CN106878451A

CN106878451A - A kind of based on performance requirement and sequence cloud monitoring system and construction method

Info

Publication number: CN106878451A
Application number: CN201710147376.9A
Authority: CN
Inventors: 张彦彬; 林铭杰; 叶政晟
Original assignee: Guangzhou High Energy Computer Technology Co Ltd
Current assignee: Guangzhou High Energy Computer Technology Co Ltd
Priority date: 2016-12-01
Filing date: 2017-03-13
Publication date: 2017-06-20
Anticipated expiration: 2037-03-13
Also published as: CN106878451B

Abstract

The invention provides a kind of based on performance requirement and sequence cloud monitoring system, including：Computing cluster, monitoring module and simultaneously sequence module；Wherein, the computing cluster includes at least one physical node；Described and sequence module builds to instruct according to default parameter, the hardware parameter of each physical node and default virtual machine and build at least one virtual machine on physical node described at least one；Present invention also offers a kind of based on performance requirement and sequence cloud monitoring system construction method, instruction can be built according to default virtual machine virtual machine is distributed on multiple physical nodes in several ways, so that the performance of virtual machine better meets the mission requirements of user, allow users to more flexibly, efficiently arrange virtual machine.

Description

A kind of based on performance requirement and sequence cloud monitoring system and construction method

Patent Office of the People's Republic of China is submitted to this application claims on December 1st, 2016, application number 201611088969.4 is entitled The priority of the Chinese patent application of " a kind of based on performance requirement and sequence cloud monitoring system and construction method ", it is all interior Appearance is hereby incorporated by reference in the application.

Technical field

The present invention relates to field of cloud calculation, more particularly to a kind of based on performance requirement and sequence cloud monitoring system and structure Method.

Background technology

Cloud computing is another great change after the 1980's mainframe computer to the big change of client-server, It is Distributed Calculation (Distributed Computing), parallel computation (Parallel Computing), effectiveness calculating (Utility Computing), the network storage (Network Storage Technologies), virtualization (Virtualization), the traditional calculations such as load balancing (Load Balance), hot-standby redundancy (High Available) Machine and the product of network technical development fusion.

Cloud computing platform provides a user with network access available, easily, on demand.User enters configurable calculating Resource-sharing pond (resource include network, server, storage, application software, service), can put into little management work with And in the case of seldom being interacted with service provision end, the above-mentioned resource of quick obtaining.

Existing cloud service platform is mostly in the form of multiple virtual machines are created in a physical machine, user's application is virtual Machine is the size for needing to set virtual machine, and by cloud service provider, according to user's request, physically to create this virtual corresponding Machine.Using this kind of mode, one carrys out the specific performance that user does not know physical machine, to the performance neither one of virtual machine handle well Control；Two carry out the specific mission requirements that cloud service provider does not know user yet, are that the physical machine of its distribution may in performance The mission requirements of user cannot well be met；Furthermore, using this kind of mode, the limited performance of virtual machine is in separate unit physical machine sheet Body, certain mission requirements of user if there is explosion type growth, such as when there are wide-area failures in monitored server, calamity The mission requirements of evil transfer service occur increase drastically, it is more likely that cause virtual machine overload operation and cause the machine of delaying, It could even be possible to because prolonged overload operation causes the impaired of physical machine performance.

The content of the invention

It is an object of the invention to overcome prior art not enough, there is provided a kind of based on performance requirement and sequence cloud monitoring system System and Imaginary Mechanism construction method, realize virtual machine and sequenceization process, allow virtual machine while building in many physical machines, The concurrent operation of task is realized, and the elasticity adjustment of resource can be carried out.

The present invention uses following technical scheme to achieve the above object：

In a first aspect, the invention provides a kind of based on performance requirement and sequence cloud monitoring system, including：Calculate collection Group, monitoring module and simultaneously sequence module；

Wherein, the monitoring module and described and sequence module are connected with the computing cluster, and the monitoring module is also It is connected with monitored device by network, the computing cluster includes at least one physical node；

Described and sequence module is used to obtaining the number of the physical node, the hardware kind number of each physical node and The hardware parameter of each hardware, and physics node hardware parameter matrix is built according to acquired hardware parameter, and according to institute State physical node hardware parameter matrix and default hardware parameter grade form builds physics node hardware rating matrix；

Described and sequence module is additionally operable to be built according to the hardware rating matrix and default hardware parameter weight matrix Physical node weight rating matrix；

Described and sequence module is additionally operable to according to the physical node weight rating matrix, default virtual machine initial parameter Setting and default virtual machine build instruction and at least one virtual machine, described and sequence are built at least one physical node Module is additionally operable to distribute monitor task to the virtual machine.

In an embodiment of the present invention, the monitoring module also include data acquisition module, data positive sequence module and temporarily Storing module；

The data acquisition module is connected by network with monitored device, the data acquisition module also with the data Positive sequence module is connected, and the data positive sequence module is connected with the temporary storage module, and the temporary storage module calculates collection with described Faciation connects.

In an embodiment of the present invention, the data acquisition module is used to be received by network the monitoring number of monitored device According to, and the monitoring data that will be received is sent in the data positive sequence module.

In an embodiment of the present invention, the data positive sequence module is used to adjust the lattice of the monitoring data for receiving Formula, and by adjustment after the monitoring data be sent in temporary storage module；The temporary storage module is used to store the monitoring data.

In an embodiment of the present invention, the computing cluster is used to provide hardware resource, the calculating for the virtual machine Cluster is additionally operable to read the monitoring data of storage in the temporary storage module and is sent to the corresponding virtual machine.

In an embodiment of the present invention, described and sequence module includes hardware parameter acquisition module, physical node scoring mould Block and structure module；

Wherein, the hardware parameter acquisition module is connected with the computing cluster；

The number of the hardware parameter acquisition module acquisition physical node, the hardware kind number of each physical node, And the hardware parameter of each hardware, and physics node hardware parameter matrix is built according to acquired hardware parameter, it is described hard Part parameter acquisition module is additionally operable to according to the physical node hardware parameter matrix and default hardware parameter grade form construction Reason node hardware rating matrix；The hardware parameter acquisition module is additionally operable to be sent to the physical node hardware rating matrix The physical node grading module；

The physical node grading module is used for according to the physical node hardware rating matrix and default hardware for receiving Parameters weighting matrix builds physics node weights rating matrix；The physical node grading module is additionally operable to the physical node Weight rating matrix is sent to the structure module；

The structure module is additionally operable to be set according to the physical node weight rating matrix, default virtual machine initial parameter Fixed and default virtual machine builds instruction and at least one virtual machine is built at least one physical node.

Specifically, in an embodiment of the present invention, remembering that the hardware parameter obtains the physical node hardware parameter square for building Battle array is P,

Wherein, n is the number of physical node, and m is the hardware parameter species that each physical node is included, P_ijRepresent i-th The parameter of physical node jth kind hardware, i ∈ [1, n], j ∈ [1, m].

Specifically, in an embodiment of the present invention, remembering that the hardware parameter obtains the physical node hardware scoring square for building Battle array is Ps,

Wherein, n is the number of physical node, and m is the hardware parameter species that each physical node is included, Ps_ijRepresent i-th The scoring of physical node jth kind hardware, i ∈ [1, n], j ∈ [1, m].

Further, the hardware parameter acquisition module is additionally operable to according to the physical node hardware parameter matrix and default Hardware parameter grade form builds physics node hardware rating matrix, specifically includes：

According to each hardware parameter P in the default hardware parameter grade form acquisition hardware parameter matrix P_ijCorresponding scoring Ps_ij；

The hardware scoring Ps that will be obtained_ijIt is built into hardware rating matrix Ps.

Specifically, in an embodiment of the present invention, remember that the hardware parameter weight matrix is U,

And

Wherein, m is the hardware parameter species that each physical node is included, U_jRepresent the weight of jth kind hardware machine parameter, j ∈[1,m]。

Further, in an embodiment of the present invention, the physical node weight that the physical node grading module builds is remembered Rating matrix is Pvs,

Wherein, Pvs_iRepresent i-th weight scoring of physical node, i ∈ [1, n].

In an embodiment of the present invention, the default virtual machine builds instruction includes that the first virtual machine builds instruction；Institute State the quantity of virtual machine Initial parameter sets including default virtual machine, the hardware kind number of default each virtual machine and pre- If each described hardware hardware parameter；

Wherein, described and sequence module is additionally operable to build the instruction acquisition physical node according to default first virtual machine Weight scoring highest physical node in weight rating matrix；

Described and sequence module is additionally operable to quantity, the hardware of default each virtual machine according to default virtual machine The hardware parameter for planting number and default each hardware builds virtual machine initial parameter matrix；

Described and sequence module is additionally operable to calculate all default virtual respectively according to the virtual machine initial parameter matrix The initial hardware parameter sum of each hardware of machine, and the initial hardware parameter sum of each hardware of gained is scored most with the weight The hardware parameter of physical node correspondence hardware high is compared；

When all default virtual machines each hardware initial hardware parameter sum no more than the weight score highest Physical node correspondence hardware hardware parameter when, described and sequence module is additionally operable to according to the virtual machine initial parameter matrix All virtual machines are built on weight scoring highest physical node.

