CN103473100A - Dependence-relationship-based Web service replacing method - Google Patents

Abstract

The invention relates to a dependence-relationship-based Web service replacing method, and belongs to the technical field of computer software. The method comprises the following steps of when certain Web service cannot be consistent with changes in an external environment or the requirements of a user, identifying the deployment condition and the version information of the problem service, and the direct precursor service and the direct subsequent service of the problem service in composite service; selecting proper service for replacement according to different conditions by combining a dependence relationship extracted from a problem service composite description file; when the problem service does not have any other deployment and version, and direct subsequent service sets of services in a similar service set comprise all the subsequent service of the problem service, calculating similarities between the similar services and the problem service in terms of input, output and constraint conditions and a service interaction relationship established between the problem service and the similar services in a specific field, and selecting the most matched service for replacement.

Description

A kind of Web service replacement method based on dependence

Technical field

The present invention relates to a kind of replacement method of Web service, relate in particular to the Web service replacement method based on dependence in a kind of large-scale complex software systems, belong to computer software technical field.

Background technology

Web service is the autonomy of being issued under Networks Environment by the ISP, stateless, as to have standalone feature computational entity.Service-oriented computing utilizes Web service as the essential structure unit, support Distributed Application fast, combined type exploitation cheaply.Based on Service-Oriented Architecture Based (Service-Oriented Architecture, SOA), software serves (SaaS) and enterprise is bought to software in cloud computing and maintain infrastructure changes into by a large amount of leases in internet consumption service service.Yet, along with service-oriented software systems scale is increasing, the problems such as its availability, reliability more and more are difficult to guarantee.When Web service can not be tackled the variation of external environment condition or user's request, in order to ensure the correct execution of composite services, need to replace immediately this Web service.

Through the retrieval to prior art, find, Chinese Patent Application No. 201110058676.2, put down in writing a kind of Web service replacement method based on service priority.Choose the service that in architecture, importance is strong and first replaced, and preferentially select the high service of candidate service centralized services quality, repair the composite service of inefficacy by the mode of replacing the part Web service.The method is not considered the restriction relation between Web service, and directly replacement may exert an influence to the Web service with its Constrained relation.

In service-oriented computing environment, between the service of distributed, different grain size, different field, call mutually, all exist complementary contact.Chinese Patent Application No. 201210159395.0, put down in writing under a kind of open network the method for building up of incidence relation between serving, execution route, anabolic process and service implementation process according to composite services, analyze the incidence relation of composite services in different step, set up substep relationship maps model, build the association service resource pool.Chinese Patent Application No. 201110168027.8, put down in writing a kind of Web service combined method based on dependence, proposed to describe the model of the data dependence information of Web service, in the Service Matching process, utilize the dependence of Web service that the Web service combination is changed into to dynamic programming problems, and find out optimum service combining method.Chinese Patent Application No. 201010289364.8, put down in writing a kind of Web application deployment constraint automatic testing method, deployment to Web application deployment scheme model retrains automatic examinations, when the current Web application being detected to after disposing running counter to of retraining, automatically locate and point out the user to carry out the modification of deployment scheme.

In sum, for dependence, current existing relevant programme is applied to the dependence between Web service in the Services Composition process mostly, and the Web service replacement method research approach based on dependence is less.

Summary of the invention

The object of the invention is to: overcome the deficiencies in the prior art, a kind of Web service replacement method based on dependence is provided, guarantee the correct execution of composite services under cloud computing environment.

The technology of the present invention solution comprises: a kind of Web service replacement method based on dependence, and performing step is:

Step 1, when a certain Web service can not meet the variation of external environment condition or user's request in a certain position, deployment scenario, version information and the forerunner's set of service in composite services and the follow-up set of service of the described a certain Web service that at first identification goes wrong;

Step 2, extract I/O, constraint condition and call relation in the described a certain Web service description document gone wrong, and the input and output of the Web service of problem described in following steps, constraint condition and follow-up set of service all obtain by this step;

