CN104866382A - Virtual resource scheduling method and virtual resource scheduling device - Google Patents

Abstract

The invention discloses a virtual resource scheduling method and a virtual resource scheduling device. The method comprises the following steps of: collecting service layer parameters and network layer parameters; calculating service layer indexes according to the service layer parameters and calculating network layer indexes according to the network layer parameters; and comparing the service layer indexes and the network layer indexes with respective thresholds, and regulating the resource allocation according to comparison results. The quality required by service development can be ensured.

Description

Schedule virtual resources method and apparatus

Technical field

The invention belongs to field of cloud calculation, particularly relate to schedule virtual resources method and apparatus.

Background technology

Use resource to be the distinguishing feature of cloud computing as required, the resource of main flow scheme of stretching is the CPU, internal memory etc. of monitoring application place virtual machine, carries out scheduling of resource when reaching default threshold values.But directly might not reflect by the index such as CPU, internal memory of virtual machine the actual conditions that business is produced, the time delay of such as business processing and responsiveness have not all met business when carrying out requirement, may the expense of now CPU and internal memory not high.

Summary of the invention

The present inventor finds to have problems in above-mentioned prior art, and therefore proposes a kind of new technical scheme at least one problem in described problem.

According to an aspect of the present invention, propose schedule virtual resources method, comprising:

Capturing service layer parameter and network layer parameter;

According to operation layer parameter computing service layer index, and according to network layer parameter computational grid layer index;

Operation layer index and network layer index are compared with respective threshold value respectively, and according to the distribution of comparative result adjustment to resource.

Further, operation layer index and operation layer upper limit threshold and operation layer lower threshold are compared, when operation layer index is greater than operation layer upper limit threshold, then reduces the resource that it is distributed, when operation layer index is less than operation layer lower threshold, then increase the resource that it is distributed.

Further, network layer index and network layer upper limit threshold and network layer lower threshold are compared, when network layer index is greater than network layer upper limit threshold, then reduces the resource that it is distributed, when network layer index is less than network layer lower threshold, then increase the resource that it is distributed.

Further, the computing formula of network layer index H is:

H=[§ _A(θ _A1×H _A1+...+θ _An×H _An)+§ _B(θ _B1×H _B1+...+θ _Bm×H _Bm)+…§ _N(θ _N1×H _N1+...+

θ _Nk×H _Nk)]×100

Wherein: § _nbe the weight factor of interface protocol type N, span is 0 ~ 1; θ _nkbe the weight factor of the interface k of interface protocol type N, span is 0 ~ 1; H _nkthe health indicator H of the interface k of interface type N _itf;

$H_{\mathrm{itf}}=\frac{R_p\·R_s}{\max (\frac{K_d}{1+\sqrt[3]{K_b}},1)}$

R _prepresent interface requests responsiveness, according to the ratio that number of request and number of responses calculate; R _srepresent interface requests success ratio, according to the ratio that success response number and failure response number calculate; K _drepresent interface requests time delay rate, according to the ratio of average response time delay and normal response time-delay calculation; K _brepresent interface busy rate, according to the ratio that concurrent user number and maximum support number of users calculate.

Further, interface requests success ratio Rs:

Wherein: n is request kind; α is the weight factor of certain class request; M is failure cause kind; β is weight factor; CAU is the failure response number of certain reason; SUC is success response number.

Further, interface requests time delay rate K _d:

$K_d=\underset{i=1}{\overset{N}{\mathrm{\Σ}}}\frac{{\mathrm{\α}}_i\·{\mathrm{DLY}}_i}{{\mathrm{DLS}}_i}$

Wherein: n is request kind; α is the weight factor of certain class request; DLY is the average response time delay of certain class request, minimumly equals DLS; DLS is the normal response time delay of certain class request.

Further, interface busy rate K _b:

$K_b=\frac{U}{U_{\max }}$

Wherein: U _maxfor the maximum number of user that network element design is supported; U is active user number, flow, carrying number, message or message number.

Further, interface requests responsiveness R _p:

$R_p=\frac{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}({\mathrm{\α}}_i\·{\mathrm{RSP}}_i)}{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}({\mathrm{\α}}_i\·{\mathrm{RQT}}_i)}$

Wherein: n is request kind; α is the weight factor of certain class request; RSP is the number of responses sent; RQT is the number of request received.

Further, operation layer index H _ncomputing formula be:

H _N=(T _s×P _s+T _d×P _d)÷log ₂L _d

Wherein: T _sfor sending success ratio weight; T _dfor sending success ratio weight; P _sfor sending success ratio; P _dfor sending success ratio; L _dfor sending time delay rate.

Further, success ratio P is sent _s:

$P_s=[1-\frac{\underset{i=1}{\overset{m}{\mathrm{\Σ}}}({\mathrm{\β}}_i\·C_{\mathrm{Si}})}{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}}]\×\frac{\mathrm{RSPs}}{\mathrm{RQTs}}$

Wherein: m is the reason kind of service error; β is reason weight; Cs sends failure cause number of times for certain; Ss is for sending number of success; RQTs is the number of times of service request; RSPs is the number of times of service response.

Further, success ratio P is sent _d:

$\mathrm{Pd}=[1-\frac{\underset{j=1}{\overset{n}{\mathrm{\Σ}}}(a_i\·{\mathrm{Cd}}_i)}{\underset{j=1}{\overset{n}{\mathrm{\Σ}}}(a_i\·{\mathrm{Cd}}_i)+\mathrm{Sd}}]\×\frac{\mathrm{RQTd}}{\mathrm{RQTs}}$

Wherein: n is the reason kind of business processing; α is reason weight; Cd is certain reason number of times of delivery of notifications failure; Sd is the successful number of times of delivery of notifications; RQTs is the number of times of service request; RQTd is the number of times of service response.

Further, time delay rate L is sent _d:

$L_d=\frac{\mathrm{Davg}}{\mathrm{Dstd}}$

Wherein: Davg is average delivered time delay; Dstd is standard delivery time delay.

According to a further aspect of the invention, propose schedule virtual resources device, comprising:

Operation layer acquisition interface, capturing service layer parameter;

Network layer acquisition interface, collection network layer parameter;

Calculation processing unit, according to operation layer parameter computing service layer index, and according to network layer parameter computational grid layer index; Operation layer index and network layer index are compared with respective threshold value respectively, and according to the distribution of comparative result adjustment to resource.