Further, in an embodiment of the present invention, the structure module also includes that physical node chooses module, hardware money Source computing module and Imaginary Mechanism modeling block；

The physical node chooses module to be used to build the instruction acquisition physical node according to default first virtual machine Weight scoring highest physical node in weight rating matrix；

The hardware resource computing module is used to calculate all default respectively according to the virtual machine initial parameter matrix The initial hardware parameter sum of each hardware of virtual machine, and the initial hardware parameter sum of each hardware of gained and the weight are commented The hardware parameter of point highest physical node correspondence hardware is compared；

When all default virtual machines each hardware initial hardware parameter sum no more than the weight score highest Physical node correspondence hardware hardware parameter when, the hardware resource computing module to the Imaginary Mechanism modeling block send structure Build instruction；

Imaginary Mechanism modeling block be used for quantity according to default virtual machine, default each virtual machine it is hard The hardware parameter of part kind number and default each hardware builds virtual machine initial parameter matrix；

The Imaginary Mechanism modeling block is additionally operable to according to the structure instruction for receiving and the virtual machine initial parameter Matrix builds all virtual machines on physical node described in weight scoring highest, and the Imaginary Mechanism modeling block is additionally operable to Monitor task is distributed to the virtual machine.

Specifically, in being example in the present invention one, remembering that the physical node chooses the weight scoring highest physics for obtaining Node is h₁, h₁∈[1,n]；

Remember that the default virtual machine quantity is k；The hardware kind number for remembering each virtual machine is m；Remember that q-th virtual machine is pre- If jth kind hardware parameter be VM_qj, q ∈ [1, k], j ∈ [1, m]；

The virtual machine initial parameter matrix that the Imaginary Mechanism builds module construction is VM,

Wherein, VM_qjRepresent q-th parameter of virtual machine jth kind hardware, q ∈ [1, k], j ∈ [1, m]；

The default jth kind hardware initial hardware parameter sum of k virtual machine that the hardware resource computing module is calculated is

Remember h₁Jth kind hardware parameter is on individual physical node

WhenIt is satisfied byWhen,

The hardware resource computing module sends to Imaginary Mechanism modeling block and builds instruction；

The Imaginary Mechanism modeling root tuber is according to the structure instruction for receiving and the virtual machine initial parameter matrix V M In weight scoring highest physical node h₁K virtual machine of upper structure, the Imaginary Mechanism modeling block is additionally operable to described virtual Machine distributes monitor task.

Further, in an embodiment of the present invention, the structure module also includes that physical node loads computing module；

Imaginary Mechanism modeling block is additionally operable to for the virtual machine initial parameter matrix to be sent to the physical node to be born Carry computing module；

When the default one of which initial hardware parameter sum of all described virtual machine for needing to build is more than the weight On scoring highest physical node during corresponding hardware parameter, the hardware resource computing module is loaded to the physical node and counted Calculate module and send control instruction；

The physical node load computing module is used for according to the control instruction and virtual machine initial parameter for receiving The maximum virtual machine quantity that can be set up on weight scoring highest physical node described in matrix computations, and will be described maximum virtual Machine quantity is sent to the Imaginary Mechanism modeling block；

The Imaginary Mechanism modeling block is used for initial according to the described maximum virtual machine quantity and the virtual machine for receiving Parameter matrix VM builds the virtual machine of respective numbers on physical node described in weight scoring highest.

Specifically, in being example in the present invention one, whenWhen；

The hardware resource computing module sends control instruction to physical node load computing module；

Remember that the physical node load computing module calculates physical node h₁On the virtual machine number that can at most build be L₁, L₁∈[1,k)；

Wherein, L is remembered₁The individual default jth kind hardware initial hardware parameter sum of virtual machine is

L₁Meet

The described maximum virtual machine quantity L that the physical node load computing module will be calculated₁It is sent to the virtual machine Build module.

The Imaginary Mechanism models root tuber according to the virtual machine initial parameter matrix V M by the 1st to L₁Individual virtual frame It is located at physical node h₁On, the Imaginary Mechanism modeling block is additionally operable to distribute monitor task to the virtual machine.

Further, in an embodiment of the present invention, the structure module also includes loop module；

The Imaginary Mechanism modeling root tuber is according to the described maximum virtual machine quantity for receiving and the initial ginseng of the virtual machine Number is set in after building the virtual machine of respective numbers on physical node described in the weight scoring highest,

The Imaginary Mechanism modeling is additionally operable to be sent to the loop module virtual machine quantity and the counting instruction for having built；

The loop module is used for loop initialization number of times, and the initial value of the cycle-index is zero；

Circulation starts：

The loop module is used to count instruction record cycle-index according to described, makes the cycle-index plus one；

The loop module is additionally operable to quantity and the virtual machine quantity for having built according to the default virtual machine Calculating also needs to the virtual machine quantity for building；

The loop module is additionally operable to choose module transmission selection instruction again to the physical node, and loop module is also used In to the virtual machine quantity for also needing to build described in hardware resource computing module transmission；

The physical node chooses module and is additionally operable to choose instruction again according to, in physical node weight scoring In matrix in the physical node of unselected mistake weight selection scoring highest physical node, and by the weight score highest The numbering of physical node is sent to the hardware resource computing module, the physical node load computing module and the virtual machine Build module；

The hardware resource computing module is additionally operable to, according to the virtual machine quantity for also needing to build for receiving, calculate It is described also need to build the default various initial hardware parameter sums of virtual machine, and with the weight score highest physics section Corresponding hardware parameter is compared on point；

The Imaginary Mechanism modeling block is used for initial according to the described maximum virtual machine quantity and the virtual machine for receiving Parameter matrix VM builds the virtual machine of respective numbers on physical node described in weight scoring highest；

The virtual machine quantity that the Imaginary Mechanism modeling block will build counts instruction and is sent to the loop module；

From circulation beginning circulation above-mentioned steps；

When the initial hardware parameter sum of each hardware of the virtual machine for also needing to and building is commented no more than the weight When dividing the hardware parameter of highest physical node correspondence hardware, the hardware resource computing module models block to the Imaginary Mechanism Send and build instruction；

The Imaginary Mechanism modeling block is additionally operable to according to structure instruction, the virtual machine initial parameter square for receiving Battle array and the virtual machine quantity for also needing to build also are needed described in being built on physical node described in weight scoring highest The virtual machine of structure；The Imaginary Mechanism modeling block is additionally operable to distribute monitor task to the virtual machine；

The Imaginary Mechanism modeling block sends circulation halt instruction to the loop module；

The loop module stops circulation according to the circulation halt instruction.

Specifically, in an embodiment of the present invention, note cycle-index is r, and the initial value of r is 0, r ∈ [0, n-1]；

WhenWhen, the Imaginary Mechanism models root tuber according to the virtual machine initial parameter square VM is by the 1st to L for battle array₁Individual virtual machine is erected at physical node h₁On, Imaginary Mechanism modeling block is additionally operable to have built Virtual machine quantity L₁And counting instruction is sent to the loop module；

Circulation starts：

The loop module counts instruction calculating cycle-index r according to described, makes r=r+1；

The loop module is according to the quantity according to the default virtual machine and the virtual machine number for having built for receiving Amount L₁Calculating also needs to the virtual machine quantity f for building,

The loop module is additionally operable to choose module transmission selection instruction, the loop module again to the physical node It is additionally operable to the virtual machine quantity f for also needing to build described in hardware resource computing module transmission；

The physical node chooses module and chooses instruction again according in the physical node weight rating matrix Pvs In unselected mistake physical node in weight selection scoring highest physical node h again_r+1, remember the physical node h_r+1's Score and be And h_r+1≠{h_r,h_r-1……h₁}；

Remember that the hardware resource computing module calculates the f default jth kind hardware initial hardware parameter sum of virtual machine ForRemember h_r+1Jth kind hardware parameter is on individual physical node

WhenWhen,

Remember that the physical node load computing module calculates physical node h_r+1On the virtual machine number that can at most build be L_r+1, L_r+1∈[1,k)；

Wherein, L is remembered_r+1The individual default jth kind hardware initial hardware parameter sum of virtual machine is

