WO2021068581A1 - Automatic service composition method based on critical path spanning tree - Google Patents
Automatic service composition method based on critical path spanning tree Download PDFInfo
- Publication number
- WO2021068581A1 WO2021068581A1 PCT/CN2020/101390 CN2020101390W WO2021068581A1 WO 2021068581 A1 WO2021068581 A1 WO 2021068581A1 CN 2020101390 W CN2020101390 W CN 2020101390W WO 2021068581 A1 WO2021068581 A1 WO 2021068581A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- service
- qos
- response time
- critical path
- services
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F8/00—Arrangements for software engineering
- G06F8/30—Creation or generation of source code
Abstract
An automatic service composition method based on a critical path spanning tree. A service QoS attribute is added as a service measurement basis in service selection. In order to complete a complex function request, a corresponding OWL-S description document is first generated from a WSDL document of a service; an QoS attribute value is added to make full use of the advantages of a semantic ontology in service matching; then, a search algorithm WSACA based on a breadth-first algorithm is proposed to search a composition scheme by taking an optimal global QoS attribute value as a target, and an OWL-S description document of a composite service is generated for service re-publishing and calling. The method avoids tedious manual composition, proposes a critical path spanning tree concept for generation of a composite service document, simplifies generation of a composite service control structure, and creates conditions for re-publishing of the composite service.
Description
本发明涉及一种Web服务组合方法，一种自动服务组合方法，属于互联网服务技术领域。The invention relates to a Web service composition method, an automatic service composition method, and belongs to the technical field of Internet services.
Web服务概念以及面向服务体系结构(SOA)的提出，使得服务的提供者可以将自己的软件以服务形式提供给用户。Web服务的接口是采用中立的方式进行定义的，独立于实现服务的平台、操作系统和编程语言，这使得各种服务能够以一种统一的通用方式进行交互。由于Web服务的封装性、松耦合和跨平台性的优点，基于Web服务的应用越来越广泛。而随着Web服务标准的完善与支持服务的软件平台不断成熟，越来越多的企业将其业务功能和流程包装成标准的Web服务发布出去，比如Amazon平台上面的图书查询服务。The concept of Web services and the service-oriented architecture (SOA) enable service providers to provide their software to users in the form of services. The interface of Web service is defined in a neutral way, independent of the platform, operating system and programming language that implement the service, which enables various services to interact in a unified and common way. Due to the encapsulation, loose coupling, and cross-platform advantages of Web services, the applications based on Web services are becoming more and more widespread. With the improvement of Web service standards and the continuous maturity of software platforms that support services, more and more companies package their business functions and processes into standard Web services, such as the book query service on the Amazon platform.
随着Web服务越来越多，有许多的Web服务功能其实类似，这就使我们在区分这些功能类似的Web服务时遇到困难。而QoS(Quality of Service)的提出较好的解决了这个问题，也就是Web服务的实现、运行平台以及Web服务器的差异使得不同服务的QoS例如响应时间，运行代价，吞吐量有所不同。With more and more Web services, there are many Web service functions that are actually similar, which makes it difficult for us to distinguish these Web services with similar functions. The proposal of QoS (Quality of Service) better solves this problem, that is, the differences in the implementation of web services, operating platforms, and web servers make the QoS of different services such as response time, operating cost, and throughput different.
由于网络上的服务越来越多，用户请求实现的功能日益复杂，单个服务已经不能满足复杂的客户需求，同时人工组合与选择服务也变得不现实。为了提高网络共享的Web服务利用率，降低开发新服务的成本，需要将多个功能有限的简单Web服务按照服务描述、服务约束、可用资源等进行服务选择和组合，实现用户对复杂服务功能请求的目标，从而产生增值服务，这就是服务的组合。As there are more and more services on the network, the functions requested by users are becoming more and more complex. A single service can no longer meet the needs of complex customers. At the same time, manual combination and selection of services have become unrealistic. In order to improve the utilization of web services shared by the network and reduce the cost of developing new services, it is necessary to select and combine multiple simple web services with limited functions according to service descriptions, service constraints, and available resources, so as to realize user requests for complex service functions. The goal of creating value-added services, which is the combination of services.