Further, operation layer index and operation layer upper limit threshold and operation layer lower threshold are compared, when operation layer index is greater than operation layer upper limit threshold, then reduces the resource that it is distributed, when operation layer index is less than operation layer lower threshold, then increase the resource that it is distributed.

Further, network layer index and network layer upper limit threshold and network layer lower threshold are compared, when network layer index is greater than network layer upper limit threshold, then reduces the resource that it is distributed, when network layer index is less than network layer lower threshold, then increase the resource that it is distributed.

Further, the computing formula of network layer index H is:

H=[§ _A(θ _A1×H _A1+...+θ _An×H _An)+§ _B(θ _B1×H _B1+...+θ _Bm×H _Bm)+…§ _N(θ _N1×H _N1+...+

θ _Nk×H _Nk)]×100

Wherein: § _nbe the weight factor of interface protocol type N, span is 0 ~ 1; θ _nkbe the weight factor of the interface k of interface protocol type N, span is 0 ~ 1; H _nkthe health indicator H of the interface k of interface type N _itf;

$H_{\mathrm{itf}}=\frac{R_p\·R_s}{\max (\frac{K_d}{1+\sqrt[3]{\mathrm{Kb}}},1)}$

R _prepresent interface requests responsiveness, according to the ratio that number of request and number of responses calculate; R _srepresent interface requests success ratio, according to the ratio that success response number and failure response number calculate; K _drepresent interface requests time delay rate, according to the ratio of average response time delay and normal response time-delay calculation; K _brepresent interface busy rate, according to the ratio that concurrent user number and maximum support number of users calculate.

Further, interface requests success ratio Rs:

$\mathrm{Rs}=1-\frac{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}[{\mathrm{\α}}_i\·\underset{j=1}{\overset{m}{\mathrm{\Σ}}}({\mathrm{\β}}_j\·{\mathrm{CAU}}_j)]}{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}[{\mathrm{\α}}_i\·\underset{j=1}{\overset{m}{\mathrm{\Σ}}}({\mathrm{\β}}_j\·{\mathrm{CVU}}_j+\mathrm{SUC})]}$

Wherein: n is request kind; α is the weight factor of certain class request; M is failure cause kind; β is weight factor; CAU is the failure response number of certain reason; SUC is success response number.

Further, interface requests time delay rate K _d:

$\mathrm{Kd}=\underset{i=1}{\overset{n}{\mathrm{\Σ}}}\frac{{\mathrm{\α}}_i\·{\mathrm{DLY}}_i}{{\mathrm{DLS}}_i}$

Wherein: n is request kind; α is the weight factor of certain class request; DLY is the average response time delay of certain class request, minimumly equals DLS; DLS is the normal response time delay of certain class request.

Further, interface busy rate K _b:

$K_b=\frac{U}{U_{\max }}$

Wherein: U _maxfor the maximum number of user that network element design is supported; U is active user number, flow, carrying number, message or message number.

Further, interface requests responsiveness R _p:

$R_p=\frac{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}({\mathrm{\α}}_i\·{\mathrm{RSP}}_i)}{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}({\mathrm{\α}}_i\·{\mathrm{RQT}}_i)}$

Wherein: n is request kind; α is the weight factor of certain class request; RSP is the number of responses sent; RQT is the number of request received.

Further, operation layer index H _ncomputing formula be:

H _N=(T _s×P _s+T _d×P _d)÷log ₂L _d

Wherein: T _sfor sending success ratio weight; T _dfor sending success ratio weight; P _sfor sending success ratio; P _dfor sending success ratio; L _dfor sending time delay rate.

Further, success ratio P is sent _s:

$\mathrm{Ps}=[1-\frac{\underset{i=1}{\overset{m}{\mathrm{\Σ}}}({\mathrm{\β}}_i\·{\mathrm{Cs}}_i)}{\underset{i=1}{\overset{m}{\mathrm{\Σ}}}({\mathrm{\β}}_i\·{\mathrm{Cs}}_i)+\mathrm{Ss}}]\×\frac{\mathrm{RSPs}}{\mathrm{RQTs}}$

Wherein: m is the reason kind of service error; β is reason weight; Cs sends failure cause number of times for certain; Ss is for sending number of success; RQTs is the number of times of service request; RSPs is the number of times of service response.

Send success ratio P _d:

$\mathrm{Pd}=[1-\frac{\underset{j=1}{\overset{n}{\mathrm{\Σ}}}(a_i\·{\mathrm{Cd}}_i)}{\underset{j=1}{\overset{n}{\mathrm{\Σ}}}(a_i\·{\mathrm{Cd}}_i)+\mathrm{Sd}}]\×\frac{\mathrm{RQTd}}{\mathrm{RQTs}}$

Wherein: n is the reason kind of business processing; α is reason weight; Cd is certain reason number of times of delivery of notifications failure; Sd is the successful number of times of delivery of notifications; RQTs is the number of times of service request; RQTd is the number of times of service response.

Further, time delay rate L is sent _d:

${L_d}_{}=\frac{\mathrm{Davg}}{\mathrm{Dstd}}$

Wherein: Davg is average delivered time delay; Dstd is standard delivery time delay.

The present invention can ensure that business carries out required quality.

By referring to the detailed description of accompanying drawing to exemplary embodiment of the present invention, further feature of the present invention and advantage thereof will become clear.

Accompanying drawing explanation

What form a part for instructions drawings describes embodiments of the invention, and together with the description for explaining principle of the present invention.

With reference to accompanying drawing, according to detailed description below, clearly the present invention can be understood, wherein:

Figure 1 shows that the schematic flow sheet of the schedule virtual resources method in the embodiment of the present invention.

Figure 2 shows that the structural representation of the schedule virtual resources device in the embodiment of the present invention.

Embodiment

Various exemplary embodiment of the present invention is described in detail now with reference to accompanying drawing.It should be noted that: unless specifically stated otherwise, otherwise positioned opposite, the numerical expression of the parts of setting forth in these embodiments and step and numerical value do not limit the scope of the invention.

Meanwhile, it should be understood that for convenience of description, the size of the various piece shown in accompanying drawing is not draw according to the proportionate relationship of reality.