Step 3, if the described a certain Web service gone wrong exists and disposes in other positions, check whether the Web service that is deployed in other positions meets external environment condition or user's request, if exist a Web service to meet, this problem Web service is replaced with to a described Web service that meets external environment condition or user's needs, jump to step 8, otherwise perform step 4;

Step 4, if there are other versions in the described a certain Web service gone wrong, using above-mentioned every other version as similar set, compare each Web service in similar set and the described a certain Web service gone wrong, and performs step 6; If there are not other versions in the described a certain Web service gone wrong, perform step 5;

Step 5, if other of the described a certain Web service that this goes wrong are disposed and other versions do not exist, judge one by one whether the immediate successor set of service of other Web services comprises all follow-up service of problem Web service, will comprise all other the follow-up Web services of problem Web service and put similar set under;

Step 6, calculate the similarity of each similar Web service and problem Web service in similar set, obtains the input and output matrix M _{in_out}, constraint matrix M _constraint, the service interaction matrix M _correlation;

Step 7, according to the input and output matrix M in step 6 _{in_out}, constraint matrix M _constraint, the service interaction matrix M _correlationcalculate the dependence matrix M, Web service is sorted, recommend out to replace the described a certain Web service gone wrong;

Step 8, the use request that the interception user sends to the problem service, replace this problem Web service.

The described direct precursor service of step 1 and immediate successor service are: suppose to have dependence to be<s ₁, s ₂,<s ₂, s ₃, mean Web service s ₁depend on Web service s ₂, Web service s ₂depend on Web service s ₃, s so ₁s ₂direct precursor, s ₃s ₂immediate successor.

The method of extracting the described a certain Web service composite descriptor gone wrong in described step 2 is: the composite file of SCA is expanded,<service > label means the service that assembly can provide, its level attribute representation should serve residing level;<function > functional method that provides from interface and behavior two aspects definition services with the form of operation of label;<reference > label means the service that assembly relies on, the service that its relation attribute representation quotes is in interlayer or layer;<function > label means that needs are provided input parameter, the output parameter requirement of operation by the service that relies on, refType, for the dependence type, has sequential relationship and mandatory relationship.

The described a certain Web service in described step 3, this gone wrong replaces with the described concrete grammar that this meets a Web service of external environment condition or user's needs: relatively whether the QoS of Web service meets tenant's requirement of using the described a certain Web service gone wrong, selects best one of meeting property.

The input and output matrix representation of similar Web service in described step 6 is as formula (1), io _ijthe similarity that means input and output between Web service i and Web service j, 0<io _ij<1; Computing method are as follows:

The matching degree of the input and output between two Web services calculates by formula (2), equates to refer to that the output parameter of Web service i is equal to the input parameter of Web service j, io fully _ij=1; Comprise the input parameter set that the output parameter set that refers to Web service i belongs to Web service j; The input parameter set that part comprises the output parameter set-inclusion Web service j that refers to Web service i; The output parameter set of unequal finger Web service i and the input parameter set of Web service j do not comprise mutually concerns io _ij=0; For comprising or part while comprising situation, need to calculate similarity io when matching degree _ijas formula (3), d means two distances between Web service; H means the height of two Web service child Web services common with it; μ, ν represents respectively d and the shared weight of h, 0<μ<1,0<ν<1, μ+ν=1;

$M_{\mathrm{in}\_\mathrm{out}}=\left[\begin{array}{cccc}{\mathrm{io}}_{11}& {\mathrm{io}}_{12}& ..& {\mathrm{io}}_{1n}\\ {\mathrm{io}}_{21}& {\mathrm{io}}_{22}& ..& {\mathrm{io}}_{2n}\\ ..& ..& ..& ..\\ {\mathrm{io}}_{n1}& {\mathrm{io}}_{n2}& ..& {\mathrm{io}}_{\mathrm{nn}}\end{array}\right]---\left(1\right)$

$\mathrm{similarity}=e^{-\mathrm{\&mu;d}}\left(\frac{e^{\mathrm{vh}}-e^{-\mathrm{vh}}}{e^{\mathrm{vh}}+e^{\mathrm{vh}}}\right)---\left(3\right).$

