub-distribution resource, thus reduce client into
This, improves business continuance, meanwhile, avoid prior art distribution and adjust the wasting of resources brought again with influencing business continuity
The shortcomings that property.
As shown in figure 5, the matching unit 12 includes：
First matching unit 121, has identical services for being retrieved on Resources requirement model storehouse with target service system
The Resources requirement model of type, obtains first time matching result.
Second matching unit 122, for being retrieved in first time matching result with target service system with identical virtual
The Resources requirement model of machine purposes, obtains second of matching result.
3rd matching unit 123, has identical services for being retrieved in second of matching result with target service system
The Resources requirement model of scale, obtains third time matching result.
Screening unit 124, for being screened to the third time matching result, extracts optimal Resources requirement model.
It should be noted that pass through the first matching unit 121, the second matching unit 122 and the 3rd matching unit 123 successively
Matching after, if multiple Resources requirement models are included in third time matching result, pass through 124 advantage distillation of screening unit provide
Source Resources requirement model the most sufficient, to ensure to obtain unique optimal Resources requirement model.
Extraction unit 125, for the element information according to corresponding to the optimal Resources requirement model, extraction standard demand
Quantity.
When extracting optimal Resources requirement model, can the optimal Resources requirement model as referring to standard, from optimal
In element information corresponding to Resources requirement model, standard requirement quantity is extracted.The standard requirement quantity includes cpu demand
Quantity, memory requirements quantity, memory space requirements quantity, storage IO quantity requireds and network I/O quantity required.
Whether judging unit 126, the application time for judging target service system are in optimal Resources requirement model institute
In corresponding time to peak, when being judged as NO, the standard requirement quantity is the resource quantity that should apply, when being judged as YES,
According to peak demand growth ratio corresponding to optimal Resources requirement model, the resource quantity that should apply is calculated, wherein, it should apply
Resource quantity=standard requirement quantity ×（1+ peak demand growth ratios）.
It should be noted that peak demand growth ratio includes cpu demand growth ratio, memory requirements growth ratio, deposits
Store up space requirement growth ratio, storage IO demand growth ratio and network I/O demand growth ratio.
Fig. 6 is another structure diagram of the cloud computing resources forecasting system 100 of the invention based on business model, with Fig. 4
Unlike the shown cloud computing resources forecasting system 100 based on business model, the cloud computing resources prediction meanss 1 are also wrapped
Include Resources requirement model storehouse construction unit 13.
As shown in fig. 7, Resources requirement model storehouse construction unit 13 includes：
Collecting unit 131, for gathering the basic information of initial service system.
Basic information in Resources requirement model storehouse comes from a large amount of investigations to each profession system, different industry
Service type, different virtual machine purposes, different business scale, corresponding different resource requirement.The basic information includes business
Type, number of users, virtual machine purposes, cpu demand, memory requirements, memory space requirements, storage IO demands, network I/O need
Ask, deposited during time to peak, peak value during cpu demand, peak value during memory requirements, peak value during memory space requirements, peak value
Network I/O demand during storage IO demands and peak value.
Division unit 132, for dividing business scale according to number of users.
Business scale refers to the range intervals of number of users.Preferably, every kind of type of service can be divided into 7 according to number of users
Kind scale, so that it is determined that business scale, but it is not limited system.
Ratio computing unit 133, for calculating peak demand growth ratio according to the basic information.
Wherein, peak demand growth ratio, which includes cpu demand growth ratio, memory requirements growth ratio, memory space, needs
Ask growth ratio, storage IO demand growth ratio and network I/O demand growth ratio.
Specifically, subtract cpu demand again with cpu demand during peak value divided by cpu demand can obtain cpu demand and increase ratio
Example；Memory requirements is subtracted with memory requirements during peak value again divided by memory requirements can obtain memory requirements growth ratio；Use peak
Memory space requirements subtract memory space requirements again during value divided by memory space requirements can obtain memory space requirements growth
Ratio；With being stored during peak value, IO demands subtract storage IO demands again divided by storage IO demands can obtain storage IO demand growth
Ratio；Network I/O demand is subtracted with network I/O demand during peak value again divided by network I/O demand can obtain network I/O demand growth
Ratio.