Illustrative to the description only actually of at least one exemplary embodiment below, never as any restriction to the present invention and application or use.

May not discuss in detail for the known technology of person of ordinary skill in the relevant, method and apparatus, but in the appropriate case, described technology, method and apparatus should be regarded as a part of authorizing instructions.

In all examples with discussing shown here, any occurrence should be construed as merely exemplary, instead of as restriction.Therefore, other example of exemplary embodiment can have different values.

It should be noted that: represent similar terms in similar label and letter accompanying drawing below, therefore, once be defined in an a certain Xiang Yi accompanying drawing, then do not need to be further discussed it in accompanying drawing subsequently.

For making the object, technical solutions and advantages of the present invention clearly understand, below in conjunction with specific embodiment, and with reference to accompanying drawing, the present invention is described in more detail.

Figure 1 shows that the schematic flow sheet of the schedule virtual resources method in the embodiment of the present invention.

In step 11, capturing service layer parameter and network layer parameter.Wherein, operation layer parameter relates to the correlation parameter of service-aware, as business pass out success ratio, business sends the parameters such as time delay; Network layer parameter is the network of relation index parameter that business relates to, as interface protocol accounting, parameter such as network turnover flow, interface requests success ratio etc.

In step 12, according to operation layer parameter computing service layer index, and according to network layer parameter computational grid layer index.

In step 13, operation layer index and network layer index are compared with respective threshold value respectively, and according to the distribution of comparative result adjustment to resource.

Wherein, operation layer index and operation layer upper limit threshold and operation layer lower threshold are compared, when operation layer index is greater than operation layer upper limit threshold, then reduces the resource that it is distributed, when operation layer index is less than operation layer lower threshold, then increase the resource that it is distributed.Resource mentioned here mainly refers to service resources.In this embodiment, operation layer upper limit threshold can also be greater than in operation layer index and continue setting-up time, then reducing the resource that it is distributed; Be less than operation layer lower threshold when operation layer index and continue setting-up time, then increasing the resource that it is distributed, such as, expanding software license, expand number of concurrent.In lasting setting-up time trigger request, the benefit of such setting is exactly that " burr " situation when preventing indivedual causes frequent requests and releasing resource, and generally all consider surplus factor when considering that bound is arranged, the requirement of trigger request after a certain time can be met.In addition, when operation layer index is greater than operation layer lower threshold and is less than upper limit threshold, when namely operation layer index is between lower threshold and upper limit threshold, can not adjust, also can adjust as the case may be.The upper limit threshold mentioned in above-described embodiment and lower threshold all can carry out changing and changing.

Network layer index and network layer upper limit threshold and network layer lower threshold are compared, when network layer index is greater than network layer upper limit threshold, then reduces the resource that it is distributed, when network layer index is less than network layer lower threshold, then increase the resource that it is distributed.Resource mentioned here mainly refers to Internet resources.In this embodiment, network layer upper limit threshold can also be greater than in network layer index and continue setting-up time, then reducing the resource that it is distributed; Be less than network layer lower threshold when network layer index and continue setting-up time, then increasing the resource that it is distributed.In lasting setting-up time trigger request, the benefit of such setting is exactly that " burr " situation when preventing indivedual causes frequent requests and releasing resource, and generally all consider surplus factor when considering that bound is arranged, the requirement of trigger request after a certain time can be met.In addition, when network layer index is greater than network layer lower threshold and is less than upper limit threshold, when namely network layer index is between lower threshold and upper limit threshold, can not adjust, also can adjust as the case may be.The upper limit threshold mentioned in above-described embodiment and lower threshold all can carry out changing and changing.

Index such as hardware such as the single employing of existing resource scheduling strategy CPU, internal memory etc., accurate evaluation business cannot carry out required real resource.This embodiment considers business and carries out the indexs such as the processing power of operation layer and the network layer related to, increase the collection to operation layer and network layer handles ability newly, the business of calculating carries out required stock number, and result of calculation is used for judgement and the control of resource elastic telescopic.Such as, the system of a processing messages conversion distribution, possible situation is that CPU, memory usage are all very low, but simultaneity factor response delay is high especially again, and (service request may need to take delay in 6 seconds, but receptible delay is 100 milliseconds in actual requirement), original in the case technology can't think that business is carried out and had problem.And pass through the above embodiment of the present invention, Problems existing is found by operation layer index or network layer index, if the words of network problem can increase Internet resources, if the words of traffic issues can increase the concurrent mandate number of application to promote business processing speed, the business of guarantee carries out required quality.

Below in conjunction with embodiment, be described in detail to according to the process of network layer parameter computational grid layer index.It will be understood by those skilled in the art that embodiment mentioned here is just for illustrating, should not be construed as limitation of the present invention.On the basis of this embodiment, can revise accordingly and modification.

Define grid layer index H, reflects the health degree of network needed for whole network element, the processing power reflected according to each interface of the external network interaction of network element, stability, busy degree and calculating.Have several several, interface link, so network layer index H just needs to calculate how many items.

The computing formula of network layer index H is:

H=[§ _A(θ _A1×H _A1+...+θ _An×H _An)+§ _B(θ _B1×H _B1+...+θ _Bm×H _Bm)+…§ _N(θ _N1×H _N1+...+

θ _Nk×H _Nk)]×100

Wherein:

§ _athe weight factor of interface protocol type A, § _bthe weight factor of interface protocol type B, § _nit is the weight factor of interface protocol type N.The span 0 ~ 1 of weight factor, that is, be more than or equal to 0 and be less than or equal to 1, and weight factor in other embodiments of the present invention can with reference to the value of the weight factor in this embodiment.Such factor mainly comprises the balance of docking port importance, interface capacity.

θ _a1the weight factor of the interface 1 of interface protocol type A, θ _bmthe weight factor of the interface m of interface protocol type B, θ _nkit is the weight factor of the interface k of interface protocol type N.The span 0 ~ 1 of weight factor, that is, be more than or equal to 0 and be less than or equal to 1, and such factor mainly considers number of users, size of message etc.

H _a1the health indicator of the interface 1 of interface protocol type A, H _bmthe health indicator of the interface m of interface type B, H _nkit is the health indicator of the interface k of interface type N.The computing method of this index are shown in interface health index H _itf.

