Service combination verification method and system based on probabilistic model inspection in cloud network convergence environment

Abstract

The invention discloses a service combination verification method and system based on probabilistic model verification in a cloud network convergence environment. According to the method, cloud network user position information and server distribution and service quality files are collected, and a cloud network service sample data set is obtained by fusing the cloud network user position information and the server distribution and service quality files. On the basis, formalized modeling is carried out on a service combination process in cloud network convergence, and meanwhile, changes, caused by user movement, of service attributes in a cloud network convergence environment are considered. User position information is used as historical data, and a Markov process is utilized to establish a user movement model. In order to accurately describe user demands, multi-target formalized description is carried out on the user demands in a cloud network integration scene, and service combination verification in a cloud network integration environment is carried out by utilizing a PRISM model inspection tool. According to the service combination verification method provided by the invention, whether the service combination meets the user requirement can be verified in time, so that the service combination is ensured to still meet the performance and availability requirements.

Description

Technical Field

[0001] The invention relates to a service combination verification method based on probability model verification in a network convergence environment, and belongs to the field of information technology. Background Art

[0002] As service-oriented architecture continues to mature, service composition has become a key concept. Service composition is the process of combining single services into a new service to meet specific needs. In recent years, service composition has not only been applied in traditional environments, but also plays a key role in cloud-network convergence environments. In cloud-network convergence environments, the interactions and dependencies between services become more complex, so service composition verification becomes particularly important. Model-based verification technology can discover problems in the service composition process and evaluate the performance indicators of service composition, thereby assisting the quality control process and reducing losses caused by service composition errors.

[0003] In recent years, a large number of studies have been devoted to the problem of service composition verification. Service composition is modeled, user requirements are described according to formal languages ​​such as CTL and PCTL, and service composition models and user requirements are input into model checkers for verification. These methods have made some progress in service composition verification in traditional environments. Bataineh et al. proposed a model checking method for agent-based service composition methods. In this method, a BPEL model is created to be converted into an automaton model. Although this method considers the dependencies between the various services in the service composition, this method only verifies the correctness of the service composition, and the feasibility of other attribute verification still needs further consideration. Mietal et al. proposed a probability-based model checking to combine Web sub-services using the Markova model. First, the reliability attribute is added to each sub-service in the composition method using the BPEL language. Then, the service selection process is described using the Markova decision process. Finally, the CTL formula is analyzed in the PRISM model checker. Although this method specifically considers the reliability attribute, it is easy to cause the state explosion problem and is not suitable for service composition verification with a large number of services.

[0004] There are problems with the current verification method of services in a cloud-network converged environment. In a cloud-network converged environment, users have mobility characteristics, which may cause users to move from one service scenario to another in a short period of time. This makes the selection probability and attribute value of the previous service invalid during the verification process, which directly increases the difficulty of verification. At the same time, services in a cloud-network converged environment are characterized by large scale. Therefore, it is necessary to take into account the mobility of users and the large number of services in the service combination verification process and verify the service combination. Summary of the invention

[0005] Purpose of the invention: Traditional methods do not take into account the problem of invalid service combination verification results caused by user mobility. In a cloud-network converged environment, the number and types of services requested by users may be higher than in other environments. Therefore, previous verification methods are difficult to directly apply to cloud-network converged environments. In order to solve this problem, the purpose of the present invention is to propose a service combination verification method and system based on probabilistic model verification, which can effectively verify services in a cloud-network converged environment.

[0006] Technical solution: To achieve the above-mentioned invention object, the present invention provides a service combination verification method and system based on probability model verification in a cloud-network fusion environment, comprising the following steps:

[0007] Step 1: Collect cloud network user location information, server distribution dataset, and service quality dataset.

[0008] Step 2: Integrate user location information, server distribution, and service quality datasets to obtain a cloud network service sample dataset.

[0009] Step 3: Model the service composition process in the cloud-network converged environment as a Markov decision process.

[0010] Step 4: Define user requirements as reward structures and transform the requirements into multi-objective temporal logic formulas through PRCTL.

[0011] Step 5: Based on the collected user historical location information, a user mobility model is established using the Markov process.

[0012] Step 6: Put the constructed service composition model, user mobility model and multi-objective temporal logic formula into the PRISM probabilistic model detector to obtain the verification results.