Constraint matrix M in described step 6 _constraintmean as formula (4), con _ijthe similarity that means constraint condition between Web service i and Web service j, 0<con _ij<1, computing method are as follows:

Constrained matching degree between two Web services calculates by formula (5), equates to refer to that the constraint condition of Web service i is equal to the constraint condition of Web service j, con fully _ij=1; Comprise the constraint condition set that the constraint condition set that refers to Web service i belongs to Web service j; The constraint condition set that the constraint condition set that part comprises finger Web service i comprises Web service j; The constraint condition set of unequal finger Web service i and the constraint condition set of Web service j; What do not comprise mutually concerns con _ij=0; For comprising or part while comprising situation, need to calculate similarity con when matching degree _ijas formula (6), d means two distances between Web service; H means the height of two Web service child Web services common with it; μ, ν represents respectively d and the shared weight of h;

$M_{\mathrm{constraint}}=\left[\begin{array}{cccc}{\mathrm{con}}_{11}& {\mathrm{con}}_{12}& ..& {\mathrm{con}}_{1n}\\ {\mathrm{con}}_{21}& {\mathrm{con}}_{22}& ..& {\mathrm{con}}_{2n}\\ ..& ..& ..& ..\\ {\mathrm{con}}_{n1}& {\mathrm{con}}_{n2}& ..& {\mathrm{con}}_{\mathrm{nn}}\end{array}\right]---\left(4\right)$

$\mathrm{similarity}=e^{-\mathrm{\&mu;d}}\left(\frac{e^{\mathrm{vh}}-e^{-\mathrm{vh}}}{e^{\mathrm{vh}}+e^{\mathrm{vh}}}\right)---\left(6\right)$

Service interaction matrix M in described step 6 _correlationvalue be multiplied by dependency degree and obtain by relying on the type weights.

In described step 7, dependence matrix M computing formula is:

$M=\left[\begin{array}{cccc}{d}_{11}& d_{12}& ..& d_{1n}\\ d_{21}& d_{22}& ..& d_{2n}\\ ..& ..& ..& ..\\ d_{n1}& d_{n2}& ..& d_{\mathrm{nn}}\end{array}\right]=M_{\mathrm{in}\_\mathrm{out}}*M_{\mathrm{constraint}}*M_{\mathrm{correlation}}---\left(4\right).$

The present invention's advantage and good effect compared with prior art is as follows:

(1) the present invention is divided into service substitution other deployment, other versions, other similar services of problem service, has solved the service substitution problem comprehensively, guarantees the true(-)running of large-scale complex software systems.

(2) the present invention is based on complicated dependence between Web service, except the coupling of carrying out input and output, also carry out the similarity of restriction relation and calculate, and the business dependence of setting up between the inner question service of computing service field and similar service, the degree of accuracy of replacement improved.

The accompanying drawing explanation

Fig. 1 is the embodiments of the invention process flow diagram;

Embodiment

As shown in Figure 1, the Web service replacement method based on dependence that the embodiment of the present invention is set up is: when a certain Web service can not meet the variation of external environment condition or user's request, identify deployment scenario, version information and the direct precursor service in composite services and the immediate successor service of problem service; In conjunction with the dependence of extracting from problem service composite description document, select suitable service according to different situations and replaced; When other deployment of problem service do not exist with other versions, the all follow-up service of the immediate successor set-inclusion problem service of the service in similar set of service, calculate they and the similarity of problem service on input and output, constraint condition, interactive relation, select the service of coupling to be replaced.

The specific implementation step is as follows:

Step 1, when a certain Web service can not meet the variation of external environment condition or user's request, at first identify deployment scenario, version information and the direct precursor service in composite services and the immediate successor service of problem service.Suppose to have dependence to be<s ₁, s ₂,<s ₂, s ₃, mean Web service s ₁depend on Web service s ₂, Web service s ₂depend on Web service s ₃, s so ₁s ₂direct precursor, s ₃s ₂immediate successor.