服务组合被广泛用来改进企业软件系统的敏捷性、灵活性和可用性，在新的业务需求驱动的促使下，以开放的Web服务方式进行的、由服务质量QoS驱动的服务组合与选择已经成为Web服务领域的研究重点。Service composition is widely used to improve the agility, flexibility and availability of enterprise software systems. Driven by new business requirements, service composition and selection driven by open Web services and QoS have become Research focus in the field of Web services.
发明内容Summary of the invention
本发明的目的是提供一种基于关键路径生成树的自动服务组合方法，实现更高的组合性能，并加强灵活度，为组合服务的再发布创造条件。The purpose of the present invention is to provide an automatic service composition method based on a critical path spanning tree, to achieve higher composition performance, strengthen flexibility, and create conditions for the republishing of composite services.
本发明的目的是这样实现的：一种基于关键路径生成树的自动服务组合方法，包括以下步骤：The purpose of the present invention is achieved as follows: an automatic service composition method based on a critical path spanning tree, including the following steps:
步骤1)计算组合服务的QoS模型；Step 1) Calculate the QoS model of the combined service;
步骤2)构建服务组合与选择的问题模型；Step 2) Construct a problem model of service composition and selection;
步骤3)创建服务自动组合与选择算法WSACA；Step 3) Create automatic service combination and selection algorithm WSACA;
步骤4)构设服务组合关键路径生成树算法。Step 4) Construct a spanning tree algorithm for the critical path of service composition.
作为本发明的进一步限定，步骤1)具体步骤为：As a further limitation of the present invention, the specific steps of step 1) are:
步骤1-1)提出响应时间的全局QoS计算方法；Step 1-1) Propose a global QoS calculation method for response time;
步骤1-2)通过对可靠性与可得性的取对数与归一化操作使得其全局QoS计算方法与响应时间一样；Step 1-2) Through the logarithm and normalization of reliability and availability, the global QoS calculation method is the same as the response time;
步骤1-3)通过用户权重分配从而完成各项属性QoS值的聚合。Step 1-3) Through user weight distribution, the aggregation of the QoS values of various attributes is completed.
作为本发明的进一步限定，步骤1-1)中的全局响应时间计算公式如下表所示：As a further limitation of the present invention, the global response time calculation formula in step 1-1) is shown in the following table:
其中，服务与服务前后关系有三种：序列(Sequence)，分支(Split)和汇合(Join)，假设getLocQos(S)为得到服务S的本地响应时间(即服务的单独调用响应时间)，而getGlbQos(S)为获取服务S的全局响应时间(在组合服务过程中运行完此服务所需的响应时间)，glbQos(S)表示得到服务S的全局响应时间为qos，并能改变它的值；Among them, there are three types of service and service context: Sequence, Split and Join. Assume that getLocQos(S) is the local response time of service S (that is, the response time of a single service call), and getGlbQos (S) To obtain the global response time of service S (the response time required to run this service in the process of combining services), glbQos(S) indicates that the global response time of service S is qos, and its value can be changed;
步骤1-2)中的响应时间的归一化公式：The normalized formula of response time in step 1-2):
其中qos*为服务的原始响应时间值(真实响应时间值)，而qos
max为所有服务当中的最大响应时间，qos
min为所有服务当中的最小响应时间，qos为归一化之后的响应时间。
Where qos* is the original response time value of the service (real response time value), and qos max is the maximum response time among all services, qos min is the minimum response time among all services, and qos is the normalized response time.
作为本发明的进一步限定，步骤2)具体步骤为：As a further limitation of the present invention, the specific steps of step 2) are:
步骤2-1)服务匹配，基于服务输入参数集、输出参数集与QoS属性值给出两个服务匹配的判断规则，用以辅助服务组合与选择问题模型的构建；Step 2-1) Service matching, based on the service input parameter set, output parameter set and QoS attribute value, the judgment rule of two service matching is given to assist the construction of service combination and selection problem model;
步骤2-2)服务组合与选择问题模型，在给定服务集内寻找满足服务请求的服务组合方案问题定义为寻找一个特定目标服务集，且一旦该服务集确定，根据服务之间的匹配关系，服务集组合调用流程图也相应确定。Step 2-2) Service composition and selection problem model, find a service composition solution that satisfies the service request in a given service set. The problem is defined as finding a specific target service set, and once the service set is determined, according to the matching relationship between services , The service set combination call flow chart is also determined accordingly.