$H_{\mathrm{itf}}=\frac{R_p\·R_s}{\max (\frac{K_d}{1+\sqrt[3]{K_b}},1)}$

R _prepresent interface requests responsiveness, according to the ratio that number of request and number of responses calculate;

R _srepresent interface requests success ratio, according to the ratio that success response number and failure response number calculate;

K _drepresent interface requests time delay rate, according to the ratio of average response time delay and normal response time-delay calculation;

K _brepresent interface busy rate, according to the ratio that concurrent user number and maximum support number of users calculate.

In this embodiment, R _pr _srepresent R _p× R _s, that is, point here represents multiplication sign.In other embodiments of the invention, also multiplication sign can be represented.

Below by respectively to R _p, R _s, K _dand K _bcomputation process be illustrated.

1) interface requests success ratio Rs:

$\mathrm{Rs}=1-\frac{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}[\mathrm{\αi}\·\underset{j=1}{\overset{m}{\mathrm{\Σ}}}({\mathrm{\β}}_j\·\mathrm{CA}{U}_j)]}{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}[{\mathrm{\α}}_i\·\underset{j=1}{\overset{m}{\mathrm{\Σ}}}({\mathrm{\β}}_j\·\mathrm{CA}{U}_j+\mathrm{SUC})]}$

Wherein:

N is request kind;

α is the weight factor of certain class request;

M is failure cause kind;

β is weight factor, the main degree in close relations considering a variety of causes and this network element;

CAU is the failure response number of certain reason;

SUC is success response number.

2) interface requests time delay rate K _d:

$K_d=\underset{i=1}{\overset{N}{\mathrm{\Σ}}}\frac{{\mathrm{\α}}_i\·{\mathrm{DLY}}_i}{{\mathrm{DLS}}_i}$

Wherein:

N is request kind;

α is the weight factor of certain class request, comprises the balance of the priority to request, disturbance degree;

DLY is the average response time delay of certain class request, minimumly equals DLS, and average response time delay can obtain after receiving request and sending the mistiming statistical summaries calculating mean value between response;

DLS is the normal response time delay of certain class request, obtains in can being required by involved Interface design, also can be arranged according to operation needs by managerial personnel.

3) interface busy rate K _b:

$K_b=\frac{U}{U_{\max }}$

Wherein:

U _maxfor the maximum number of user that network element design is supported;

U is active user number.According to different network elements, number of users can change flow, carrying number, message or message number etc. into.

4) interface requests responsiveness R _p:

$R_p=\frac{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}({\mathrm{\α}}_i\·{\mathrm{RSP}}_i)}{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}({\mathrm{\α}}_i\·{\mathrm{RQT}}_i)}$

Wherein:

N is the request kind received;

α is the weight factor of certain class request, main consideration message count accounting;

RSP is the number of responses sent;

RQT is the number of request received.

In one embodiment, according to above-mentioned parameters computational grid layer index H, suppose that the upper limit threshold of network layer index H is 85, lower threshold is 60, be then respectively:

As H>85, Internet resources can be reduced to resource management platform request, can also determine in conjunction with each classification indicators the type and the quantity that reduce Internet resources;

As H<60, Internet resources can be increased to resource management platform request, can also determine in conjunction with each classification indicators the type and the quantity that increase resource.The type of Internet resources comprises increases bandwidth (upstream bandwidth and downlink bandwidth), amendment routed path etc.

Below in conjunction with embodiment, be described in detail to according to the process of operation layer parameter computing service layer index.It will be understood by those skilled in the art that embodiment mentioned here is just for illustrating, should not be construed as limitation of the present invention.On the basis of this embodiment, can revise accordingly and modification.

Operation layer index H _ncomputing formula be:

H _N=(T _s×P _s+T _d×P _d)÷log ₂L _d

Wherein:

T _sfor sending success ratio weight;

T _dfor sending success ratio weight;

P _sfor sending success ratio;

P _dfor sending success ratio;

L _dfor sending time delay rate.

Below by respectively to P _s, P _dand L _bcomputation process be illustrated.

A) success ratio P is sent _s:

$P_s=[1-\frac{\underset{i=1}{\overset{m}{\mathrm{\&Sigma;}}}({\mathrm{\&beta;}}_i\&CenterDot;C_{\mathrm{Si}})}{\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}}]\&times;\frac{\mathrm{RSPs}}{\mathrm{RQTs}}$

Wherein:

M is the reason kind of service error;

β is reason weight;

Cs sends failure cause number of times for certain;

Ss is for sending number of success;

RQTs is the number of times of service request;

RSPs is the number of times of service response.

B) success ratio P is sent _d:

$\mathrm{Pd}=[1-\frac{\underset{j=1}{\overset{n}{\mathrm{\&Sigma;}}}(a_i\&CenterDot;{\mathrm{Cd}}_i)}{\underset{j=1}{\overset{n}{\mathrm{\&Sigma;}}}(a_i\&CenterDot;{\mathrm{Cd}}_i)+\mathrm{Sd}}]\&times;\frac{\mathrm{RQTd}}{\mathrm{RQTs}}$

Wherein:

N is the reason kind of business processing;

α is reason weight;

Cd is certain reason number of times of delivery of notifications failure;

Sd is the successful number of times of delivery of notifications;

RQTs is the number of times of service request;

RQTd is the number of times of service response.

C) time delay rate L is sent _d:

$L_d=\frac{\mathrm{Davg}}{\mathrm{Dstd}}$

Wherein:

Davg is average delivered time delay.

Dstd is standard delivery time delay.

In the above-described embodiments, the process for computational grid layer index and operation layer index illustrates.Certainly, be not limited to each above-mentioned embodiment, corresponding variants and modifications can be carried out accordingly.The above embodiment of the present invention, consider business and carry out the indexs such as the processing power of operation layer and the network layer related to, the business of calculating carries out required stock number, and result of calculation is used for judgement and the control of resource elastic telescopic.The present invention can ensure that business carries out required quality.

Figure 2 shows that the structural representation of the schedule virtual resources device in the embodiment of the present invention.

Operation layer acquisition interface, capturing service layer parameter.Wherein, operation layer index relates to the correlation parameter of service-aware, as business pass out success ratio, business sends the parameters such as time delay.