Step 2, extract I/O, constraint condition and call relation in this problem service composite description document.The composite file of service component architecture SCA (Service Composite Architecture, SCA) is expanded.<service > label means the service that assembly can provide, its level attribute representation should serve residing level;<function > functional method that provides from interface and behavior two aspects definition services with the form of operation of label.<reference > label means the service that assembly relies on, the service that its relation attribute representation quotes is in interlayer or layer;<function > label means that needs are provided the requirements such as input parameter, output parameter of operation by the service that relies on, refType is the dependence type, and sequential relationship, mandatory relationship etc. are arranged.

Step 3, if this problem service exists and disposes in other positions, check whether the Web service that is deployed in other positions meets external environment condition or user's request, relatively whether the QoS of Web service meets tenant's requirement of use problem Web service, selects best one of meeting property.If exist a Web service to meet, by the problem service substitution, be this service, jump to step 8, otherwise perform step 4.

Step 4, if there are other versions in the problem service, be similar set by the version sets cooperation, and the Web service in the version set and problem service are compared, and performs step 6; If there are not other versions in the problem service, perform step 5;

Step 5, if other of this problem service dispose and other versions do not exist, judge that whether the immediate successor set of other Web services comprises all follow-up service of problem service, puts this Web service under similar set of service if comprise one by one;

Step 6, calculate respectively the input and output matrix M of similar service _{in_out}, constraint matrix M _constraint, the service interaction matrix M _correlation.The input and output matrix representation of similar service is as formula (1), io _ijthe dependency degree that means input and output between service i and service j, 0<io _ij<1; Computing method are as follows:

The matching degree of the input and output between two services calculates by formula (2), equates to refer to the output parameter set out_s of service i fully _ibe equal to the input parameter set in_s of service j _j, io _ij=1; Comprise the output parameter set out_s that refers to service i _ibelong to the input parameter set in_s that serves j _j; Part comprises the output parameter set out_s that refers to service i _icomprise the input parameter set in_s that serves j _j; The output parameter set out_s of unequal finger service i _iinput parameter set in_s with service j _jwhat do not comprise mutually concerns io _ij=0; For comprising or part while comprising situation, need to calculate similarity io when matching degree _ij, seeing formula (3), d means two distances between service; H means the height of two service child services common with it; μ, ν represents respectively d and the shared weight of h, 0<μ<1,0<ν<1, μ+ν=1.

Constraint matrix means to serve the matching degree of constraint condition between i and service j, M _constraintcomputing method are with reference to the computing method of input and output matrix.

Service interaction refers to the business dependence of setting up between the service of specific area inner question and similar service, service interaction matrix M _correlationvalue be multiplied by dependency degree and obtain by relying on the type weights.

$M_{\mathrm{in}\_\mathrm{out}}=\left[\begin{array}{cccc}{\mathrm{io}}_{11}& {\mathrm{io}}_{12}& ..& {\mathrm{io}}_{1n}\\ {\mathrm{io}}_{21}& {\mathrm{io}}_{22}& ..& {\mathrm{io}}_{2n}\\ ..& ..& ..& ..\\ {\mathrm{io}}_{n1}& {\mathrm{io}}_{n2}& ..& {\mathrm{io}}_{\mathrm{nn}}\end{array}\right]---\left(1\right)$

$\mathrm{similarity}=e^{-\mathrm{\&mu;d}}\left(\frac{e^{\mathrm{vh}}-e^{-\mathrm{vh}}}{e^{\mathrm{vh}}+e^{\mathrm{vh}}}\right)---\left(3\right)$

Step 7, calculate the dependence matrix M, and service is sorted, recommend out can the replacement problem service Web service.

$M=\left[\begin{array}{cccc}{d}_{11}& d_{12}& ..& d_{1n}\\ d_{21}& d_{22}& ..& d_{2n}\\ ..& ..& ..& ..\\ d_{n1}& d_{n2}& ..& d_{\mathrm{nn}}\end{array}\right]=M_{\mathrm{in}\_\mathrm{out}}*M_{\mathrm{constraint}}*M_{\mathrm{correlation}}---\left(4\right).$

Step 8, the interception service request, replace this problem service.