Element extraction unit 134, according to the basic information, business scale and peak demand growth ratio, extracts element
Information.
Element information in the Resources requirement model include type of service, business scale, virtual machine purposes, cpu demand,
Memory requirements, memory space requirements, storage IO demands, network I/O demand, time to peak and peak demand growth ratio.
Construction unit 135, for building Resources requirement model storehouse according to the element information.
Pass sequentially through collecting unit 131, division unit 132, ratio computing unit 133 and element extraction unit 134
Build multiple independent Resources requirement models.Wherein, different type of service, different virtual machine purposes, different business scale
Different resource requirements is corresponded to respectively, meanwhile, resource needs when every kind of type of service has traffic peak time and the peak value of oneself
Seek growth ratio.Construction unit 135 stores Resources requirement model in the form of database table can form Resources requirement model storehouse.
From the foregoing, it will be observed that the present invention builds the resource requirement of industry universal by Resources requirement model storehouse construction unit 13 in advance
Model, perfect matching basis is provided for cloud computing resources prediction.Meanwhile by matching unit 12 application when according to resource
Demand model carries out rational resource allocation, improves accuracy during first sub-distribution resource, so as to reduce client cost, improves
Business continuance, avoids prior art distribution and adjusts the shortcomings that wasting of resources brought is with influencing business continuance again.
The above is the preferred embodiment of the present invention, it is noted that for those skilled in the art
For, various improvements and modifications may be made without departing from the principle of the present invention, these improvements and modifications are also considered as
Protection scope of the present invention.
Claims (4)
- A kind of 1. cloud computing resources Forecasting Methodology based on business model, it is characterised in that including：The demand information of the target service system corresponding to user is obtained, the demand information includes the virtual machine institute of user's application Belong to the type of service of operation system, virtual machine purposes of the virtual machine in operation system of user's application, user's application it is virtual The application time of the business scale of the affiliated operation system of machine and the virtual machine of user's application；Resources requirement model storehouse is matched according to the demand information, obtains the resource quantity that target service system should apply；Resources requirement model is stored with the Resources requirement model storehouse；Element information in the Resources requirement model includes industry Service type, business scale, virtual machine purposes, cpu demand, memory requirements, memory space requirements, storage IO demands, network I/O need Ask, time to peak and peak demand growth ratio；The step of Resources requirement model storehouse of information matches according to demand, includes：Retrieval and target on Resources requirement model storehouse Operation system has the Resources requirement model of identical services type, obtains first time matching result；In first time matching result The Resources requirement model that there is identical virtual machine purposes with target service system is retrieved, obtains second of matching result；Second Retrieval has the Resources requirement model of identical services scale with target service system in secondary matching result, obtains third time matching knot Fruit；The third time matching result is screened, extracts optimal Resources requirement model；According to the optimal resource requirement mould Element information corresponding to type, extraction standard quantity required, the standard requirement quantity include cpu demand quantity, memory requirements Quantity, memory space requirements quantity, storage IO quantity requireds and network I/O quantity required；When judging the application of target service system Between whether in the time to peak corresponding to the optimal Resources requirement model, when being judged as NO, the standard requirement quantity is The resource quantity that should apply, when being judged as YES, according to peak demand growth ratio corresponding to optimal Resources requirement model, calculates The resource quantity that should apply.
- 2. the cloud computing resources Forecasting Methodology based on business model as claimed in claim 1, it is characterised in that further include：Structure Build Resources requirement model storehouse；The method in the structure Resources requirement model storehouse includes：Gather initial service system basic information, the basic information include type of service, number of users, virtual machine purposes, CPU during cpu demand, memory requirements, memory space requirements, storage IO demands, network I/O demand, time to peak, peak value Store network I/O during IO demands and peak value during demand, peak value during memory requirements, peak value during memory space requirements, peak value Demand；Business scale is divided according to number of users；Peak demand growth ratio is calculated according to the basic information；According to the basic information, business scale and peak demand growth ratio, element information is extracted；Resources requirement model storehouse is built according to the element information.