L_r+1Meet

The described maximum virtual machine quantity L that the physical node load computing module will be calculated_r+1It is sent to the virtual machine Build module；

Imaginary Mechanism modeling root tuber is according to the virtual machine initial parameter matrix V M by (L_r+ 1) it is individual to (L_r+L_r+1) Individual virtual machine is erected at physical node h_r+1On；

The virtual machine quantity L that the Imaginary Mechanism modeling block will build_r+1And counting instruction is sent to the cyclic module Block；

From circulation beginning circulation above-mentioned steps；

WhenIt is satisfied byWhen, the hardware resource computing module is to described virtual Mechanism Modeling block sends and builds instruction；

The Imaginary Mechanism modeling root tuber is according to the structure instruction for receiving and the virtual machine initial parameter matrix V M By (L_r+ 1) individual to k-th virtual machine is erected at physical node h_r+1On；The Imaginary Mechanism modeling block is additionally operable to the void Plan machine distributes monitor task；

In an embodiment of the present invention, the default virtual machine builds instruction includes that the second virtual machine builds instruction；Institute State the quantity of virtual machine Initial parameter sets including default virtual machine, the hardware kind number of default each virtual machine and pre- If each described hardware hardware parameter；

Wherein, described and sequence module is additionally operable to quantity, default each virtual machine according to default virtual machine Hardware kind number and default each hardware hardware parameter build virtual machine initial parameter matrix；

Described and sequence module is additionally operable to build instruction by physical node weight scoring according to second virtual machine The weight scoring of each physical node is compared with default value in matrix, and physical node mark is built according to the comparative result Quasi- rating matrix；

Described and sequence module is additionally operable to the structure according to each physical node of physical node scale matrix computations Weight；

Described and sequence module is additionally operable to structure weight and the virtual machine initial parameter according to each physical node The hardware parameter that each virtual machine of matrix computations occupies on each physical node, described and sequence module is additionally operable to according to obtained by calculating Virtual machine is built on corresponding physical node, described and sequence module is additionally operable to distribute monitor task to the virtual machine.

Further, in an embodiment of the present invention, the structure module also includes physical node standards of grading module；

The physical node standards of grading module is used to build instruction by the physics section according to second virtual machine The weight scoring of each physical node is compared with default value in point weight rating matrix, is built according to the comparative result Physical node scale matrix；

Wherein, the standardization is specially the weight scoring when the physical node not less than the default value When, the scale of the physical node is equal to the weight scoring of the physical node；Otherwise, the physical node Scale is zero.

Specifically, in an embodiment of the present invention, remembering that the default value is virtual machine standard point Pst, i-th physics is remembered The weight scoring of node is Pvs_i, i ∈ [1, n]；

The physical node scale matrix of the physical node standards of grading module construction is Pvss,

Wherein, Pvss_iRepresent i-th scale of physical node, i ∈ [1, n]；

Further, in an embodiment of the present invention, the structure module also includes building weight computation module；

The structure weight computation module is used for according to each physical node of physical node scale matrix computations Build weight.

Specifically, in an embodiment of the present invention, the structure for building i-th physical node that weight computation module is calculated Weight is built for VMW_i,

Further, in an embodiment of the present invention, the Imaginary Mechanism modeling block is used for according to described each physics section The structure weight and the virtual machine initial parameter matrix of point, calculate the hardware ginseng that each virtual machine is occupied on each physical node Number；

The Imaginary Mechanism modeling block is additionally operable to the hardware ginseng occupied on each physical node according to each virtual machine Number builds corresponding virtual machine on corresponding physical node.

Specifically, in an embodiment of the present invention, remembering that the default virtual machine quantity is k；Remember each virtual machine Hardware kind number is m；Remember that q-th default jth kind hardware parameter of virtual machine is VM_qj, q ∈ [1, k], j ∈ [1, m]；

The hardware that is occupied ginseng of y-th virtual machine that the Imaginary Mechanism modeling block is calculated on s-th physical node Number is [VM_y1…VM_ym]×VMW_s, rounding up, y ∈ [1, k], s ∈ [1, n], the Imaginary Mechanism modeling block are additionally operable to institute State virtual machine distribution monitor task.

Preferably, in an embodiment of the present invention, the Imaginary Mechanism models root tuber according to y-th virtual machine obtained by calculating The hardware parameter for being occupied on s-th physical node builds y-th virtual machine on s-th physical node, described virtual Mechanism Modeling block is additionally operable to distribute monitor task to the virtual machine.

In an embodiment of the present invention, based on performance requirement and sequence cloud monitoring system also include human-computer interaction module；

The human-computer interaction module is used to providing default parameter and the virtual machine and builds instruction input interface, and by institute State default parameter and the virtual machine builds instruction and is sent to described and sequence module, described and sequence module is additionally operable to receive The default parameter and the virtual machine build instruction.

In an embodiment of the present invention, described based on performance requirement and sequence cloud monitoring system also includes external storage mould Block, the computing cluster is connected by network with the outer memory module, the outer memory module by store it is described based on Calculate the data of cluster.

In an embodiment of the present invention, the physical node is computer, and the hardware of the physical node includes but do not limit In CPU, internal memory, hard disk, mainboard.

Second aspect, present invention also offers a kind of based on performance requirement and sequence cloud monitoring system construction method, Including:

Obtain physical node number, the hardware kind number of each physical node, and each hardware hardware parameter, And physics node hardware parameter matrix is built according to acquired hardware parameter, and according to the physical node hardware parameter matrix And default hardware parameter grade form builds physics node hardware rating matrix；

Physics node weights rating matrix is built according to the hardware rating matrix and default hardware parameter weight matrix；

According to the physical node weight rating matrix, default virtual machine Initial parameter sets and default virtual machine Build instruction and at least one virtual machine is built at least one physical node, and monitor task is distributed to the virtual machine.

In an embodiment of the present invention, remember that constructed physical node hardware parameter matrix is P,

In an embodiment of the present invention, remember that constructed physical node hardware rating matrix is Ps,

Further, the structure physics node hardware rating matrix Ps is specifically included：

And

Wherein, Pvs_iRepresent i-th weight scoring of physical node, i ∈ [1, n].

In an embodiment of the present invention, the default virtual machine builds instruction includes that the first virtual machine builds instruction；

The virtual machine Initial parameter sets include quantity, the hardware of default each virtual machine of default virtual machine Plant the hardware parameter of number and default each hardware；

It is described according to the physical node weight rating matrix, default virtual machine Initial parameter sets and default Virtual machine builds instruction and at least one virtual machine is built at least one physical node, specifically includes：

Weight during instruction obtains the physical node weight rating matrix is built according to default first virtual machine to score most Physical node high；

Quantity, the hardware kind number of default each virtual machine according to default virtual machine and it is default it is each described in The hardware parameter of hardware builds virtual machine initial parameter matrix；

Calculate the initial hardware of each hardware of all default virtual machines respectively according to the virtual machine initial parameter matrix Parameter sum, and by the initial hardware parameter sum of each hardware of gained and the weight scoring corresponding hardware of highest physical node Hardware parameter be compared；

When all default virtual machines each hardware initial hardware parameter sum no more than the weight score highest Physical node correspondence hardware hardware parameter when, according to the virtual machine initial parameter matrix the weight score highest All virtual machines are built on physical node, and monitor task is distributed to the virtual machine.

Specifically, in an embodiment of the present invention, remember the number of the physical node for n, each physical node it is hard Part kind number m；

Remember that the weight scoring highest physical node is h₁, h₁∈[1,n]；

The constructed virtual machine initial parameter matrix of note is VM,

Remember that the k default jth kind hardware initial hardware parameter sum of virtual machine is

Remember h₁Jth kind hardware parameter is on individual physical node

WhenIt is satisfied byWhen,

According to the virtual machine initial parameter matrix V M in weight scoring highest physical node h₁Upper structure k virtual Machine, and distribute monitor task to the virtual machine.

In an embodiment of the present invention, when the default one of which initial hardware ginseng of all described virtual machine for needing to build When number sum is more than corresponding hardware parameter on weight scoring highest physical node；

Can be set up on the weight scoring highest physical node according to the virtual machine initial parameter matrix computations Maximum virtual machine quantity；

Respective counts are built on physical node according to the virtual machine initial parameter matrix and the maximum virtual machine quantity The virtual machine of amount, and distribute monitor task to the virtual machine.

Specifically, in being example in the present invention one, whenWhen,

Note L₁The individual default jth kind hardware initial hardware parameter sum of virtual machine is

Wherein, L₁Meet

According to the virtual machine initial parameter matrix V M by the 1st to L₁Individual virtual machine is erected at physical node h₁On, and Monitor task is distributed to the virtual machine.