Non-elaborated part of the present invention belongs to those skilled in the art's known technology.

The above; it is only the embodiment in the present invention; but protection scope of the present invention is not limited to this; anyly be familiar with the people of this technology in the disclosed technical scope of the present invention; can understand conversion or the replacement expected; all should be encompassed in of the present invention comprise scope within, therefore, protection scope of the present invention should be as the criterion with the protection domain of claims.

Claims (8)

1. the Web service replacement method based on dependence is characterized in that comprising the following steps:

Step 1, when a certain Web service can not meet the variation of external environment condition or user's request in a certain position, deployment scenario, version information and the forerunner's set of service in composite services and the follow-up set of service of the described a certain Web service that at first identification goes wrong;

Step 2, extract I/O, constraint condition and call relation in the described a certain Web service description document gone wrong, and input/output argument, the constraint condition of the Web service of problem described in following steps all obtain by this step;

Step 3, if the described a certain Web service gone wrong exists and disposes in other positions, check whether the Web service that is deployed in other positions meets external environment condition or user's request, if exist a Web service to meet, this problem Web service is replaced with to a described Web service that meets external environment condition or user's needs, jump to step 8, otherwise perform step 4;

Step 4, if there are other versions in the described a certain Web service gone wrong, using above-mentioned every other version as similar set, compare each Web service in similar set and the described a certain Web service gone wrong, and performs step 6; If there are not other versions in the described a certain Web service gone wrong, perform step 5;

Step 5, if other of the described a certain Web service that this goes wrong are disposed and other versions do not exist, judge one by one whether the immediate successor set of service of other Web services comprises all follow-up service of problem Web service, will comprise all other the follow-up Web services of problem Web service and put similar set under;

Step 6, calculate the similarity of each similar Web service and problem Web service in similar set, obtains the input and output matrix M _{in_out}, constraint matrix M _constraint, the service interaction matrix M _correlation;

Step 7, according to the input and output matrix M in step 6 _{in_out}, constraint matrix M _constraint, the service interaction matrix M _correlationcalculate the dependence matrix M, Web service is sorted, recommend out to replace the described a certain Web service gone wrong;

Step 8, the use request that the interception user sends to the problem service, replace this problem Web service.

2. the Web service replacement method based on dependence according to claim 1, it is characterized in that: the described direct precursor service of step 1 and immediate successor service are: suppose to have dependence to be<s ₁, s ₂,<s ₂, s ₃, mean Web service s ₁depend on Web service s ₂, Web service s ₂depend on Web service s ₃, s so ₁s ₂direct precursor, s ₃s ₂immediate successor.

3. the Web service replacement method based on dependence according to claim 1, it is characterized in that: the method for extracting the described a certain Web service composite descriptor gone wrong in described step 2 is: the composite file of SCA is expanded,<service > label means the service that assembly can provide, its level attribute representation should serve residing level;<function > functional method that provides from interface and behavior two aspects definition services with the form of operation of label;<reference > label means the service that assembly relies on, the service that its relation attribute representation quotes is in interlayer or layer;<function > label means that needs are provided input parameter, the output parameter requirement of operation by the service that relies on, refType, for the dependence type, has sequential relationship and mandatory relationship.

4. the Web service replacement method based on dependence according to claim 1, it is characterized in that: the described a certain Web service in described step 3, this gone wrong replaces with the described concrete grammar that this meets a Web service of external environment condition or user's needs and is: relatively whether the QoS of Web service meets tenant's requirement of using the described a certain Web service gone wrong, selects best one of meeting property.