- 3. a kind of cloud computing resources forecasting system based on business model, it is characterised in that including cloud computing resources prediction meanss And Resources requirement model storehouse；The cloud computing resources prediction meanss include：Acquiring unit, for obtaining the demand information of the target service system corresponding to user, the demand information includes user Virtual machine purposes in operation system of the type of service of the affiliated operation system of virtual machine of application, the virtual machine of user's application, The application time of the business scale of the affiliated operation system of virtual machine of user's application and the virtual machine of user's application；Matching unit, for matching Resources requirement model storehouse according to the demand information, obtains what target service system should apply Resource quantity；Resources requirement model is stored with the Resources requirement model storehouse；Element information in the Resources requirement model includes industry Service type, business scale, virtual machine purposes, cpu demand, memory requirements, memory space requirements, storage IO demands, network I/O need Ask, time to peak and peak demand growth ratio；The matching unit includes：First matching unit, for the retrieval on Resources requirement model storehouse and target service system tool There is the Resources requirement model of identical services type, obtain first time matching result；Second matching unit, for being matched in first time As a result middle retrieval has the Resources requirement model of identical virtual machine purposes with target service system, obtains second of matching result； 3rd matching unit, need to resource of the target service system with identical services scale for the retrieval in second of matching result Modulus type, obtains third time matching result；Screening unit, for being screened to the third time matching result, extraction is optimal Resources requirement model；Extraction unit, for the element information according to corresponding to the optimal Resources requirement model, extraction standard needs Quantity is sought, the standard requirement quantity, which includes cpu demand quantity, memory requirements quantity, memory space requirements quantity, storage IO, to be needed Ask quantity and network I/O quantity required；Whether judging unit, the application time for judging target service system are in optimal money In time to peak corresponding to the demand model of source, when being judged as NO, the standard requirement quantity is the resource quantity that should apply, When being judged as YES, according to peak demand growth ratio corresponding to optimal Resources requirement model, the resource quantity that should apply is calculated.
- 4. the cloud computing resources forecasting system based on business model as claimed in claim 3, it is characterised in that the cloud computing Resources device further includes Resources requirement model storehouse construction unit；Resources requirement model storehouse construction unit includes：Collecting unit, for gathering the basic information of initial service system, the basic information includes type of service, number of users When amount, virtual machine purposes, cpu demand, memory requirements, memory space requirements, storage IO demands, network I/O demand, peak value Between, during peak value during cpu demand, peak value during memory requirements, peak value during memory space requirements, peak value store IO demands and Network I/O demand during peak value；Division unit, for dividing business scale according to number of users；Ratio computing unit, for calculating peak demand growth ratio according to the basic information；Element extraction unit, according to the basic information, business scale and peak demand growth ratio, extracts element information；Construction unit, for building Resources requirement model storehouse according to the element information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510537328.1A CN105207812B (en) | 2015-08-28 | 2015-08-28 | A kind of cloud computing resources Forecasting Methodology and system based on business model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510537328.1A CN105207812B (en) | 2015-08-28 | 2015-08-28 | A kind of cloud computing resources Forecasting Methodology and system based on business model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105207812A CN105207812A (en) | 2015-12-30 |