In an embodiment of the present invention, when the default one of which initial hardware ginseng of all described virtual machine for needing to build It is described initial according to the virtual machine when number sum is more than corresponding hardware parameter on weight scoring highest physical node After the virtual machine that parameter matrix and the maximum virtual machine quantity build respective numbers on physical node, also comprise the following steps Loop initialization number of times, the initial value of the cycle-index is zero；

Circulation starts：

Cycle-index adds one；

Quantity according to the default virtual machine and the virtual machine quantity for having built calculate the virtual machine number for also needing to build Amount；

Weight selection scoring is most again in the physical node of unselected mistake in the physical node weight rating matrix Physical node high；

The initial hardware parameter of each hardware of the virtual machine for building also is needed according to the virtual machine initial parameter matrix computations Sum, and by the hard of the initial hardware parameter sum of each hardware of gained and the weight scoring corresponding hardware of highest physical node Part parameter is compared；

When the default one of which initial hardware parameter sum of virtual machine for also needing to build is commented more than the weight When dividing corresponding hardware parameter on highest physical node；

Obtain the virtual machine quantity that can be at most built on the weight scoring highest physical node；

Scored in the weight according to the virtual machine initial parameter matrix and the virtual machine quantity that can at most build The virtual machine of respective numbers is built on highest physical node；

Repeated the above steps from circulation beginning；

When the default initial hardware parameter sum of virtual machine for also needing to build scores most no more than the weight On physical node high during corresponding hardware parameter；

According to the virtual machine initial parameter matrix the weight scoring highest physical node on build it is all described in The virtual machine for building also is needed, and monitor task is distributed to the virtual machine；

End loop.

Specifically, in being example in the present invention one, remembering that the cycle-index is r, the initial value of r is 0, r ∈ [0, n-1]；

Circulation starts：

R=r+1；

Note also needs to the virtual machine quantity f for building,

The weight selection scoring again in the physical node of unselected mistake in the physical node weight rating matrix Pvs Highest physical node h_r+1, remember the physical node h_r+1Scoring beh_r+1∈ [1, n], and h_r+1≠{h_r, h_r-1……h₁}；

Remember that the f default jth kind hardware initial hardware parameter sum of virtual machine isRemember h_r+1Individual physics Jth kind hardware parameter is on node

WhenWhen,

Note physical node h_r+1On the virtual machine number that can at most build be L_r+1, L_r+1∈[1,k)；

L_r+1Meet

According to the virtual machine initial parameter matrix V M by (L_r+ 1) it is individual to (L_r+L_r+1) individual virtual machine is erected at physics Node h_r+1On；

Repeated the above steps from circulation beginning；

WhenIt is satisfied byWhen,

According to the virtual machine initial parameter matrix V M by (L_r+ 1) individual to k-th virtual machine is erected at physical node h_r+1On, and distribute monitor task to the virtual machine；

Stop circulation.

Instruction is built by each physical node in the physical node weight rating matrix according to second virtual machine Weight scoring is compared with default value, and physical node scale matrix is built according to the comparative result；

According to the structure weight of each physical node of physical node scale matrix computations；

Structure weight and each virtual machine of virtual machine initial parameter matrix computations according to each physical node is each The hardware parameter occupied on physical node, virtual machine is built according to gained is calculated on corresponding physical node, and to the void Plan machine distributes monitor task.

Specifically, in being example in the present invention one, remembering that the default virtual machine quantity is k；Remember each virtual machine Hardware kind number is m；Remember that q-th default jth kind hardware parameter of virtual machine is VM_qj, q ∈ [1, k], j ∈ [1, m]；

The constructed virtual machine initial parameter matrix of note is VM,

The number for remembering the physical node is n, the hardware kind number m of each physical node；Remember that the default value is void Plan machine standard scores Pst, note physical node weight rating matrix is Pvs, and the weight scoring of i-th physical node of note is Pvs_i, i ∈ [1,n]；

The physical node scale matrix is remembered for Pvss,

Wherein, Pvss_iRepresent i-th scale of physical node, i ∈ [1, n]；

Remember that i-th structure weight of physical node is VMW_i,

The hardware parameter that is occupied of y-th virtual machine on s-th physical node is [VM_y1…VM_ym]×VMW_s, upwards Round, y ∈ [1, k], s ∈ [1, n].

Preferably, in an embodiment of the present invention, the Imaginary Mechanism models root tuber according to y-th virtual machine obtained by calculating The hardware parameter for being occupied on s-th physical node builds y-th virtual machine on s-th physical node.

Beneficial effects of the present invention：

Cloud monitoring system provided by the present invention, its using and sequence by the way of build virtual machine, can be according to user's need Seek the building mode of adjustment virtual machine so that the performance of virtual machine better meets the mission requirements of user, relative to traditional For building mode, the extendible ability of resource of virtual machine provided by the present invention is stronger, and each virtual machine can be enjoyed entirely The resource of resource pool, allow users to more flexibly, efficiently arrange virtual machine.

Brief description of the drawings

Fig. 1 is a kind of based on performance requirement and sequence cloud monitoring system the structural representation in one embodiment of the invention Figure；

Fig. 2 is the structural representation of in one embodiment of the invention and sequence module；

Fig. 3 is the structural representation of the structure module in one embodiment of the invention；

Fig. 4 is a kind of based on performance requirement and sequence cloud monitoring system the construction method stream in one embodiment of the invention Cheng Tu；

Fig. 5 is the construction method flow chart of the kind virtual machine in first embodiment of the invention；

Fig. 6 is the construction method flow chart of the kind virtual machine in one embodiment of the invention；

Fig. 7 is the construction method flow chart of the kind virtual machine in another embodiment of the present invention；

Fig. 8 is the construction method flow chart of the kind virtual machine in second embodiment of the invention；

Fig. 9 is a kind of method flow diagram of the standardization in one embodiment of the invention.

Specific embodiment

Below in conjunction with the accompanying drawings and specific embodiment the present invention will be further described, illustrative examples therein and Illustrate only to be used for explaining the present invention, but it is not as a limitation of the invention.

In a first aspect, the invention provides a kind of based on performance requirement and sequence cloud monitoring system, such as Fig. 1, the bag Include monitoring module, and sequence module 400, computing cluster 500 and human-computer interaction module 600；

Wherein, the monitoring module includes, data acquisition module 100, data positive sequence module 200, temporary storage module 300,

Data acquisition module 100 is connected by LAN or internet with monitored device, data acquisition module 100 also with Data positive sequence module 200 and, data positive sequence module 200 is connected with temporary storage module 300, human-computer interaction module 600 with it is described And sequence module 400 is connected, temporary storage module 300 and simultaneously sequence module 400 is connected with computing cluster 500, computing cluster 500 includes At least one physical node；

Data acquisition module 100 is used to obtain the monitoring data of monitored device, and the monitoring data is sent into institute State data positive sequence module 200；

Data positive sequence module 200 is used to adjusting the form of the monitoring data for receiving, and by described in after adjustment Monitoring data is sent in temporary storage module 300；

Temporary storage module 300 is used to store the monitoring data；

Human-computer interaction module 300 is used to providing default parameter and the virtual machine and builds instruction input interface, and by institute State default parameter and the virtual machine builds instruction and is sent to and sequence module 400, and sequence module 400 is additionally operable to receive institute State default parameter and the virtual machine builds instruction；

And sequence module 400 is used to obtain default parameter, and sequence module 400 is additionally operable to obtain the physical node The hardware parameter of number n, the hardware class m of the physical node and each physical node, and sequence module 400 is additionally operable to Instruction is built at least one institute according to the default parameter, the hardware parameter of each physical node and the virtual machine State and built on physical node at least one virtual machine, and sequence module 400 is additionally operable to distribute monitor task to the virtual machine；

Computing cluster 500 is used to provide hardware resource for the virtual machine, and computing cluster 500 is additionally operable to read temporary storage module In 300 store the monitoring data and be sent to the corresponding virtual machine.

In an embodiment of the present invention, as shown in Fig. 2 simultaneously sequence module 400 includes hardware parameter acquisition module 410, thing Reason node grading module 420 and structure module 430；

Wherein, hardware parameter acquisition module 410 is connected with computing cluster 200；

Hardware parameter acquisition module 410 is used to obtain the quantity n of the physical node and hardware class m of each physical node, Hardware parameter acquisition module 410 is additionally operable to obtain the various hardware parameters of each physical node, and according to default hardware Parameter scores table obtains the corresponding hardware parameter scoring of each described hardware parameter, and by the described hard of each physical node Part parameter scores are sent to physical node grading module 420；

Physical node grading module 420 is used to be scored and pre- according to the hardware parameter of each physical node for receiving If hardware parameter weight matrix calculate the weight scoring of each physical node, and the weight of each physical node is scored It is sent to structure module 430；

Build weight scoring, the initial parameter of the virtual machine that module 430 is additionally operable to each physical node according to Setting and the virtual machine build instruction and Imaginary Mechanism are built on physical node described at least one.