Network layer acquisition interface, collection network layer parameter.Wherein, network layer index is the network of relation index parameter that business relates to, as interface protocol accounting, parameter such as network turnover flow, interface requests success ratio etc.

Calculation processing unit, according to operation layer parameter computing service layer index, and according to network layer parameter computational grid layer index; Operation layer index and network layer index are compared with respective threshold value respectively, and according to the distribution of comparative result adjustment to resource.

Wherein, operation layer index and operation layer upper limit threshold and operation layer lower threshold are compared, when operation layer index is greater than operation layer upper limit threshold, then reduces the resource that it is distributed, when operation layer index is less than operation layer lower threshold, then increase the resource that it is distributed.Resource mentioned here mainly refers to service resources.In this embodiment, operation layer upper limit threshold can also be greater than in operation layer index and continue setting-up time, then reducing the resource that it is distributed; Be less than operation layer lower threshold when operation layer index and continue setting-up time, then increasing the resource that it is distributed, such as, expanding software license, expand number of concurrent.In lasting setting-up time trigger request, the benefit of such setting is exactly that " burr " situation when preventing indivedual causes frequent requests and releasing resource, and generally all consider surplus factor when considering that bound is arranged, the requirement of trigger request after a certain time can be met.In addition, when operation layer index is greater than operation layer lower threshold and is less than upper limit threshold, when namely operation layer index is between lower threshold and upper limit threshold, can not adjust, also can adjust as the case may be.The upper limit threshold mentioned in above-described embodiment and lower threshold all can carry out changing and changing.

Network layer index and network layer upper limit threshold and network layer lower threshold are compared, when network layer index is greater than network layer upper limit threshold, then reduces the resource that it is distributed, when network layer index is less than network layer lower threshold, then increase the resource that it is distributed.Resource mentioned here mainly refers to Internet resources.In this embodiment, network layer upper limit threshold can also be greater than in network layer index and continue setting-up time, then reducing the resource that it is distributed; Be less than network layer lower threshold when network layer index and continue setting-up time, then increasing the resource that it is distributed.In lasting setting-up time trigger request, the benefit of such setting is exactly that " burr " situation when preventing indivedual causes frequent requests and releasing resource, and generally all consider surplus factor when considering that bound is arranged, the requirement of trigger request after a certain time can be met.In addition, when network layer index is greater than network layer lower threshold and is less than upper limit threshold, when namely network layer index is between lower threshold and upper limit threshold, can not adjust, also can adjust as the case may be.The upper limit threshold mentioned in above-described embodiment and lower threshold all can carry out changing and changing.

Index such as hardware such as the single employing of existing resource scheduling strategy CPU, internal memory etc., accurate evaluation business cannot carry out required real resource.This embodiment considers business and carries out the indexs such as the processing power of operation layer and the network layer related to, increase the collection to operation layer and network layer handles ability newly, the business of calculating carries out required stock number, and result of calculation is used for judgement and the control of resource elastic telescopic.Such as, the system of a processing messages conversion distribution, possible situation is that CPU, memory usage are all very low, but simultaneity factor response delay is high especially again, and (service request may need to take delay in 6 seconds, but receptible delay is 100 milliseconds in actual requirement), original in the case technology can't think that business is carried out and had problem.And pass through the above embodiment of the present invention, Problems existing is found by operation layer index or network layer index, if the words of network problem can increase Internet resources, if the words of traffic issues can increase the concurrent mandate number of application to promote business processing speed, the business of guarantee carries out required quality.

Below in conjunction with embodiment, be described in detail to according to the process of network layer parameter computational grid layer index.It will be understood by those skilled in the art that embodiment mentioned here is just for illustrating, should not be construed as limitation of the present invention.On the basis of this embodiment, can revise accordingly and modification.

Define grid layer index H, reflects the health degree of network needed for whole network element, the processing power reflected according to each interface of the external network interaction of network element, stability, busy degree and calculating.Have several several, interface link, so network layer index H just needs to calculate how many items.

The computing formula of network layer index H is:

H=[§ _A(θ _A1×H _A1+...+θ _An×H _An)+§ _B(θ _B1×H _B1+...+θ _Bm×H _Bm)+…§ _N(θ _N1×H _N1+...+

θ _Nk×H _Nk)]×100

Wherein:

§ _athe weight factor of interface protocol type A, § _bthe weight factor of interface protocol type B, § _nit is the weight factor of interface protocol type N.The span 0 ~ 1 of weight factor, such factor mainly comprises the balance of docking port importance, interface capacity.

θ _a1the weight factor of the interface 1 of interface protocol type A, θ _bmthe weight factor of the interface m of interface protocol type B, θ _nkit is the weight factor of the interface k of interface protocol type N.The span 0 ~ 1 of weight factor, such factor mainly considers number of users, size of message etc.

H _a1the health indicator of the interface 1 of interface protocol type A, H _bmthe health indicator of the interface m of interface type B, H _nkit is the health indicator of the interface k of interface type N.The computing method of this index are shown in interface health index H _itf.

$H_{\mathrm{itf}}=\frac{R_p\&CenterDot;R_s}{\max (\frac{K_d}{1+\sqrt[3]{\mathrm{Kb}}},1)}$

R _prepresent interface requests responsiveness, according to the ratio that number of request and number of responses calculate;

R _srepresent interface requests success ratio, according to the ratio that success response number and failure response number calculate;

K _drepresent interface requests time delay rate, according to the ratio of average response time delay and normal response time-delay calculation;

K _brepresent interface busy rate, according to the ratio that concurrent user number and maximum support number of users calculate.

Below by respectively to R _p, R _s, K _dand K _bcomputation process be illustrated.

1) interface requests success ratio Rs:

$\mathrm{Rs}=1-\frac{\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}[{\mathrm{\&alpha;}}_i\&CenterDot;\underset{j=1}{\overset{m}{\mathrm{\&Sigma;}}}({\mathrm{\&beta;}}_j\&CenterDot;{\mathrm{CAU}}_j)]}{\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}[{\mathrm{\&alpha;}}_i\&CenterDot;\underset{j=1}{\overset{m}{\mathrm{\&Sigma;}}}({\mathrm{\&beta;}}_j\&CenterDot;{\mathrm{CVU}}_j+\mathrm{SUC})]}$

Wherein:

N is request kind;

α is the weight factor of certain class request;

M is failure cause kind;

β is weight factor, the main degree in close relations considering a variety of causes and this network element;

CAU is the failure response number of certain reason;

SUC is success response number.