[0013] Preferably, the user location information and server distribution collection in step 1 includes the following steps:

[0014] Step 11: Collect geographic location information including taxi time, longitude and latitude from the mobile trajectory dataset.

[0015] Step 12: Collect the base station geographic location data set, including the base station's geographic location information, access user ID, and access time.

[0016] Step 13: Collect a quality of service dataset containing real-world service information, including response time and throughput.

[0017] Preferably, the data sets are fused to obtain a data set that meets the requirements, including the following steps:

[0018] Step 21: Assign a unique base station number to each base station based on the geographical location of the base station.

[0019] Step 22: associate the service information contained in each service according to the base station number, and add the corresponding service information to the base station information.

[0020] Step 23: Add the corresponding service quality for each service to form a new data set.

[0021] Preferably, step 3 comprises the following steps:

[0022] Step 31: Define cloud network specific services and cloud network abstract services. The cloud network specific services are defined as follows:

[0023] CNCS=<i,n,Q,C>

[0024] Where i represents the service number of the abstract service set, which represents the abstract service identified by i. <q1,q2,q3,...q n > is the QoS set of the current specific service, including multiple QoS attributes. C represents the category to which the current specific service belongs, namely, air, space, land, and sea services.

[0025] The cloud network abstract services are as follows:

[0026]

[0027] in Represents a cloud network specific service, i is the identifier, and n is the sequence number of the current specific service under the abstract service.

[0028] Step 32: Define the cloud network service composition process as follows:

[0029] M= <I,S i , A.S. i , A i , P i , R i , Q>

[0030] Where, is the identifier of the abstract service, which is a positive integer. S is a set of finite states, s i Represents the state set when i∈I. AS i Represents the concrete services contained in the current abstract service. i is a set of actions, i.e., the action of selecting a specific service from an abstract service. P is the transition probability describing the Markov decision process, P: S×A×S→[0,1], P(s′|s,a) represents the probability of transitioning from the current state s to the state s′ through action a, i.e., the probability of service selection. R is the reward of the action, R: S×A×S, where R is a real number, representing the reward value obtained by calling action a in state s. Q is the QoS value.

[0031] Step 33: After the service combination is modeled, the corresponding service QoS values ​​in the cloud network service sample data set are filled into the model.

[0032] Preferably, step 4 comprises the following:

[0033] Step 41: Define the Probabilistic Reward Tree Logic (PRCTL), which is divided into syntax and semantics. The syntax is defined as follows:

[0034] Φ=::P ＝？ [φ]|C ＝？ [FA]

[0035] φ=::XA|AUA|AU n A

[0036]

[0037] Among them, T represents true, ⊥ represents false, and they are all PRCTL formulas. When φ is a PRCTL formula, φ∧φ, φ→φ, Xφ, etc. are all PRCTL formulas. X represents the next state, U represents until a certain state, F represents a certain future state, and is a tense connective. P is the probability value, and C is the reward value.

[0038] The semantics are defined as follows:

[0039]

[0040] in, It is a service composition model about PRCTL. Indicated in the model The path π satisfies the PRCTL formula φ.

[0041] Step 42: Describe the service combination verification result through reward output and probability output, which are defined as follows:

[0042]

[0043] in, is the PRCTL formula, represents the service composition model, P represents the output probability, and C represents the output reward.

[0044] Preferably, step 5 comprises the following:

[0045] Step 51: Define the user mobility model through the Markov process, which is defined as follows:

[0046] M=<S,A,P>

[0047] Where S is the set of base stations. A is the set of actions, representing the base station change actions provided by the user's movement. P is the transition probability describing the Markov decision process, representing the probability of a user moving from one base station to another.

[0048] Step 52: Extract user location information through the cloud network service sample data set and construct the above Markov model.

[0049] Preferably, step 6 comprises the following steps:

[0050] Step 61: The cloud-network service combination model and the user mobility model are integrated to obtain the cloud-network service combination model under user mobility, and converted into PRISM language for description.

[0051] Step 62: Convert the multi-objective temporal logic formula into PRISM language for model checking.

[0052] Based on the same inventive concept, the present invention provides a service combination verification system based on probability model checking in a cloud-network fusion environment, comprising:

[0053] Data collection module: used to collect cloud network user location information, server distribution files and service quality files.