作为本发明的进一步限定，步骤3)具体步骤为：As a further limitation of the present invention, the specific steps of step 3) are:
步骤3-1)建立算法当中所用到的数据结构和输出的哈希表；Step 3-1) Establish the data structure and output hash table used in the algorithm;
步骤3-2)构设服务自动组合选择算法WSACA，利用现有的服务输入查找能够调用的服务列表，然后将可调用服务的输出加入可得输入之中，更新服务的全局QoS及RefTable哈希表，并进行下一层次的搜索，具体算法描述如下：Step 3-2) Construct the service automatic combination selection algorithm WSACA, use the existing service input to find the list of services that can be called, and then add the output of the callable service to the available input, and update the global QoS and RefTable hash of the service Table, and perform the next level of search, the specific algorithm is described as follows:
由上面描述的算法WSACA能够在用户给定输入并请求特定输出后，根据用户的功能请求与具有全局最优QoS值的目标驱动下自动寻找组合服务方案。The algorithm WSACA described above can automatically find a combined service plan based on the user's function request and the goal with the global optimal QoS value after the user has given input and requested a specific output.
作为本发明的进一步限定，步骤4)的具体步骤为：As a further limitation of the present invention, the specific steps of step 4) are:
提出服务流程图的关键路径生成树算法，选择步骤3中搜索出的服务用以组合；在服务中添加了一个输入关联服务哈希表，即存储了每个服务的输入与提供此输入数据对应的服务输出，从而实现服务间数据关联绑定；随后采用关键路径树生成算法，同时生成组合服务的OWL-S文件以方便调用或者发布；具体生成关键路径树的算法如下：The key path spanning tree algorithm of the service flowchart is proposed, and the services searched in step 3 are selected for combination; an input correlation service hash table is added to the service, that is, the input of each service is stored corresponding to the input data provided To realize the data association and binding between services; then use the critical path tree generation algorithm, and generate the OWL-S file of the combined service to facilitate the call or release; the specific algorithm for generating the critical path tree is as follows:
由于响应时间的限制，在构造了关键路径生成树之后产生的控制流程能保证服务被调用时与之有数据相关的服务输出都已经具备了，这就是构造关键路径生成树的必要性。在服务中添加了一个输入关联服务哈希表，即存储了每个服务的输入与提供此输入数据对应的服务输出，从而实现服务间数据关联绑定。Due to the limitation of response time, the control flow generated after the critical path spanning tree is constructed can ensure that the service output related to the data is available when the service is called. This is the necessity of constructing the critical path spanning tree. An input correlation service hash table is added to the service, that is, the input of each service and the service output corresponding to the input data are stored, so as to realize the data association and binding between services.
本发明采用以上技术方案与现有技术相比，具有以下技术效果：本发明提出了一种拥有较优性能的服务自动组合算法，避免了手工组合的繁琐；为组合服务文档生成提出了关键路径生成树概念，简化组合服务控制结构的生成，为组合服务的再发布创造条件。Compared with the prior art, the present invention adopts the above technical solutions and has the following technical effects: the present invention proposes an automatic service composition algorithm with better performance, avoids the cumbersome manual composition; and proposes a critical path for the generation of combined service documents The concept of spanning tree simplifies the generation of composite service control structure and creates conditions for the re-release of composite services.
图1为本发明工作流程图。Figure 1 is a working flow chart of the present invention.
图2为本发明中服务组合控制模式图。Figure 2 is a diagram of the service composition control mode in the present invention.
图3为本发明中WSDL与OWL-S的映射关系。Figure 3 shows the mapping relationship between WSDL and OWL-S in the present invention.
图4为本发明中样例服务组合流程图。Figure 4 is a flow chart of sample service composition in the present invention.
图5为本发明中样例组合服务关键路径生成树。Figure 5 shows the critical path spanning tree of the sample composite service in the present invention.
图6为本发明中样例组合服务控制结构。Figure 6 is a sample composite service control structure in the present invention.