2) interface requests time delay rate K _d:

$\mathrm{Kd}=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}\frac{{\mathrm{\&alpha;}}_i\&CenterDot;{\mathrm{DLY}}_i}{{\mathrm{DLS}}_i}$

Wherein:

N is request kind;

α is the weight factor of certain class request, comprises the balance of the priority to request, disturbance degree;

DLY is the average response time delay of certain class request, minimumly equals DLS, and average response time delay can obtain after receiving request and sending the mistiming statistical summaries calculating mean value between response;

DLS is the normal response time delay of certain class request, obtains in can being required by involved Interface design, also can be arranged according to operation needs by managerial personnel.

3) interface busy rate K _b:

$K_b=\frac{U}{U_{\max }}$

Wherein:

U _maxfor the maximum number of user that network element design is supported;

U is active user number.According to different network elements, number of users can change flow, carrying number, message or message number etc. into.

4) interface requests responsiveness R _p:

$R_p=\frac{\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}({\mathrm{\&alpha;}}_i\&CenterDot;{\mathrm{RSP}}_i)}{\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}({\mathrm{\&alpha;}}_i\&CenterDot;{\mathrm{RQT}}_i)}$

Wherein:

N is the request kind received;

α is the weight factor of certain class request, main consideration message count accounting;

RSP is the number of responses sent;

RQT is the number of request received.

In one embodiment, according to above-mentioned parameters computational grid layer index H, suppose that the upper limit threshold of network layer index H is 85, lower threshold is 60, be then respectively:

As H>85, Internet resources can be reduced to resource management platform request, can also determine in conjunction with each classification indicators the type and the quantity that reduce Internet resources;

As H<60, Internet resources can be increased to resource management platform request, can also determine in conjunction with each classification indicators the type and the quantity that increase resource.The type of Internet resources comprises increases bandwidth (upstream bandwidth and downlink bandwidth), amendment routed path etc.

Below in conjunction with embodiment, be described in detail to according to the process of operation layer parameter computing service layer index.It will be understood by those skilled in the art that embodiment mentioned here is just for illustrating, should not be construed as limitation of the present invention.On the basis of this embodiment, can revise accordingly and modification.

Operation layer index H _ncomputing formula be:

H _N=(T _s×P _s+T _d×P _d)÷log ₂L _d

Wherein:

T _sfor sending success ratio weight;

T _dfor sending success ratio weight;

P _sfor sending success ratio;

P _dfor sending success ratio;

L _dfor sending time delay rate.

Below by respectively to P _s, P _dand L _bcomputation process be illustrated.

A) success ratio P is sent _s:

$\mathrm{Ps}=[1-\frac{\underset{i=1}{\overset{m}{\mathrm{\&Sigma;}}}({\mathrm{\&beta;}}_i\&CenterDot;{\mathrm{Cs}}_i)}{\underset{i=1}{\overset{m}{\mathrm{\&Sigma;}}}({\mathrm{\&beta;}}_i\&CenterDot;{\mathrm{Cs}}_i)+\mathrm{Ss}}]\&times;\frac{\mathrm{RSPs}}{\mathrm{RQTs}}$

Wherein:

M is the reason kind of service error;

β is reason weight;

Cs sends failure cause number of times for certain;

Ss is for sending number of success;

RQTs is the number of times of service request;

RSPs is the number of times of service response.

B) success ratio P is sent _d:

$\mathrm{Pd}=[1-\frac{\underset{j=1}{\overset{n}{\mathrm{\&Sigma;}}}(a_i\&CenterDot;{\mathrm{Cd}}_i)}{\underset{j=1}{\overset{n}{\mathrm{\&Sigma;}}}(a_i\&CenterDot;{\mathrm{Cd}}_i)+\mathrm{Sd}}]\&times;\frac{\mathrm{RQTd}}{\mathrm{RQTs}}$

Wherein:

N is the reason kind of business processing;

α is reason weight;

Cd is certain reason number of times of delivery of notifications failure;

Sd is the successful number of times of delivery of notifications;

RQTs is the number of times of service request;

RQTd is the number of times of service response.

C) time delay rate L is sent _d:

${L_d}_{}=\frac{\mathrm{Davg}}{\mathrm{Dstd}}$

Wherein:

Davg is average delivered time delay.

Dstd is standard delivery time delay.

In the above-described embodiments, the process for computational grid layer index and operation layer index illustrates.Certainly, be not limited to each above-mentioned embodiment, corresponding variants and modifications can be carried out accordingly.The above embodiment of the present invention, consider business and carry out the indexs such as the processing power of operation layer and the network layer related to, the business of calculating carries out required stock number, and result of calculation is used for judgement and the control of resource elastic telescopic.The present invention can ensure that business carries out required quality.

So far, the present invention is described in detail.In order to avoid covering design of the present invention, details more known in the field are not described.Those skilled in the art, according to description above, can understand how to implement technical scheme disclosed herein completely.

Method of the present invention and device may be realized in many ways.Such as, any combination by software, hardware, firmware or software, hardware, firmware realizes method of the present invention and device.Said sequence for the step of described method is only to be described, and the step of method of the present invention is not limited to above specifically described order, unless specifically stated otherwise.In addition, in certain embodiments, can be also record program in the recording medium by the invention process, these programs comprise the machine readable instructions for realizing according to method of the present invention.Thus, the present invention also covers the recording medium stored for performing the program according to method of the present invention.

Although be described in detail specific embodiments more of the present invention by example, it should be appreciated by those skilled in the art, above example is only to be described, instead of in order to limit the scope of the invention.It should be appreciated by those skilled in the art, can without departing from the scope and spirit of the present invention, above embodiment be modified.Scope of the present invention is limited by claims.

Claims (24)

1. schedule virtual resources method, is characterized in that:

Capturing service layer parameter and network layer parameter;

According to operation layer parameter computing service layer index, and according to network layer parameter computational grid layer index;

Operation layer index and network layer index are compared with respective threshold value respectively, and according to the distribution of comparative result adjustment to resource.