[0054] Data fusion module: used to fuse user location information, server distribution and service quality data sets to obtain cloud network service sample data sets.

[0055] Combination modeling module: used to use cloud-network service information as data to build a service combination model in a cloud-network converged environment.

[0056] Mobility modeling module: It is used to use user location information as data and build a user mobility model using the Markov model.

[0057] Demand conversion module: used to convert user needs into corresponding temporal logic formulas and formally describe user needs.

[0058] Model verification module: used to verify the service combination results according to the service combination model, mobility model and target temporal logic formula to determine whether they meet user needs.

[0059] Based on the same inventive concept, the present invention provides a computer system, including a memory, a processor, and a computer program / instructions stored in the memory and executable on the processor. When the computer program / instructions are executed by the processor, the steps of the service combination verification method based on probabilistic model verification in a cloud-network fusion environment are implemented.

[0060] Based on the same inventive concept, the present invention provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of a service combination verification method based on probabilistic model verification in a cloud-network fusion environment.

[0061] Beneficial effects: Compared with the existing technology, the present invention has the following advantages: The present invention fully considers the mobility of users, models the user mobility as a Markov process, and applies the generated mobility model to the verification process of the service combination in the cloud-network fusion environment, solving the problem of service verification failure caused by user mobility. Considering the large number of services in the cloud-network fusion environment, the present invention adopts the probability model verification method to verify the service combination, which can quickly obtain the verification result of the service combination and improve the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 The figure is a flow chart of a method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0063] The present invention is further explained below in conjunction with the accompanying drawings and specific embodiments. It should be understood that these embodiments are only used to illustrate the present invention and are not used to limit the scope of the present invention. After reading the present invention, various equivalent forms of modifications to the present invention by those skilled in the art all fall within the scope defined by the claims attached to this application.

[0064] like Figure 1 As shown, the embodiment of the present invention discloses a service combination verification method based on probability model verification in a cloud-network fusion environment, which mainly includes the following steps:

[0065] Step 1: Collect cloud network user location information, server distribution dataset, and service quality dataset.

[0066] Step 2: Integrate user location information, server distribution, and service quality datasets to obtain a cloud network service sample dataset.

[0067] Step 3: Model the service composition process in the cloud-network converged environment as a Markov decision process.

[0068] Step 4: Define user requirements as reward structures and transform the requirements into multi-objective temporal logic formulas through PRCTL.

[0069] Step 5: Based on the collected user historical location information, a user mobility model is established using the Markov process.

[0070] Step 6: Put the constructed service composition model, user mobility model and multi-objective temporal logic formula into the PRISM probabilistic model detector to obtain the verification results.

[0071] The embodiment of the present invention provides a service combination verification method based on probability model verification in a cloud-network fusion environment. The method collects GPS trajectory data sets, server distribution information, and service quality data sets, and integrates the obtained data sets according to certain rules to obtain a service sample data set that meets the cloud-network fusion environment. The data collected in step 1 includes three aspects: a trajectory data set containing geographical location information of taxi time, longitude and latitude; a network base station data set containing network base station latitude and longitude information and time slice information; service quality information; the specific steps are as follows:

[0072] Step 11: Collect geographic location information including taxi time, longitude and latitude from the mobile trajectory dataset.

[0073] Step 12: Collect the base station geographic location data set, including the base station's geographic location information, access user ID, and access time.

[0074] Step 13: Collect a quality of service dataset containing real-world service information, including response time and throughput.

[0075] Based on the above-obtained data set, in step 2, a cloud network service sample data set is obtained by integrating user location information and server distribution. This step specifically includes:

[0076] Step 21: Assign a unique base station number to each base station based on the geographical location of the base station.

[0077] Step 22: associate the service information contained in each service according to the base station number, and add the corresponding service information to the base station information.

[0078] Step 23: Add the corresponding service quality for each service to form a new data set.

[0079] Step 3: Considering the characteristics of services in the cloud-network fusion environment, define cloud-network specific services and cloud-network abstract services respectively, and define the cloud-network service composition process. Use the service request response process in the cloud-network fusion environment to illustrate the method steps, including proxy services, cache services, application services, file services, mail services, space-based data services, ground-based data services and other abstract services. Each abstract service contains different numbers of services of the same type. The service composition process is the process of users calling these services.