下面结合附图对本发明的技术方案做进一步的详细说明：The technical scheme of the present invention will be further described in detail below in conjunction with the accompanying drawings:
一种基于关键路径生成树的自动服务组合方法，包含如下步骤：An automatic service composition method based on critical path spanning tree, including the following steps:
1)计算组合服务的QoS模型；1) Calculate the QoS model of the combined service;
2)生成OWL-S文档；2) Generate OWL-S documents;
3)构建服务组合与选择问题模型；3) Build a service portfolio and selection problem model;
4)创设服务自动组合与选择算法WSACA；4) Create automatic service combination and selection algorithm WSACA;
5)构建组合服务控制流程。5) Build a control process for composite services.
步骤1的具体步骤为：The specific steps of step 1 are:
步骤1.1)在本发明的QoS计算模型中，服务与服务前后关系有三种，分别是：序列 (Sequence)，分支(Split)和汇合(Join)；控制模式三种情况介绍如下图1所示，以响应时间(Response Time)为例进行全局QoS计算；Step 1.1) In the QoS calculation model of the present invention, there are three types of service and service context, namely: Sequence, Split and Join; the three control modes are introduced as shown in Figure 1. Take Response Time as an example for global QoS calculation;
假设getLocQos(S)为得到服务S的本地响应时间(即本服务的单独调用响应时间)，而getGlbQos(S)为获取服务S的全局响应时间(在组合服务过程中运行完此服务所需的响应时间)，glbQos(S)表示得到服务S的全局响应时间为qos，并能改变它的值。由上图可知三种控制模式中下一层服务的全局响应时间计算公式如下表所示：Assume that getLocQos(S) is to obtain the local response time of service S (that is, the response time of this service alone), and getGlbQos(S) is to obtain the global response time of service S (the service required to run this service in the combined service process) Response time), glbQos(S) indicates that the global response time of service S is qos, and its value can be changed. From the above figure, it can be seen that the global response time calculation formula of the next layer of the three control modes is shown in the following table:
表1控制模式响应时间计算公式Table 1 Calculation formula of control mode response time
由上面控制模式的响应时间计算公式可知，如果控制模式是Sequence的话就是简单的响应时间值累加，因为服务都是顺序执行的。而如果一个服务的输入必须由多个服务的输出提供时即Join模式，则可将其看成多个独立的Sequence取其中具有最大延迟时间的Sequence作为关键延迟即这服务的全局相应时间，因为多个独立的Sequence之间的服务是可以并行运行的。而类似的Split结构其实就可以直接看成是互不干扰的Sequence，故而计算也很类似下一层的每个服务互不影响，只与上一层服务与本服务的响应时间有关。由此可看出整个组合服务流程的响应时间只与其中的几条关键服务路径的响应时间有关，故而，利用这一原理，后文会介绍如何据此生成相应的服务调用过程，以达到充分利用生成服务方案的全局QoS属性。From the response time calculation formula of the above control mode, it can be seen that if the control mode is Sequence, the response time value is simply accumulated, because the services are executed sequentially. And if the input of a service must be provided by the output of multiple services, that is, the Join mode, it can be regarded as multiple independent Sequences, and the sequence with the largest delay time is used as the key delay, which is the global corresponding time of the service, because Services between multiple independent Sequences can run in parallel. The similar Split structure can actually be directly regarded as a sequence that does not interfere with each other, so the calculation is also very similar to that each service of the next layer does not affect each other, and is only related to the response time of the upper layer service and this service. It can be seen that the response time of the entire composite service process is only related to the response time of several key service paths. Therefore, using this principle, the following article will introduce how to generate the corresponding service invocation process accordingly to achieve sufficient Utilize the global QoS attributes of the generated service plan.
步骤1.2)由于服务的QoS属性之间单位各不相同，而且取值范围也不一样，未免在QoS属性聚合时产生某一属性值占主导优势而直接决定聚合后的QoS值，故本发明对服务的各种属性进行归一化，即将各种QoS属性值映射到区间[0,1]范围，而且要保证每个QoS属性全局属性值的计算方法都与上述响应时间的计算方法一致，以保证服务组合的准确性；Step 1.2) Since the units of the QoS attributes of the service are different, and the value ranges are also different, it is inevitable that a certain attribute value will be dominant when the QoS attributes are aggregated and directly determine the aggregated QoS value. The various attributes of the service are normalized, that is, the various QoS attribute values are mapped to the range [0,1], and the calculation method of the global attribute value of each QoS attribute must be consistent with the calculation method of the above response time. Ensure the accuracy of service portfolio;
1.2.1)首先是响应时间的归一化公式：1.2.1) First, the normalized formula of response time:
其中qos*为服务的原始响应时间值(真实响应时间值)，而qosmax为所有服务当中的最大响应时间，qosmin为所有服务当中的最小响应时间，qos为归一化之后的响应时间。Where qos* is the original response time value of the service (real response time value), and qosmax is the maximum response time among all services, qosmin is the minimum response time among all services, and qos is the normalized response time.