Specifically, hardware parameter acquisition module 410 is according to the quantity n of the physical node for getting, the hardware of physical node The various hardware parameters of species m and each physical node build physics node hardware parameter matrix P：

Wherein, P_ijRepresent i-th parameter of physical node jth kind hardware, i ∈ [1, n], j ∈ [1, m]；

Hardware parameter acquisition module 410 is according to each in the default hardware parameter grade form acquisition hardware parameter matrix P Hardware parameter P_ijCorresponding scoring Ps_ij, and it is built into hardware rating matrix Ps：

Wherein, Ps_ijRepresent i-th scoring of physical node jth kind hardware, i ∈ [1, n], j ∈ [1, m]；

The hardware rating matrix Ps is sent to physical node grading module 420 by hardware parameter acquisition module 410.

Specifically, the default hardware parameter weight matrix,

And

Wherein, U_jRepresent the weight of jth kind hardware parameter, j ∈ [1, m]；

Specifically, physical node grading module 420 is joined according to the hardware rating matrix Ps for receiving and the hardware Number weight matrix U, calculates the weight scoring of each physical node, obtains physical node weight rating matrix Pvs,

Wherein,

Wherein, Pvs_iRepresent i-th weight scoring of physical node, i ∈ [1, n]；

The physical node weight rating matrix Pvs is sent to structure module 430 by physical node grading module 420.

In an embodiment of the present invention, as shown in figure 3, building module 430 also includes that Imaginary Mechanism models block 431, physics Node selection module 432, hardware resource computing module 433, physical node load computing module 434, loop module 435, physics Node standards of grading module 436, builds weight computation module 437；

Wherein, Imaginary Mechanism modeling block 431 is used for quantity, default each virtual machine according to default virtual machine Hardware kind number and default each hardware hardware parameter build virtual machine initial parameter matrix；

Physical node chooses module 432 to be used to obtain weight scoring highest thing in the physical node weight rating matrix Node is managed, and weight scoring highest physical node numbering is sent to hardware resource computing module 433, physical node and born Carry computing module 434 and Imaginary Mechanism modeling block 431；

Hardware resource computing module 433 is used to be needed according to the virtual machine initial parameter matrix computations all institutes of structure State the default various initial hardware parameter sums of virtual machine, and with corresponding hardware on weight scoring highest physical node Parameter is compared；

When the default various initial hardware parameter sums of all described virtual machine for needing to build are no more than the power Again on scoring highest physical node during corresponding hardware parameter, hardware resource computing module 433 is additionally operable to be built to virtual machine Module 431 sends and builds instruction；

Imaginary Mechanism modeling block 431 is additionally operable to according to the structure instruction for receiving and the virtual machine initial parameter square Battle array building all virtual machines described in weight scoring highest on physical node, Imaginary Mechanism modeling block 431 be additionally operable to The virtual machine distributes monitor task.

When the default one of which initial hardware parameter sum of all described virtual machine for needing to build is more than the weight On scoring highest physical node during corresponding hardware parameter, hardware resource computing module 433 calculates mould to physical node load Block 434 sends control instruction；

Physical node load computing module 434 is used for according to the control instruction and the virtual machine initial parameter square for receiving Battle array calculates the maximum virtual machine quantity that can be set up on weight scoring highest physical node, and by the maximum virtual machine Quantity is sent to Imaginary Mechanism modeling block 431；

Loop module 435 is used for loop initialization number of times, and loop module 435 is additionally operable to record cycle-index, loop module 435 are additionally operable to calculate the virtual machine quantity for also needing to build；

Loop module 435 is additionally operable to choose module transmission selection instruction again to the physical node；

Loop module 435 is additionally operable to the virtual machine number for also needing to build described in hardware resource computing module transmission Amount；

Physical node selection module 432 is additionally operable to be chosen again according to instruction in the physical node of unselected mistake Again weight selection scoring highest physical node, and weight scoring highest physical node numbering is sent to hardware money Source computing module 433, physical node loads computing module 434 and Imaginary Mechanism modeling block 431；

Hardware resource computing module 433 is additionally operable to be also needed to described in calculating the default each of all described virtual machine of structure Initial hardware parameter sum is planted, and is compared with corresponding hardware parameter on weight scoring highest physical node；

Physical node standards of grading module 436 is used to obtain the default virtual machine standard point Pst；

Physical node standards of grading module 436 is additionally operable to according to the virtual machine standard point Pst to described in receiving The scale to each physical node that the weight scoring of each physical node is standardized, and will described each thing The scale for managing node is sent to structure weight computation module 437；

Building weight computation module 437 is used to calculate each according to the scale of each physical node for receiving The structure weight of physical node, and the structure weight of each physical node is sent to Imaginary Mechanism modeling block 431.

In an embodiment of the present invention, the virtual machine builds instruction includes that the first virtual machine builds instruction, builds mould Built in Imaginary Mechanism on physical node described in weight scoring highest by block 430, and Imaginary Mechanism modeling block 431 is additionally operable to described Virtual machine distributes monitor task.

Specifically, in an embodiment of the present invention, remembering that the default virtual machine quantity is k；Remember each virtual machine Hardware kind number is m；Imaginary Mechanism models block 431 and builds virtual machine initial parameter according to acquired virtual machine Initial parameter sets Matrix V M,

Wherein, VM_qjQ-th parameter of virtual machine jth kind hardware is represented, q ∈ [1, k], j ∈ [1, m], k are virtual machine Quantity；

And the virtual machine initial parameter matrix V M is sent to hardware resource computing module 433 and physical node load meter Calculate module 434；

Physical node is chosen module 432 and is chosen wherein according to the described hard physical node weight rating matrix Pvs for receiving Weight scoring highest physical nodeh₁∈ [1, n], and by weight scoring highest physical node numbering h₁Send To hardware resource computing module 433 and Imaginary Mechanism modeling block 431；

Hardware resource computing module 433 is used to be calculated according to the virtual machine initial parameter matrix V M for receiving needs structure The default various initial hardware parameter sums of all described virtual machine built, and scored on highest physical node with the weight Corresponding hardware parameter is compared, specifically, the k default jth kind hardware initial hardware parameter sum of virtual machine of note isRemember h₁Jth kind hardware parameter is on individual physical node

When the default various initial hardware parameter sums of all described virtual machine for needing to build are no more than the power Again on scoring highest physical node during corresponding hardware parameter, i.e. whenIt is satisfied byWhen, firmly Part Resource Calculation module 433 sends to Imaginary Mechanism modeling block 431 and builds instruction；

Imaginary Mechanism modeling block 431 is additionally operable to according to the structure instruction for receiving and the virtual machine initial parameter square Battle array VM, build k Imaginary Mechanism in weight scoring highest physical node h₁On, Imaginary Mechanism modeling block 431 is additionally operable to institute State virtual machine distribution monitor task；

When the default one of which initial hardware parameter sum of all described virtual machine for needing to build is more than the weight On scoring highest physical node during corresponding hardware parameter, i.e. whenMeetWhen, hardware resource Computing module 433 sends control instruction to physical node load computing module 434；

The physical node load computing module 434 is initially joined according to the control instruction and the virtual machine that receive Matrix number VM calculates the maximum virtual machine quantity that can be set up on the weight scoring highest physical node, and will be described Maximum virtual machine quantity is sent to Imaginary Mechanism modeling block 431；

Specifically, note physical node h₁On the virtual machine number that can at most build be L₁,L₁∈[1,k)；

The virtual machine number L that can at most build₁Meet physical node h₁Upper jth kind hardware parameterSubtract L₁Individual void The default jth kind hardware initial hardware parameter sum of plan machine is the minimum value more than 0, wherein, remember L₁The individual default jth of virtual machine Planting hardware initial hardware parameter sum is

I.e.

Imaginary Mechanism models block 431 according to the described maximum virtual machine quantity L for receiving₁And the virtual machine initial parameter Matrix V M, by the 1st to L₁Individual virtual machine is erected at physical node h₁On, Imaginary Mechanism modeling block 431 is additionally operable to the void Plan machine distributes monitor task.