5. the Web service replacement method based on dependence according to claim 1, it is characterized in that: the input and output matrix representation of similar Web service in described step 6 is shown in formula (1), io _ijthe similarity that means input and output between Web service i and Web service j, 0<io _ij<1; Computing method are as follows:

The matching degree of the input and output between two Web services calculates by formula (2), equates to refer to the output parameter set out_s of Web service i fully _ibe equal to the input parameter set in_s of Web service j _j, io _ij=1; Comprise the output parameter set out_s that refers to Web service i _ithe input parameter set in_s that belongs to Web service j _j; Part comprises the output parameter set out_s that refers to Web service i _ithe input parameter set in_s that comprises Web service j _j; The output parameter set out_s of unequal finger Web service i _iinput parameter set in_s with Web service j _jwhat do not comprise mutually concerns io _ij=0; For comprising or part while comprising situation, need to calculate similarity io when matching degree _ijsee formula (3), d means two distances between Web service; H means the height of two Web service child Web services common with it; μ, ν represents respectively d and the shared weight of h, 0<μ<1,0<ν<1, μ+ν=1;

$M_{\mathrm{in}\_\mathrm{out}}=\left[\begin{array}{cccc}{\mathrm{io}}_{11}& {\mathrm{io}}_{12}& ..& {\mathrm{io}}_{1n}\\ {\mathrm{io}}_{21}& {\mathrm{io}}_{22}& ..& {\mathrm{io}}_{2n}\\ ..& ..& ..& ..\\ {\mathrm{io}}_{n1}& {\mathrm{io}}_{n2}& ..& {\mathrm{io}}_{\mathrm{nn}}\end{array}\right]---\left(1\right)$

$\mathrm{similarity}=e^{-\mathrm{\&mu;d}}\left(\frac{e^{\mathrm{vh}}-e^{-\mathrm{vh}}}{e^{\mathrm{vh}}+e^{\mathrm{vh}}}\right)---\left(3\right).$

6. the Web service replacement method based on dependence according to claim 1, is characterized in that: constraint matrix M in described step 6 _constraintmean to see formula (4), con _ijthe similarity that means constraint condition between Web service i and Web service j, 0<con _ij<1, computing method are as follows:

Constraint condition matching degree between two Web services calculates by formula (5), equates to refer to the constraint condition set con_s of Web service i fully _ibe equal to the constraint condition set con_s of Web service j _j, con _ij=1; Comprise the constraint condition set con_s that refers to Web service i _ithe constraint condition set con_s that belongs to Web service j _j; Part comprises the constraint condition set con_s that refers to Web service i _ithe constraint condition set con_s that comprises Web service j _j; The constraint condition set con_s of unequal finger Web service i _iconstraint condition set con_s with Web service j _j; What do not comprise mutually concerns con _ij=0; For comprising or part while comprising situation, need to calculate similarity con when matching degree _ijsee formula (6), d means two distances between Web service; H means the height of two Web service child Web services common with it; μ, ν represents respectively d and the shared weight of h;

$M_{\mathrm{constraint}}=\left[\begin{array}{cccc}{\mathrm{con}}_{11}& {\mathrm{con}}_{12}& ..& {\mathrm{con}}_{1n}\\ {\mathrm{con}}_{21}& {\mathrm{con}}_{22}& ..& {\mathrm{con}}_{2n}\\ ..& ..& ..& ..\\ {\mathrm{con}}_{n1}& {\mathrm{con}}_{n2}& ..& {\mathrm{con}}_{\mathrm{nn}}\end{array}\right]---\left(4\right)$

$\mathrm{similarity}=e^{-\mathrm{\&mu;d}}\left(\frac{e^{\mathrm{vh}}-e^{-\mathrm{vh}}}{e^{\mathrm{vh}}+e^{\mathrm{vh}}}\right)---\left(6\right).$

7. the Web service replacement method based on dependence according to claim 1, is characterized in that: service interaction matrix M in described step 6 _correlationvalue be multiplied by dependency degree and obtain by relying on the type weights.

8. the Web service replacement method based on dependence according to claim 1, it is characterized in that: in described step 7, dependence matrix M computing formula is:

$M=\left[\begin{array}{cccc}{d}_{11}& d_{12}& ..& d_{1n}\\ d_{21}& d_{22}& ..& d_{2n}\\ ..& ..& ..& ..\\ d_{n1}& d_{n2}& ..& d_{\mathrm{nn}}\end{array}\right]=M_{\mathrm{in}\_\mathrm{out}}*M_{\mathrm{constraint}}*M_{\mathrm{correlation}}---\left(4\right).$

Images