1.2.2)其次是可靠性与可得性的归一化，由于二者都是比例，取值为[0,1]，但是在序列控制结构情况下的全局QoS属性值的计算公式不是相加，而是相乘，而且组合服务的可靠性或可得性越大越好(与响应时间越小越好不一致)，故而要使二者的组合服务值计算 拥有与组合服务响应时间一样的计算方法必须加以变化：1.2.2) The second is the normalization of reliability and availability. Since both are proportions, the value is [0,1], but the calculation formula of the global QoS attribute value in the case of the sequence control structure is not the same. Add, but multiply, and the greater the reliability or availability of the combined service, the better (the smaller the response time, the better), so the combined service value calculation of the two should have the same calculation as the response time of the combined service The method must be changed:
1.2.2.1)首先对原始值取对数，假设可靠性或者可得性原始值为1≧a>0，这样使得序列中组合服务的可靠性与可得性计算也演化为加法运算：1.2.2.1) First, take the logarithm of the original value, assuming that the original value of reliability or availability is 1≧a>0, so that the reliability and availability calculations of combined services in the sequence also evolve into addition operations:
qos
*＝ln(a)
qos * =ln(a)
1.2.2.2)其次取对数之后再进行归一化，同样规定qos
max为所有最大可靠性或可得性值，所不同的是可靠性与可得性的初始值具有理论最大值1，所以可以设定qos
max＝ln(1)＝0，对于qos
min则为全部服务可靠性或可得性的最小值取对数(值为0的不予考虑，不参加组合)，故其归一化操作为：
1.2.2.2) Secondly, normalization is performed after taking the logarithm. The qos max is also specified as all the maximum reliability or availability values. The difference is that the initial values of reliability and availability have the theoretical maximum value of 1, so You can set qos max =ln(1)=0. For qos min , the minimum value of all service reliability or availability is taken as the logarithm (the value 0 will not be considered and will not participate in the combination), so it is normalized The operation is:
此时可发现，服务的可靠性与可得性归一化后所得值不仅具有越小越优的性质，且其全局QoS的计算与响应时间的原理一样，故而简化了组合服务的QoS值计算；At this point, it can be found that the value obtained after normalization of the reliability and availability of the service not only has the property of the smaller the better, and the calculation of its global QoS is the same as the principle of response time, which simplifies the calculation of the QoS value of the combined service ；
1.2.3)完成各种属性的归一化操作后，将其进行聚合，以达到简化组合服务整体QoS值计算的目的，在此定义一个服务s的本地QoS值为各属性的权重与属性值的乘积和，及下面等式：1.2.3) After completing the normalization operation of various attributes, aggregate them to achieve the purpose of simplifying the calculation of the overall QoS value of the combined service. Here, define the local QoS value of a service s as the weight and attribute value of each attribute And the following equation:
locQos＝w
1×responseTime(s)+w
2×availability(s)+w
3×reliability(s)
locQos=w 1 ×responseTime(s)+w 2 ×availability(s)+w 3 ×reliability(s)
其中w
1,w
2,w
3分别是用户给定的响应时间、可得性与可靠性的比例权重，范围为[0,1]，且三者相加和为1，可以看出本地QoS值的范围为[0,3]。通过上述对服务的QoS操作，服务的各类属性值便都聚合为一个值，但是组合之后组合服务的各类QoS值依然与分开计算各个QoS属性的组合值是相等的。
Among them, w 1 , w 2 , and w 3 are the proportional weights of response time, availability and reliability given by the user, in the range of [0,1], and the sum of the three is 1, which shows the local QoS The range of values is [0,3]. Through the above QoS operations on the service, the various attribute values of the service are all aggregated into one value, but after the combination, the various QoS values of the combined service are still equal to the combined value of each QoS attribute separately calculated.