Further, in an embodiment of the present invention, Imaginary Mechanism models block 431 by the 1st to L₁Individual virtual machine It is erected at physical node h₁After upper, it is additionally operable to be sent to loop module 435 the virtual machine quantity L for counting instruction and having built₁,

The initial value that loop module 435 is used for loop initialization number of times r, r is 0, and r ∈ [0, n-1]；

Circulation starts：

Loop module 435 counts instruction calculating cycle-index r according to described, makes r=r+1

Loop module 435 calculates the virtual machine quantity f for also needing to build,

Loop module 435 chooses module 432 and sends selection instruction again to physical node, and loop module 435 is provided to hardware Source computing module 433 sends the virtual machine quantity f for also needing to build；

Physical node is chosen module 432 and is selected again in the physical node of unselected mistake according to the instruction of selection again Weighting scores highest physical node againThat is h_r+1≠{h_r,h_r-1……h₁, h_r+1∈ [1, n], and the weight is commented Divide highest physical node numbering h_r+1It is sent to hardware resource computing module 433, physical node load computing module 434 And the Imaginary Mechanism modeling block 431；

Hardware resource computing module 433 be additionally operable to calculate f virtual machines default various initial hardware parameters it With, and be compared with corresponding hardware parameter on weight scoring highest physical node；

Specifically, the f default jth kind hardware initial hardware parameter sum of virtual machine of note isRemember h_r+1 Jth kind hardware parameter is on individual physical node

WhenWhen,

Hardware resource computing module 433 sends control instruction to physical node load computing module 434；

Physical node loads computing module 434 according to the control instruction for receiving and the virtual machine initial parameter square Battle array VM calculates physical node h_r+1On the maximum virtual machine quantity L that can set up_r+1, L₊₁∈[1,k)；

Specifically, the maximum virtual machine quantity L_r+1Meet physical node h_r+1Upper jth kind hardware parameterSubtract (L_r+ 1) it is individual to (L+L_r+1) the individual default jth kind hardware initial hardware parameter sum of virtual machine is minimum value more than 0, its In, remember (L_r+ 1) it is individual to (L_r+L_r+1) the individual default jth kind hardware initial hardware parameter sum of virtual machine is

I.e.

The described maximum virtual machine quantity L that physical node load computing module 434 is calculated_r+1It is sent to virtual machine structure Module 431；

Imaginary Mechanism models block 431 according to the described maximum virtual machine quantity L for receiving_r+1And the virtual machine is initially joined Matrix number VM, by (L_r+ 1) it is individual to (L_r+L_r+1) individual virtual machine is erected at physical node h_r+1On, and to loop module 435 again It is secondary to send the virtual machine quantity L for counting instruction and having built_r+1；

From circulation beginning circulation above-mentioned steps；

WhenIt is satisfied byWhen,

Hardware resource computing module 433 sends to Imaginary Mechanism modeling block 431 and builds instruction；

Imaginary Mechanism models block 431 according to the structure instruction for receiving and the virtual machine initial parameter matrix V M, By (L_r+ 1) individual to k-th virtual machine is erected at physical node h_r+1On, Imaginary Mechanism modeling block 431 is additionally operable to the void Plan machine distributes monitor task；

Imaginary Mechanism models block 431 and sends circulation halt instruction to loop module 435；

Loop module 435 stops circulation according to the circulation END instruction；

In another implementation method of the invention, the virtual machine builds instruction includes that the second virtual machine builds instruction, builds Build in Imaginary Mechanism on the physical node of the scoring higher than default virtual machine standard point by module 430.

Specifically, in an embodiment of the present invention, physical node grading module 420 scores square the physical node weight Battle array Pvs is sent to physical node standards of grading module 436；

Physical node standards of grading module 436 is according to the virtual machine standard point Pst to physical node weight scoring square Battle array Pvs is standardized, and specifically includes:

By the weight scoring Pvs of each physical node in physical node weight rating matrix Pvs_iWith the virtual machine mark Standard point Pst is compared, and obtains the scale of each physical node

According to the scale Pvss of each physical node_iPhysical node scale matrix Pvss is built,

The physical node scale matrix Pvss is sent to structure weight by physical node standards of grading module 436 Computing module 437.

Specifically, build weight computation module 437 calculates each physics according to the physical node scale matrix Pvss The structure weight VMW of node_i,

Wherein, VMW_iRepresent the structure weight of the i-th physical node；

Build weight computation module 437 and the structure weight of each physical node is sent to Imaginary Mechanism modeling block 460。

Specifically, the note default virtual machine quantity is k；The hardware kind number for remembering each virtual machine is m；Imaginary Mechanism Modeling block 431 builds virtual machine initial parameter matrix V M according to acquired virtual machine Initial parameter sets,

Imaginary Mechanism models block 431 according to the virtual machine initial parameter matrix V M and the structure weight of the physical node VMW_i, the hardware parameter that each virtual machine is occupied on each physical node is calculated, i.e.,

The hardware parameter that is occupied of y-th virtual machine on s-th physical node is [VM_y1 … VM_ym]×VMW_s, to On round, y ∈ [1, k], s ∈ [1, n]；

Imaginary Mechanism modeling block 431 builds corresponding virtual machine according to above-mentioned result of calculation on corresponding physical node, Imaginary Mechanism modeling block 431 is additionally operable to distribute monitor task to the virtual machine.

In an embodiment of the present invention, described based on load balance and sequence cloud monitoring system also includes external storage mould Block, computing cluster 200 can be also connected by network with the outer memory module, and the outer memory module is used to store described The data of computing cluster.

Second aspect, present invention also offers a kind of based on performance requirement and sequence cloud monitoring system construction method, As shown in figure 4, including:

S100：Obtain physical node number, the hardware kind number of each physical node, and each hardware hardware Parameter, and physics node hardware parameter matrix is built according to acquired hardware parameter, and joined according to the physical node hardware Matrix number and default hardware parameter grade form build physics node hardware rating matrix；

S200：The scoring of physics node weights is built according to the hardware rating matrix and default hardware parameter weight matrix Matrix；

S300：According to the physical node weight rating matrix, default virtual machine Initial parameter sets and default void Plan mechanism builds instruction and at least one virtual machine is built at least one physical node, and appoints to virtual machine distribution monitoring Business.

In an embodiment of the present invention, step S100 is completed by the system that first aspect present invention is provided, specifically by hard Part parameter acquisition module 410 is completed, hardware parameter acquisition module 410 be used to obtaining the quantity n of physical node, physical node it is hard The various hardware parameters of part species m and each physical node, and quantity n, physics section according to the physical node for getting The hardware class m of point and the various hardware parameters of each physical node build physics node hardware parameter matrix P：

Wherein, Ps_ijRepresent i-th scoring of physical node jth kind hardware, i ∈ [1, n], j ∈ [1, m].

In an embodiment of the present invention, remember that the hardware parameter weight matrix is U,

And

Wherein, U_jRepresent the weight of jth kind hardware parameter, j ∈ [1, m].

In an embodiment of the present invention, step S200 is completed by the system that first aspect present invention is provided, specifically by thing Reason node grading module 420 is completed, physical node grading module 420 according to hardware parameter acquisition module 410 sends it is described firmly Hardware parameter weight matrix U described in part rating matrix Ps, calculates the weight scoring of each physical node, obtains physical node weight Rating matrix Pvs,

Wherein,

Wherein, Pvs_iRepresent i-th weight scoring of physical node, i ∈ [1, n].

In an embodiment of the present invention, the virtual machine builds instruction includes that the first virtual machine builds instruction, the void The quantity of plan machine Initial parameter sets including default virtual machine, the hardware kind number of default each virtual machine and default The hardware parameter of each hardware；Step S300 also includes step as shown in Figure 5：

S311：Weight during instruction obtains the physical node weight rating matrix is built according to first virtual machine to score Highest physical node；

S312：Quantity, the hardware kind number of default each virtual machine according to default virtual machine and default The hardware parameter of each hardware builds virtual machine initial parameter matrix；

S313：Calculated respectively according to the virtual machine initial parameter matrix all default virtual machines each hardware it is initial Hardware parameter sum, and the initial hardware parameter sum of each hardware of gained is corresponding with weight scoring highest physical node The hardware parameter of hardware is compared；

S314：When the initial hardware parameter sum of each hardware of all default virtual machines scores no more than the weight During the hardware parameter of highest physical node correspondence hardware, scored most in the weight according to the virtual machine initial parameter matrix All virtual machines are built on physical node high, and monitor task is distributed to the virtual machine.