[0080] Step 31: Define cloud network specific services and cloud network abstract services. The cloud network specific services are defined as follows:

[0081] CNCS=<i,n,Q,C>

[0082] Where i represents the service number of the abstract service set, which represents the abstract service identified by i. <q1,q2,q3,...q n > is the QoS set of the current specific service, including multiple QoS attributes. C represents the category to which the current specific service belongs, namely, air, space, land, and sea services.

[0083] The cloud network abstract services are as follows:

[0084]

[0085] in Represents a cloud network specific service, i is the identifier, and n is the sequence number of the current specific service under the abstract service.

[0086] Step 32: Define the cloud network service composition process as follows:

[0087] M= <I,S i , A.S. i , A i , P i , R i , Q>

[0088] Where, is the identifier of the abstract service, which is a positive integer. S is a set of finite states, s i Represents the state set when i∈I. AS i Represents the concrete services contained in the current abstract service. i is a set of actions, i.e., the action of selecting a specific service from an abstract service. P is the transition probability describing the Markov decision process, P: S×A×S→[0,1], P(s′|s,a) represents the probability of transitioning from the current state s to the state s′ through action a, i.e., the probability of service selection. R is the reward of the action, R: S×A×S, where R is a real number, representing the reward value obtained by calling action a in state s. Q is the QoS value.

[0089] Step 33: After the service combination is modeled, the corresponding service QoS values ​​in the cloud network service sample data set are filled into the model.

[0090] Step 4: Define user demand as a reward structure, and transform the demand into a multi-objective temporal logic formula through PRCTL. In this model, the probability of service selection is predetermined. Data from the cloud network service sample data set is selected as the response time and throughput of the service, and the response time and throughput are used as the reward structure to illustrate the model. It includes the following steps:

[0091] Step 41: Define the Probabilistic Reward Tree Logic (PRCTL), which is divided into syntax and semantics. The syntax is defined as follows:

[0092] Φ=∷P＝？ [φ]|C ＝？ [FA]

[0093] φ=∷XA|AUA|AU n A

[0094]

[0095] Among them, T represents true, ⊥ represents false, and they are all PRCTL formulas. When φ is a PRCTL formula, φ∧φ, φ→φ, Xφ, etc. are all PRCTL formulas. X represents the next state, U represents until a certain state, F represents a certain future state, and is a tense connective. P is the probability value, and C is the reward value.

[0096] The semantics are defined as follows:

[0097]

[0098] in, It is a service composition model about PRCTL. Indicated in the model The path π satisfies the PRCTL formula φ.

[0099] Step 42: Describe the service combination verification result through reward output and probability output, which are defined as follows:

[0100]

[0101] in, is the PRCTL formula, represents the service composition model, P represents the output probability, and C represents the output reward.

[0102] Preferably, step 5 comprises the following:

[0103] Step 43: According to step 41, determine the task model: Where N is the total number of tasks, n is the index of the task, S is the offloading decision result of the task, and I n ,O n ,P n They represent the input data size, instruction number size and task type of the task respectively.

[0104] Step 5: Based on the collected historical location information of users, a user mobility model is established using the Markov process, including the following steps:

[0105] Step 51: Define the user mobility model through the Markov process, which is defined as follows:

[0106] M=<S,A,P>

[0107] Where S is the set of base stations. A is the set of actions, representing the base station change actions provided by the user's movement. P is the transition probability describing the Markov decision process, representing the probability of a user moving from one base station to another.

[0108] Step 52: Extract user location information by selecting a cloud network service sample data set and construct the above Markov model.

[0109] Step 6 puts the constructed service composition model, user mobility model and multi-objective temporal logic formula into the PRISM probabilistic model detector to obtain the verification result.

[0110] Step 61: The cloud-network service composition model and the user mobility model are integrated to obtain the cloud-network service composition model under user mobility, and converted into PRISM for description.

[0111] Step 62: Convert the multi-objective temporal logic formula into PRISM verification language for model checking.