2. schedule virtual resources method according to claim 1, is characterized in that:

Operation layer index and operation layer upper limit threshold and operation layer lower threshold are compared, when operation layer index is greater than operation layer upper limit threshold, then reduces the resource that it is distributed, when operation layer index is less than operation layer lower threshold, then increase the resource that it is distributed.

3. schedule virtual resources method according to claim 1, is characterized in that:

Network layer index and network layer upper limit threshold and network layer lower threshold are compared, when network layer index is greater than network layer upper limit threshold, then reduces the resource that it is distributed, when network layer index is less than network layer lower threshold, then increase the resource that it is distributed.

4. schedule virtual resources method according to claim 1 or 2 or 3, is characterized in that:

The computing formula of network layer index H is:

H=[§ _A(θ _A1×H _A1+...+θ _An×H _An)+§ _B(θ _B1×H _B1+...+θ _Bm×H _Bm)+…§ _N(θ _N1×H _N1+...+

θ _Nk×H _Nk)]×100

Wherein:

§ _nbe the weight factor of interface protocol type N, span is 0 ~ 1;

θ _nkbe the weight factor of the interface k of interface protocol type N, span is 0 ~ 1;

H _nkthe health indicator H of the interface k of interface type N _itf;

$H_{\mathrm{itf}}=\frac{R_p\&CenterDot;R_s}{\max (\frac{K_d}{1+\sqrt[3]{K_b}},1)}$

R _prepresent interface requests responsiveness, according to the ratio that number of request and number of responses calculate;

R _srepresent interface requests success ratio, according to the ratio that success response number and failure response number calculate;

K _drepresent interface requests time delay rate, according to the ratio of average response time delay and normal response time-delay calculation;

K _brepresent interface busy rate, according to the ratio that concurrent user number and maximum support number of users calculate.

5. schedule virtual resources method according to claim 4, is characterized in that:

Interface requests success ratio Rs:

$\mathrm{Rs}=1-\frac{\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}[\mathrm{\&alpha;i}\&CenterDot;\underset{j=1}{\overset{m}{\mathrm{\&Sigma;}}}({\mathrm{\&beta;}}_j\&CenterDot;\mathrm{CA}{U}_j)]}{\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}[{\mathrm{\&alpha;}}_i\&CenterDot;\underset{j=1}{\overset{m}{\mathrm{\&Sigma;}}}({\mathrm{\&beta;}}_j\&CenterDot;\mathrm{CA}{U}_j+\mathrm{SUC})]}$

Wherein:

N is request kind;

α is the weight factor of certain class request;

M is failure cause kind;

β is weight factor;

CAU is the failure response number of certain reason;

SUC is success response number.

6. schedule virtual resources method according to claim 4, is characterized in that:

Interface requests time delay rate K _d:

$K_d=\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}\frac{{\mathrm{\&alpha;}}_i\&CenterDot;{\mathrm{DLY}}_i}{{\mathrm{DLS}}_i}$

Wherein:

N is request kind;

α is the weight factor of certain class request;

DLY is the average response time delay of certain class request, minimumly equals DLS;

DLS is the normal response time delay of certain class request.

7. schedule virtual resources method according to claim 4, is characterized in that:

Interface busy rate K _b:

$K_b=\frac{U}{U_{\max }}$

Wherein:

U _maxfor the maximum number of user that network element design is supported;

U is active user number, flow, carrying number, message or message number.

8. schedule virtual resources method according to claim 4, is characterized in that:

Interface requests responsiveness R _p:

$R_p=\frac{\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}({\mathrm{\&alpha;}}_i\&CenterDot;{\mathrm{RSP}}_i)}{\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}({\mathrm{\&alpha;}}_i\&CenterDot;{\mathrm{RQT}}_i)}$

Wherein:

N is request kind;

α is the weight factor of certain class request;

RSP is the number of responses sent;

RQT is the number of request received.

9. schedule virtual resources method according to claim 1 or 2 or 3, is characterized in that:

Operation layer index H _ncomputing formula be:

H _N=(T _s×P _s+T _d×P _d)÷log ₂L _d

Wherein:

T _sfor sending success ratio weight;

T _dfor sending success ratio weight;

P _sfor sending success ratio;

P _dfor sending success ratio;

L _dfor sending time delay rate.

10. schedule virtual resources method according to claim 9, is characterized in that:

Send success ratio P _s:

$P_s=[1-\frac{\underset{i=1}{\overset{m}{\mathrm{\&Sigma;}}}({\mathrm{\&beta;}}_i\&CenterDot;C_{\mathrm{Si}})}{\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}}]\&times;\frac{\mathrm{RSPs}}{\mathrm{RQTs}}$

Wherein:

M is the reason kind of service error;

β is reason weight;

Cs sends failure cause number of times for certain;

Ss is for sending number of success;

RQTs is the number of times of service request;

RSPs is the number of times of service response.

11. schedule virtual resources methods according to claim 9, is characterized in that:

Send success ratio P _d:

$\mathrm{Pd}=[1-\frac{\underset{j=1}{\overset{n}{\mathrm{\&Sigma;}}}(a_i\&CenterDot;{\mathrm{Cd}}_i)}{\underset{j=1}{\overset{n}{\mathrm{\&Sigma;}}}(a_i\&CenterDot;{\mathrm{Cd}}_i)+\mathrm{Sd}}]\&times;\frac{\mathrm{RQTd}}{\mathrm{RQTs}}$

Wherein:

N is the reason kind of business processing;

α is reason weight;

Cd is certain reason number of times of delivery of notifications failure;

Sd is the successful number of times of delivery of notifications;

RQTs is the number of times of service request;

RQTd is the number of times of service response.

12. schedule virtual resources methods according to claim 9, is characterized in that:

Send time delay rate L _d:

$L_d=\frac{\mathrm{Davg}}{\mathrm{Dstd}}$

Wherein:

Davg is average delivered time delay;

Dstd is standard delivery time delay.

13. schedule virtual resources devices, is characterized in that:

Operation layer acquisition interface, capturing service layer parameter;

Network layer acquisition interface, collection network layer parameter;

Calculation processing unit, according to operation layer parameter computing service layer index, and according to network layer parameter computational grid layer index; Operation layer index and network layer index are compared with respective threshold value respectively, and according to the distribution of comparative result adjustment to resource.