Specifically, in an embodiment of the present invention, step S300 is completed by the system that first aspect present invention is provided, tool Body chooses module 432 by Imaginary Mechanism modeling block 431, physical node and hardware resource computing module 433 is completed jointly；

The note virtual machine quantity for needing to build is k；Remember that q-th default jth kind initial hardware parameter of virtual machine is VM_qj, q ∈ [1, k], j ∈ [1, m]；

Imaginary Mechanism models block 431 and builds virtual machine initial parameter matrix according to acquired virtual machine Initial parameter sets VM,

And the virtual machine initial parameter matrix V M is sent to hardware resource computing module 433；

Hardware resource computing module 433 calculates all virtual machines and presets according to the virtual machine initial parameter matrix V M for receiving Various hardware parameter sums, and with the weight scoring highest physical node in corresponding hardware parameter be compared, have Body：

Remember h₁Jth kind hardware parameter is P on individual physical node_h1j,h₁∈[1,n]；

WhenIt is satisfied byWhen,

Hardware resource computing module 433 sends to Imaginary Mechanism modeling block 431 and builds instruction,

Imaginary Mechanism models block 431 according to the structure instruction for receiving and the virtual machine initial parameter matrix V M, It build k Imaginary Mechanism in weight scoring highest physical node h₁On, and distribute monitor task to the virtual machine.

In an embodiment of the present invention, as shown in fig. 6, also including after step S340：

S315：When the default one of which initial hardware parameter sum of all described virtual machine for needing to build is more than described On weight scoring highest physical node during corresponding hardware parameter；

S316：Being capable of frame on the weight scoring highest physical node according to the virtual machine initial parameter matrix computations If maximum virtual machine quantity；

S317：Phase is built on physical node according to the virtual machine initial parameter matrix and the maximum virtual machine quantity The virtual machine of quantity is answered, and monitor task is distributed to the virtual machine.

Specifically, in an embodiment of the present invention, step S300 is completed by the system that first aspect present invention is provided, tool Body chooses module 432, hardware resource computing module 433 and physical node load meter by Imaginary Mechanism modeling block 431, physical node Module 434 is calculated to complete jointly；

Remember h₁Jth kind hardware parameter is on individual physical node

WhenMeetWhen, hardware resource computing module 433 calculates mould to physical node load Block 434 sends control instruction；

I.e.

Imaginary Mechanism models block 431 according to the described maximum virtual machine quantity L for receiving₁And the virtual machine initial parameter Matrix V M, by the 1st to L₁Individual virtual machine is erected at physical node h₁On, and distribute monitor task to the virtual machine.

Further, in a preferred embodiment, shown in Fig. 7, step S370 also comprises the following steps：

S3171：Loop initialization number of times, circulation starts；

S3172：Cycle-index adds one；

S3173：Quantity according to the default virtual machine and the virtual machine quantity for having built calculate the void for also needing to build Plan machine quantity；

S3174：The weight selection again in the physical node of unselected mistake in the physical node weight rating matrix Scoring highest physical node；

S3175：Each hardware of the virtual machine for also needing to build according to the virtual machine initial parameter matrix computations it is initial hard Part parameter sum, and the initial hardware parameter sum of each hardware of gained is corresponding hard with weight scoring highest physical node The hardware parameter of part is compared；

S3176：When the default one of which initial hardware parameter sum of virtual machine for also needing to build is more than described On weight scoring highest physical node during corresponding hardware parameter；Obtain on the weight scoring highest physical node at most The virtual machine quantity that can be built；

S3177：According to the virtual machine initial parameter matrix and the virtual machine quantity that can at most build in the power The virtual machine of respective numbers is built on scoring highest physical node again；Repeated the above steps from circulation beginning；

S3178：When the default initial hardware parameter sum of virtual machine for also needing to build is no more than the weight On scoring highest physical node during corresponding hardware parameter；Scored in the weight according to the virtual machine initial parameter matrix All virtual machines for also needing to build are built on highest physical node, and monitor task is distributed to the virtual machine；

S3179：End loop.

Specifically, in an embodiment of the present invention, above-mentioned circulation, the system provided by first aspect present invention is completed, Specifically module 432, hardware resource computing module 433, physical node is chosen by Imaginary Mechanism modeling block 431, physical node to load Computing module 434 and loop module 435 are completed jointly；

Imaginary Mechanism models block 431 by the 1st to L₁Individual virtual machine is erected at physical node h₁After upper, be additionally operable to Loop module 435 sends the virtual machine quantity L for counting instruction and having built₁,

The initial value of loop module 435 loop initialization number of times r, r is 0, and r ∈ [0, n-1]；

Circulation starts：

Loop module 435 counts instruction according to described, makes r=r+1；

WhenWhen,

Physical node loads computing module 434 according to the control instruction for receiving and the virtual machine initial parameter square Battle array VM calculates physical node h_r+1On the maximum virtual machine quantity L that can set up_r+1, L_r+1∈[1,k)；

Specifically, the maximum virtual machine quantity L_r+1Meet physical node h_r+1Upper jth kind hardware parameterSubtract (L_r+ 1) it is individual to (L_r+L_r+1) the individual default jth kind hardware initial hardware parameter sum of virtual machine is minimum value more than 0, its In, remember (L_r+ 1) it is individual to (L_r+L_r+1) the individual default jth kind hardware initial hardware parameter sum of virtual machine is

I.e.

Said process is repeated from circulation beginning；

WhenIt is satisfied byWhen,

Imaginary Mechanism models block 431 according to the structure instruction for receiving and the virtual machine initial parameter matrix V M, By (L_r+ 1) individual to k-th virtual machine is erected at physical node h_r+1On, Imaginary Mechanism modeling block 431 is sent out to loop module 435 Circulation END instruction is sent, and monitor task is distributed to the virtual machine；

Loop module 435 is according to the circulation END instruction end loop.

In another implementation method of the invention, the virtual machine builds instruction also includes that the second virtual machine builds instruction, institute Stating default virtual machine and building to instruct includes that the second virtual machine builds instruction；The virtual machine Initial parameter sets include default The hardware parameter of the quantity of virtual machine, the hardware kind number of default each virtual machine and default each hardware；Step S300 also includes step as shown in Figure 8：

S321：Quantity, the hardware kind number of default each virtual machine according to default virtual machine and default The hardware parameter of each hardware builds virtual machine initial parameter matrix；

S322：Instruction is built by each physics section in the physical node weight rating matrix according to second virtual machine The weight scoring of point is compared with default value, and physical node scale matrix is built according to the comparative result；

S323：According to the structure weight of each physical node of physical node scale matrix computations；

S324：Structure weight and each virtual machine of virtual machine initial parameter matrix computations according to each physical node The hardware parameter occupied on each physical node, virtual machine is built according to gained is calculated on corresponding physical node, and to institute State virtual machine distribution monitor task.

Specifically, in an embodiment of the present invention, step S321 is completed by the system that first aspect present invention is provided, tool Body is completed by Imaginary Mechanism modeling block 431；

Wherein, VM_qjQ-th parameter of virtual machine jth kind hardware is represented, q ∈ [1, k], j ∈ [1, m], k are virtual machine Quantity.

In an embodiment of the present invention, as shown in figure 9, the default value is virtual machine standard point Pst, step S322 tool Body includes：

S3221：By the weight scoring Pvs of each physical node in the physical node weight rating matrix Pvs_iWith institute Virtual machine standard point Pst is stated to be compared；

S3222：By the weight scoring Pvs of physical node_iPhysical node standard is designated as more than the virtual machine standard point Pst Scoring Pvss_i；

S3223:Scale Pvss according to each physical node_iBuild physical node scale matrix Pvss。

In an embodiment of the present invention, step S322 is completed by the system that first aspect present invention is provided, specifically by thing Reason node standards of grading module 436 is completed；

Physical node standards of grading module 436 obtains the virtual machine standard point Pst, and according to the virtual machine standard The physical node weight rating matrix Pvs for dividing Pst to send physical node grading module 420 is standardized:

In an embodiment of the present invention, step S323 is completed by the system that first aspect present invention is provided, specifically by structure Weight computation module 437 is built to complete；

Build the physical node mark that weight computation module 437 sends according to physical node standards of grading module 436 Quasi- rating matrix Pvss calculates the structure weight VMW of each physical node_i,

Wherein, VMW_iRepresent the structure weight of the i-th physical node.

In an embodiment of the present invention, step S324 is completed by the system that first aspect present invention is provided, specifically by void Intend Mechanism Modeling block 431 to complete；

Imaginary Mechanism models block 431 and is sent out according to the virtual machine initial parameter matrix V M and structure weight computation module 437 The structure weight VMW of the physical node for sending_i, the hardware parameter that each virtual machine is occupied on each physical node is calculated, i.e.,

Imaginary Mechanism modeling block 431 builds corresponding virtual machine according to above-mentioned result of calculation on corresponding physical node, And distribute monitor task to the virtual machine.