[0112] Table 1 shows the final experimental verification results, which show the probability of deadlock in the composite service, the probability of accessing only cached services, the probability of responding within 10 steps, and the total response time and throughput. In order to further illustrate the effectiveness of the method of the present invention, a set of experiments were conducted on the method of the present invention to prove the effectiveness of the service combination verification results under the condition of large-scale service quantity. The experimental results are shown in Table 2. The changes in model building time and model checking time as the number of services increases. From the vertical perspective of the table, as the number of services increases, the model building time also increases, and the model checking time also increases. From the horizontal perspective of the table, under the condition of the same number of services, the model checking time is much less than the model building time. In general, when the number of services is large, from the perspective of verification time cost, the experiments show that the service combination verification method adopted by the present invention is feasible and effective.

[0113] Table 1 Verification results

[0114]

[0115] Table 2 Verification time consumption under different service numbers

[0116]

[0117] Based on the same inventive concept, the present invention provides a service combination verification system based on probability model checking in a cloud-network fusion environment, comprising:

[0118] Data collection module: used to collect cloud network user location information, server distribution files and service quality files.

[0119] Data fusion module: used to fuse user location information, server distribution and service quality data sets to obtain cloud network service sample data sets.

[0120] Combination modeling module: used to use cloud-network service information as data to build a service combination model in a cloud-network converged environment.

[0121] Mobility modeling module: It is used to use user location information as data and build a user mobility model using the Markov model.

[0122] Demand conversion module: used to convert user needs into corresponding temporal logic formulas and formally describe user needs.

[0123] Model verification module: used to verify the service combination results according to the service combination model, mobility model and target temporal logic formula to determine whether they meet user needs.

[0124] The specific working process of each module described above can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here. The module division is only a logical function division, and there may be other division methods in actual implementation, such as multiple modules can be combined or integrated into another system.

[0125] Based on the same inventive concept, an embodiment of the present invention provides a computer system, including a memory, a processor, and a computer program / instructions stored in the memory and executable on the processor. When the computer program / instructions are executed by the processor, the steps of a service combination verification method based on probabilistic model verification in a cloud-network fusion environment are implemented.

[0126] Based on the same inventive concept, an embodiment of the present invention provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of a service combination verification method based on probabilistic model verification in a cloud-network fusion environment.

[0127] Those skilled in the art can understand that the technical solution of the present invention, in essence or in other words, the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for enabling a computer system (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in the embodiment of the present invention. The storage medium includes: various media that can store computer programs, such as a USB flash drive, a mobile hard disk, a read-only memory ROM, a random access memory RAM, a magnetic disk or an optical disk.

Claims

1. A service combination verification method based on probability model verification in a cloud-network fusion environment, characterized in that: The steps include: Step 1: Collect cloud network user location information, server distribution dataset, and service quality dataset. Step 2: Integrate user location information, server distribution, and service quality datasets to obtain a cloud network service sample dataset. Step 3: Model the service composition process in the cloud-network converged environment as a Markov decision process. Step 4: Define user requirements as reward structures and transform the requirements into multi-objective temporal logic formulas through PRCTL. Step 5: Based on the collected user historical location information, a user mobility model is established using the Markov process. Step 6: Put the constructed service composition model, user mobility model and multi-objective temporal logic formula into the PRISM probabilistic model detector to obtain the verification results.

2. The service combination verification method based on probability model verification in a cloud-network fusion environment according to claim 1 is characterized in that: The user location information, server distribution collection, and service quality data set in step 1 specifically include: Step 11: Collect geographic location information including taxi time, longitude and latitude from the mobile trajectory dataset. Step 12: Collect the base station geographic location data set, including the base station's geographic location information, access user ID, and access time. Step 13: Collect a quality of service dataset containing real-world service information, including response time and throughput.

3. The service combination verification method based on probability model verification in a cloud-network fusion environment according to claim 1 is characterized in that: In step 2, the data sets are fused to obtain a data set that meets the requirements, including the following steps: Step 21: Assign a unique base station number to each base station based on the geographical location of the base station. Step 22: associate the service information contained in each service according to the base station number, and add the corresponding service information to the base station information. Step 23: Add the corresponding service quality for each service to form a new data set.