14., according to schedule virtual resources device described in claim 13, is characterized in that:

Operation layer index and operation layer upper limit threshold and operation layer lower threshold are compared, when operation layer index is greater than operation layer upper limit threshold, then reduces the resource that it is distributed, when operation layer index is less than operation layer lower threshold, then increase the resource that it is distributed.

15., according to schedule virtual resources device described in claim 13, is characterized in that:

Network layer index and network layer upper limit threshold and network layer lower threshold are compared, when network layer index is greater than network layer upper limit threshold, then reduces the resource that it is distributed, when network layer index is less than network layer lower threshold, then increase the resource that it is distributed.

16. according to claim 13 or 14 or 15 schedule virtual resources device, it is characterized in that:

The computing formula of network layer index H is:

H=[§ _A(θ _A1×H _A1+...+θ _An×H _An)+§ _B(θ _B1×H _B1+...+θ _Bm×H _Bm)+…§ _N(θ _N1×H _N1+...+

θ _Nk×H _Nk)]×100

Wherein:

§ _nbe the weight factor of interface protocol type N, span is 0 ~ 1;

θ _nkbe the weight factor of the interface k of interface protocol type N, span is 0 ~ 1;

H _nkthe health indicator H of the interface k of interface type N _itf;

$H_{\mathrm{itf}}=\frac{R_p\&CenterDot;R_s}{\max (\frac{K_d}{1+\sqrt[3]{\mathrm{Kb}}},1)}$

R _prepresent interface requests responsiveness, according to the ratio that number of request and number of responses calculate;

R _srepresent interface requests success ratio, according to the ratio that success response number and failure response number calculate;

K _drepresent interface requests time delay rate, according to the ratio of average response time delay and normal response time-delay calculation;

K _brepresent interface busy rate, according to the ratio that concurrent user number and maximum support number of users calculate.

17., according to schedule virtual resources device described in claim 16, is characterized in that:

Interface requests success ratio Rs:

$\mathrm{Rs}=1-\frac{\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}[{\mathrm{\&alpha;}}_i\&CenterDot;\underset{j=1}{\overset{m}{\mathrm{\&Sigma;}}}({\mathrm{\&beta;}}_j\&CenterDot;{\mathrm{CAU}}_j)]}{\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}[{\mathrm{\&alpha;}}_i\&CenterDot;\underset{j=1}{\overset{m}{\mathrm{\&Sigma;}}}({\mathrm{\&beta;}}_j\&CenterDot;{\mathrm{CVU}}_j+\mathrm{SUC})]}$

Wherein:

N is request kind;

α is the weight factor of certain class request;

M is failure cause kind;

β is weight factor;

CAU is the failure response number of certain reason;

SUC is success response number.

18., according to schedule virtual resources device described in claim 16, is characterized in that:

Interface requests time delay rate K _d:

$\mathrm{Kd}=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}\frac{{\mathrm{\&alpha;}}_i\&CenterDot;{\mathrm{DLY}}_i}{{\mathrm{DLS}}_i}$

Wherein:

N is request kind;

α is the weight factor of certain class request;

DLY is the average response time delay of certain class request, minimumly equals DLS;

DLS is the normal response time delay of certain class request.

19., according to schedule virtual resources device described in claim 16, is characterized in that:

Interface busy rate K _b:

$K_b=\frac{U}{U_{\max }}$

Wherein:

U _maxfor the maximum number of user that network element design is supported;

U is active user number, flow, carrying number, message or message number.

20., according to schedule virtual resources device described in claim 16, is characterized in that:

Interface requests responsiveness R _p:

$R_p=\frac{\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}({\mathrm{\&alpha;}}_i\&CenterDot;{\mathrm{RSP}}_i)}{\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}({\mathrm{\&alpha;}}_i\&CenterDot;{\mathrm{RQT}}_i)}$

Wherein:

N is request kind;

α is the weight factor of certain class request;

RSP is the number of responses sent;

RQT is the number of request received.

21. according to claim 13 or 14 or 15 schedule virtual resources device, it is characterized in that:

Operation layer index H _ncomputing formula be:

H _N=(T _s×P _s+T _d×P _d)÷log ₂L _d

Wherein:

T _sfor sending success ratio weight;

T _dfor sending success ratio weight;

P _sfor sending success ratio;

P _dfor sending success ratio;

L _dfor sending time delay rate.

22., according to schedule virtual resources device described in claim 21, is characterized in that:

Send success ratio P _s:

$\mathrm{Ps}=[1-\frac{\underset{i=1}{\overset{m}{\mathrm{\&Sigma;}}}({\mathrm{\&beta;}}_i\&CenterDot;{\mathrm{Cs}}_i)}{\underset{i=1}{\overset{m}{\mathrm{\&Sigma;}}}({\mathrm{\&beta;}}_i\&CenterDot;{\mathrm{Cs}}_i)+\mathrm{Ss}}]\&times;\frac{\mathrm{RSPs}}{\mathrm{RQTs}}$

Wherein:

M is the reason kind of service error;

β is reason weight;

Cs sends failure cause number of times for certain;

Ss is for sending number of success;

RQTs is the number of times of service request;

RSPs is the number of times of service response.

23., according to schedule virtual resources device described in claim 21, is characterized in that:

Send success ratio P _d:

$\mathrm{Pd}=[1-\frac{\underset{j=1}{\overset{n}{\mathrm{\&Sigma;}}}(a_i\&CenterDot;{\mathrm{Cd}}_i)}{\underset{j=1}{\overset{n}{\mathrm{\&Sigma;}}}(a_i\&CenterDot;{\mathrm{Cd}}_i)+\mathrm{Sd}}]\&times;\frac{\mathrm{RQTd}}{\mathrm{RQTs}}$

Wherein:

N is the reason kind of business processing;

α is reason weight;

Cd is certain reason number of times of delivery of notifications failure;

Sd is the successful number of times of delivery of notifications;

RQTs is the number of times of service request;

RQTd is the number of times of service response.

24., according to schedule virtual resources device described in claim 21, is characterized in that:

Send time delay rate L _d:

${L_d}_{}=\frac{\mathrm{Davg}}{\mathrm{Dstd}}$

Wherein:

Davg is average delivered time delay;

Dstd is standard delivery time delay.