| CN105207812B true CN105207812B (en) | 2018-04-17 |
Family
ID=54955265
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510537328.1A Active CN105207812B (en) | 2015-08-28 | 2015-08-28 | A kind of cloud computing resources Forecasting Methodology and system based on business model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105207812B (en) |
Families Citing this family (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106502796A (en) * | 2016-10-27 | 2017-03-15 | 广东浪潮大数据研究有限公司 | A kind of software development test platform, system and using method |
| US11095526B2 (en) * | 2016-11-14 | 2021-08-17 | Huawei Technologies Co., Ltd. | System and method for accelerated provision of network services |
| CN106648871B (en) * | 2016-12-28 | 2020-04-03 | 北京奇艺世纪科技有限公司 | Resource management method and system |
| EP3605322A4 (en) * | 2017-03-23 | 2020-03-25 | Nec Corporation | Development operation supporting system, development management server, operation management server, method thereof, and non-transient computer readable medium storing program thereof |
| CN108039962A (en) * | 2017-12-05 | 2018-05-15 | 三盟科技股份有限公司 | A kind of cloud calculation service Resource Calculation method and system under the environment based on big data |
| CN109995677B (en) * | 2018-01-02 | 2022-11-29 | 中国移动通信有限公司研究院 | Resource allocation method, device and storage medium |
| CN110659126B (en) * | 2018-06-29 | 2023-04-14 | 中兴通讯股份有限公司 | Resource management method, device and computer readable storage medium |
| CN109213597A (en) * | 2018-08-01 | 2019-01-15 | 平安科技（深圳）有限公司 | Resource allocation method, apparatus, computer device, and computer-readable storage medium |
| CN110262869B (en) * | 2018-10-30 | 2020-05-15 | 湖北省农村信用社联合社网络信息中心 | Private cloud Paas resource supply and demand party information interaction method |
| CN109901928B (en) * | 2019-03-01 | 2022-11-29 | 厦门星界链科技有限公司 | Method for recommending cloud host configuration and cloud host |
| CN112737798B (en) * | 2019-10-14 | 2022-09-27 | 中国移动通信集团四川有限公司 | Host resource allocation method and device, scheduling server and storage medium |
| CN111381970B (en) * | 2020-03-16 | 2023-07-25 | 第四范式（北京）技术有限公司 | Cluster task resource allocation method and device, computer device and storage medium |
| CN114205249B (en) * | 2020-09-02 | 2023-07-18 | 中国联合网络通信集团有限公司 | Resource allocation method and terminal equipment |
| CN115766336B (en) * | 2022-09-27 | 2024-10-11 | 中国联合网络通信集团有限公司 | Resource allocation method, device, equipment and storage medium |
| CN117435298B (en) * | 2023-10-16 | 2024-05-28 | 常州大数据有限公司 | Data center energy efficiency intelligent tuning cloud platform based on large language model |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1693763A1 (en) * | 2005-02-18 | 2006-08-23 | International Business Machines Corporation | System, method and computer program product for providing computing service-power to Service-Users via a heterogeneous distributed computing environment |
| WO2013131186A1 (en) * | 2012-03-01 | 2013-09-12 | Cirba Inc. | System and method for providing a capacity reservation system for a virtual or cloud computing environment |
| CN103533037A (en) * | 2013-09-29 | 2014-01-22 | 浙江工商大学 | Resource scheduling method in forwarding and control separation network based on economic model |
| CN104270459A (en) * | 2014-10-20 | 2015-01-07 | 山东省计算中心（国家超级计算济南中心） | A fairness-enhanced cloud computing user resource quota allocation method |
| CN104601664A (en) * | 2014-12-22 | 2015-05-06 | 西安电子科技大学 | Cloud computing platform resource management and virtual machine dispatching control system |
-
2015
- 2015-08-28 CN CN201510537328.1A patent/CN105207812B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1693763A1 (en) * | 2005-02-18 | 2006-08-23 | International Business Machines Corporation | System, method and computer program product for providing computing service-power to Service-Users via a heterogeneous distributed computing environment |
| WO2013131186A1 (en) * | 2012-03-01 | 2013-09-12 | Cirba Inc. | System and method for providing a capacity reservation system for a virtual or cloud computing environment |
| CN103533037A (en) * | 2013-09-29 | 2014-01-22 | 浙江工商大学 | Resource scheduling method in forwarding and control separation network based on economic model |
| CN104270459A (en) * | 2014-10-20 | 2015-01-07 | 山东省计算中心（国家超级计算济南中心） | A fairness-enhanced cloud computing user resource quota allocation method |