4. The service combination verification method based on probability model verification in a cloud-network fusion environment according to claim 3 is characterized in that: The step 3 comprises the following steps: Step 31: Define cloud network specific services and cloud network abstract services. The cloud network specific services are defined as follows: CNCS=<i,n,Q,C> Where i represents the service number of the abstract service set, which represents the abstract service identified by i. <q1,q2,q3,...q n > is the QoS set of the current specific service, including multiple QoS attributes. C represents the category to which the current specific service belongs, namely, air, space, land, and sea services. The cloud network abstract services are as follows: in Represents a cloud network specific service, i is the identifier, and n is the sequence number of the current specific service under the abstract service. Step 32: Define the cloud network service composition process as follows: M=<I,S i ,AS i ,A i ,P i ,R i ,Q> Where, is the identifier of the abstract service, which is a positive integer. S is a set of finite states, S i Represents the state set when i∈I. AS i Represents the concrete services contained in the current abstract service. i is a set of actions, i.e., the action of selecting a specific service from an abstract service. P is the transition probability describing the Markov decision process, P: S×A×S→[0,1], P(s′|s,a) represents the probability of transitioning from the current state s to the state s′ through action a, i.e., the probability of service selection. R is the reward of the action, R: S×A×S, where R is a real number, representing the reward value obtained by calling action a in state s. Q is the QoS value. Step 33: After the service combination is modeled, the corresponding service QoS values ​​in the cloud network service sample data set are filled into the model.

5. The service combination verification method based on probability model verification in a cloud-network fusion environment according to claim 1 is characterized in that: The step 4 includes the following: Step 41: Define the Probabilistic Reward Tree Logic (PRCTL), which is divided into syntax and semantics. The syntax is defined as follows: Φ=::P ＝？ [φ]|C ＝？ [FA] φ=::XA|AUA|AU n A Among them, T represents true, ⊥ represents false, and they are all PRCTL formulas. When φ is a PRCTL formula, φ∧φ, φ→φ, Xφ, etc. are all PRCTL formulas. X represents the next state, U represents until a certain state, F represents a certain future state, and is a tense connective. P is the probability value, and C is the reward value. The semantics are defined as follows: in, It is a service composition model about PRCTL. Indicated in the model The path π satisfies the PRCTL formula φ. Step 42: Describe the service combination verification result through reward output and probability output, which are defined as follows: in, is the PRCTL formula, represents the service composition model, P represents the output probability, and C represents the output reward.

6. The service combination verification method based on probability model verification in a cloud-network fusion environment according to claim 1 is characterized in that: The step 5 includes the following: Step 51: Define the user mobility model through the Markov process, which is defined as follows: M=<S,A,P> Where S is the set of base stations. A is the set of actions, representing the base station change actions provided by the user's movement. P is the transition probability describing the Markov decision process, representing the probability of a user moving from one base station to another. Step 52: Extract user location information through the cloud network service sample data set and construct the above Markov model.

7. The service combination verification method based on probability model verification in a cloud-network fusion environment according to claim 1 is characterized in that: The step 6 includes the following: Step 61: The cloud-network service combination model and the user mobility model are integrated to obtain the cloud-network service combination model under user mobility, and converted into PRISM language for description. Step 62: Convert the multi-objective temporal logic formula into PRISM language for model checking.

8. A service combination verification system based on probability model verification in a cloud-network fusion environment, characterized in that: include: Data collection module: used to collect cloud network user location information, server distribution files and service quality files. Data fusion module: used to fuse user location information, server distribution and service quality data sets to obtain cloud network service sample data sets. Combination modeling module: used to use cloud-network service information as data to build a service combination model in a cloud-network converged environment. Mobility modeling module: It is used to use user location information as data and build a user mobility model using the Markov model. Demand conversion module: used to convert user needs into corresponding temporal logic formulas and formally describe user needs. Model verification module: used to verify the service combination results according to the service combination model, mobility model and target temporal logic formula to determine whether they meet user needs.

9. A computer system comprising a memory, a processor, and a computer program / instruction stored in the memory and executable on the processor, characterized in that: When the computer program / instructions are executed by the processor, the steps of a service combination verification method based on probabilistic model verification in a cloud-network fusion environment are implemented according to any one of claims 1-7.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by the processor, the steps of a service combination verification method based on probabilistic model verification in a cloud-network fusion environment are implemented according to any one of claims 1-7.

Images