Obviously, above-described embodiment expresses technical solution of the present invention example just for the sake of clearer, rather than right The restriction of embodiment of the present invention.To those skilled in the art, other can also be made on the basis of the above description The change or variation of multi-form, without departing from the inventive concept of the premise, these belong to protection scope of the present invention.Cause The protection domain of this patent of the present invention should be determined by the appended claims.

Claims

1. a kind of based on performance requirement and sequence cloud monitoring system, it is characterised in that including：Computing cluster, monitoring module and Sequence module；

Wherein, the monitoring module and described and sequence module are connected with the computing cluster, and the monitoring module also passes through Network is connected with monitored device, and the computing cluster includes at least one physical node；

Described and sequence module is used to obtain the number of the physical node, the hardware kind number of each physical node and each institute The hardware parameter of hardware is stated, and physics node hardware parameter matrix is built according to acquired hardware parameter, and according to the thing Reason node hardware parameter matrix and default hardware parameter grade form build physics node hardware rating matrix；

Described and sequence module is additionally operable to build physics according to the hardware rating matrix and default hardware parameter weight matrix Node weights rating matrix；

Described and sequence module is additionally operable to be set according to the physical node weight rating matrix, default virtual machine initial parameter Fixed and default virtual machine builds instruction and at least one virtual machine, described and sequence mould is built at least one physical node Block is additionally operable to distribute monitor task to the virtual machine.

2. a kind of based on performance requirement and sequence cloud monitoring system as claimed in claim 1, it is characterised in that its feature exists In the default virtual machine builds instruction includes that the first virtual machine builds instruction；The virtual machine Initial parameter sets include The hardware ginseng of the quantity of default virtual machine, the hardware kind number of default each virtual machine and default each hardware Number；

Wherein, described and sequence module is additionally operable to build the instruction acquisition physical node weight according to default first virtual machine Weight scoring highest physical node in rating matrix；

Described and sequence module is additionally operable to quantity, the hardware kind number of default each virtual machine according to default virtual machine And the hardware parameter of default each hardware builds virtual machine initial parameter matrix；

Described and sequence module is additionally operable to calculate all default virtual machines respectively according to the virtual machine initial parameter matrix The initial hardware parameter sum of each hardware, and by the initial hardware parameter sum of each hardware of gained and the weight scoring highest The hardware parameter of physical node correspondence hardware is compared；

When all default virtual machines each hardware initial hardware parameter sum no more than the weight score highest thing During the hardware parameter of reason node correspondence hardware, described and sequence module is additionally operable to according to the virtual machine initial parameter matrix in institute State and all virtual machines are built on weight scoring highest physical node, described and sequence module is additionally operable to be distributed to the virtual machine Monitor task.

3. a kind of based on performance requirement and sequence cloud monitoring system as claimed in claim 2, it is characterised in that the note thing The number for managing node is n, the hardware kind number m of each physical node；

The weight scoring highest physical node that described and sequence module is obtained is h₁, h₁∈[1,n]；

Remember that the default virtual machine quantity is k；The hardware kind number for remembering each virtual machine is m；Remember that q-th virtual machine is default Jth kind hardware parameter is VM_qj, q ∈ [1, k], j ∈ [1, m]；

Described and sequence module construction virtual machine initial parameter matrix is VM,

The default jth kind hardware initial hardware parameter sum of k virtual machine that described and sequence module is calculated is

Remember h₁Jth kind hardware parameter is on individual physical node

WhenIt is satisfied byWhen,

Described and sequence module is according to the virtual machine initial parameter matrix V M in weight scoring highest physical node h₁Upper structure K virtual machine, described and sequence module is additionally operable to distribute monitor task to the virtual machine.

4. a kind of based on performance requirement and sequence cloud service system as claimed in claim 1, it is characterised in that described default Virtual machine build instruction include the first virtual machine build instruct；The virtual machine Initial parameter sets include default virtual machine Quantity, the hardware kind number of default each virtual machine and default each hardware hardware parameter；

When the default one of which initial hardware parameter sum of all described virtual machine for needing to build scores more than the weight On highest physical node during corresponding hardware parameter；

Described and sequence module is additionally operable to the weight scoring highest physics according to the virtual machine initial parameter matrix computations The maximum virtual machine quantity that can be set up on node；

Described and sequence module is additionally operable to according to the virtual machine initial parameter matrix and the maximum virtual machine quantity in physics The virtual machine of respective numbers is built on node, described and sequence module is additionally operable to distribute monitor task to the virtual machine.

5. a kind of based on performance requirement and sequence cloud monitoring system as claimed in claim 1, it is characterised in that described default Virtual machine build instruction include the second virtual machine build instruct；The virtual machine Initial parameter sets include default virtual machine Quantity, the hardware kind number of default each virtual machine and default each hardware hardware parameter；

Wherein, described and sequence module be additionally operable to quantity according to default virtual machine, default each virtual machine it is hard The hardware parameter of part kind number and default each hardware builds virtual machine initial parameter matrix；

Described and sequence module is additionally operable to build instruction by the physical node weight rating matrix according to second virtual machine In each physical node weight scoring be compared with default value, according to the comparative result structure physical node standard comment Sub-matrix；

Described and sequence module is additionally operable to the structure weight according to each physical node of physical node scale matrix computations；

Described and sequence module is additionally operable to structure weight and the virtual machine initial parameter matrix according to each physical node The hardware parameter that each virtual machine occupies on each physical node is calculated, described and sequence module is additionally operable to according to obtained by calculating right Virtual machine is built on the physical node answered, described and sequence module is additionally operable to distribute monitor task to the virtual machine.

6. a kind of based on performance requirement and sequence cloud monitoring system construction method, including:

Obtain number, the hardware kind number of each physical node of physical node, and each hardware hardware parameter, and root Physics node hardware parameter matrix is built according to acquired hardware parameter, and according to the physical node hardware parameter matrix and in advance If hardware parameter grade form build physics node hardware rating matrix；

Built according to the physical node weight rating matrix, default virtual machine Initial parameter sets and default virtual machine Instruction builds at least one virtual machine at least one physical node, and distributes monitor task to the virtual machine.

7. a kind of based on performance requirement and sequence cloud monitoring system construction method as claimed in claim 6, its feature exists In the default virtual machine builds instruction includes that the first virtual machine builds instruction；

The virtual machine Initial parameter sets include quantity, the hardware kind number of default each virtual machine of default virtual machine And the hardware parameter of default each hardware；

It is described according to the physical node weight rating matrix, default virtual machine Initial parameter sets and default virtual Mechanism builds instruction and at least one virtual machine is built at least one physical node, specifically includes：

Built to instruct according to default first virtual machine and obtain weight scoring highest in the physical node weight rating matrix Physical node；

Quantity, the hardware kind number of default each virtual machine and default each hardware according to default virtual machine Hardware parameter build virtual machine initial parameter matrix；

Calculate the initial hardware parameter of each hardware of all default virtual machines respectively according to the virtual machine initial parameter matrix Sum, and by the hard of the initial hardware parameter sum of each hardware of gained and the weight scoring corresponding hardware of highest physical node Part parameter is compared；

When all default virtual machines each hardware initial hardware parameter sum no more than the weight score highest thing During the hardware parameter of reason node correspondence hardware, according to the virtual machine initial parameter matrix in weight scoring highest physics All virtual machines are built on node.

8. a kind of based on performance requirement and sequence cloud monitoring system construction method as claimed in claim 7, its feature exists In,

The number for remembering the physical node is n, the hardware kind number m of each physical node；

Remember that the weight scoring highest physical node is h₁, h₁∈[1,n]；

The constructed virtual machine initial parameter matrix of note is VM,

Remember h₁Jth kind hardware parameter is on individual physical node

WhenIt is satisfied byWhen,

According to the virtual machine initial parameter matrix V M in weight scoring highest physical node h₁K virtual machine of upper structure.

9. a kind of based on performance requirement and sequence cloud service system construction method as claimed in claim 6, its feature exists In the default virtual machine builds instruction includes that the first virtual machine builds instruction；

The maximum that can be set up on the weight scoring highest physical node according to the virtual machine initial parameter matrix computations Virtual machine quantity；

Respective numbers are built on physical node according to the virtual machine initial parameter matrix and the maximum virtual machine quantity Virtual machine.

10. a kind of based on performance requirement and sequence cloud monitoring system construction method as claimed in claim 6, its feature exists In the default virtual machine builds instruction includes that the second virtual machine builds instruction；

Instruction is built by the weight of each physical node in the physical node weight rating matrix according to second virtual machine Scoring is compared with default value, and physical node scale matrix is built according to the comparative result；

Structure weight and each virtual machine of virtual machine initial parameter matrix computations according to each physical node is in each physics The hardware parameter occupied on node, virtual machine is built according to gained is calculated on corresponding physical node.