| CN104601664A (en) * | 2014-12-22 | 2015-05-06 | 西安电子科技大学 | Cloud computing platform resource management and virtual machine dispatching control system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105207812A (en) | 2015-12-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105207812B (en) | A kind of cloud computing resources Forecasting Methodology and system based on business model | |
| CN105045818B (en) | A kind of recommendation methods, devices and systems of picture | |
| RU2018132713A (en) | METHOD AND SYSTEM FOR DETERMINING USER-DEPENDING PROPORTIONS OF CONTENTS FOR RECOMMENDATION | |
| CN109191202B (en) | Resource allocation method, device, electronic equipment and computer readable storage medium | |
| CN107360336A (en) | A kind of intelligent distribute leaflets system and method based on voice service | |
| CN108520329B (en) | Accurate automatic allocation method and device for second-hand house clients based on broker portrait | |
| Jain et al. | Cloud computing with ERP-A push business towards higher efficiency | |
| CN113919738A (en) | Business handling window distribution method and device, electronic equipment and readable storage medium | |
| US8219368B1 (en) | Capacity modeling system | |
| KR102194683B1 (en) | Method and appratus for managing legal counseling schedule | |
| CN114529209A (en) | User allocation method, device, equipment and storage medium | |
| CN110363434A (en) | Determine that self-help teller machine launches the method and device in place | |
| Lacasandile et al. | Development of an information-based dashboard: automation of barangay information profiling system (BIPS) for decision support towards e-governance | |
| CN115719270A (en) | Credit evaluation method, device, apparatus, medium, and program product | |
| CN109902196A (en) | A kind of trademark class recommended method, device, computer equipment and storage medium | |
| CN117252362A (en) | Scheduling method and device based on artificial intelligence, computer equipment and storage medium | |
| TWM605348U (en) | Financial Commodity Recommendation System | |
| CN114330847A (en) | Data processing method and related device for bill scoring calculation | |
| CN113190587A (en) | Data processing method and device for realizing service data processing | |
| CN106130757A (en) | Information getting method and device | |
| US12254030B1 (en) | Systems and methods of generating and transforming data sets for entity comparison | |
| TWI756804B (en) | Financial product recommendation system and method | |
| US20200151256A1 (en) | Conversation content generation based on user professional level | |
| Stephen et al. | Cloud Computing: Opportunities and Challenges | |
| CN112862226A (en) | Information processing method, information processing apparatus, information processing medium, and electronic device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP01 | Change in the name or title of a patent holder | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 510075 Guangdong city of Guangzhou province Yuexiu District Shuiyin Road No. 56 Building 1 room 6A (office use only) Patentee after: Guangdong three league Polytron Technologies Inc. Address before: 510075 Guangdong city of Guangzhou province Yuexiu District Shuiyin Road No. 56 Building 1 room 6A (office use only) Patentee before: GUANGDONG SANMENG INFORMATION TECHNOLOGY Co.,Ltd. |
|
| CP03 | Change of name, title or address | ||
| CP03 | Change of name, title or address |
Address after: 510095 206, 1 building 20, Taihe Gang Road, Yuexiu District, Guangzhou, Guangdong. Patentee after: SUNMNET TECHNOLOGY CO.,LTD. Address before: 510075 Guangdong city of Guangzhou province Yuexiu District Shuiyin Road No. 56 Building 1 room 6A (office use only) Patentee before: Guangdong three league Polytron Technologies Inc. |
|
| PE01 | Entry into force of the registration of the contract for pledge of patent right | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: Cloud computing resource prediction method and system based on business model Effective date of registration: 20210205 Granted publication date: 20180417 Pledgee: China Co. truction Bank Corp Guangzhou Liwan branch Pledgor: SUNMNET TECHNOLOGY Co.,Ltd. Registration number: Y2021980001059 |
|
| PP01 | Preservation of patent right | ||
| PP01 | Preservation of patent right |
Effective date of registration: 20230919 Granted publication date: 20180417 |