步骤2的具体步骤为：The specific steps of step 2 are:
在服务组合过程中，服务之间的输入输出参数匹配成为关键，而对于直接从WSDL文档解析获取输入输出参数类型进行匹配时缺少语义匹配，导致很多适配情况丢失，而且准确性无法得到保证，而如果在服务描述中增加语义说明，则能够较好的解决此类问题；因而有必要根据服务的WSDL文档说明生成对应的OWL-S文档；两种语言的映射关系如图2所示：In the service composition process, the input and output parameter matching between services becomes the key, and the lack of semantic matching when matching the input and output parameter types obtained directly from the WSDL document parsing has caused many adaptations to be lost, and the accuracy cannot be guaranteed. However, if semantic description is added to the service description, this kind of problem can be better solved; therefore, it is necessary to generate the corresponding OWL-S document according to the WSDL document description of the service; the mapping relationship between the two languages is shown in Figure 2:
如图2所示利用两种语言之间互补的优势，一方面，开发人员通过使用OWL-S的过程模型(process model)及其相对于XSD(XML Schema Definition)具有更强大表达力的OWL类型机制而获益，另一方面，开发人员还能重复使用WSDL中已经做好的大量工作(即相关的语言如SOAP)，并根据这些声明的消息交流的软件支持来定义最新的各种协议和传输机制。在此重点关注OWL-S/WSDL grounding转换关系；一个OWL-S/WSDL grounding的产生基于如下三种OWL-S与WSDL的对应关系，图2展示出了前两种：As shown in Figure 2, taking advantage of the complementary advantages between the two languages, on the one hand, developers use the process model of OWL-S and its OWL type, which is more expressive than XSD (XML Schema Definition). On the other hand, developers can reuse a large amount of work already done in WSDL (ie related languages such as SOAP), and define the latest various protocols and protocols based on the software support of these declared message exchanges. Transmission mechanism. Here we focus on the OWL-S/WSDL grounding conversion relationship; the generation of an OWL-S/WSDL grounding is based on the following three corresponding relationships between OWL-S and WSDL. Figure 2 shows the first two:
(1)一个OWL-S的原子过程(atomic process)对应一个WSDL的操作(operation)；不同类型的操作对应的OWL-S操作如下：(1) An OWL-S atomic process corresponds to a WSDL operation; the OWL-S operations corresponding to different types of operations are as follows:
A.同时拥有输入(inputs)与输出(outputs)的一个原子过程对应于一个WSDL中的请求-响应(request-response)操作；A. An atomic process with inputs and outputs at the same time corresponds to a request-response operation in WSDL;
B.只有输入没有输出的原子过程对应于一个WSDL的单向(one-way)操作；B. An atomic process with only input but no output corresponds to a one-way operation of WSDL;
C.一个只有输出没有输入的原子过程对应于一个WSDL通告(notification)操作；C. An atomic process with only output and no input corresponds to a WSDL notification operation;
D.拥有输出与输入，而且一个过程的输出传送给下一个过程的输入的一个组合过程(composite process)对应于WSDL的要求-应答(solicit-response)操作；D. A combined process that has output and input, and the output of one process is transmitted to the input of the next process (composite process) corresponds to the request-response (solicit-response) operation of WSDL;
(2)一个OWL-S的原子过程(atomic process)的输入集与输出集对应一个WSDL的消息概念(message)。确切的说，OWL-S的输入对应于WSDL操作的输入消息的part元素，OWL-S的输出对应于WSDL操作输出消息的part元素；(2) The input set and output set of an OWL-S atomic process correspond to a WSDL message concept (message). To be precise, the input of OWL-S corresponds to the part element of the input message of the WSDL operation, and the output of OWL-S corresponds to the part element of the output message of the WSDL operation;
(3)OWL-S原子过程的输入输出参数(即OWL类型)对应于WSDL中的抽象类型概念(可用于WSDL说明的消息部分)。(3) The input and output parameters of the OWL-S atomic process (that is, the OWL type) correspond to the abstract type concept in WSDL (which can be used in the message part of the WSDL description).
由于OWL-S是一种基于XML的语言，并且它的原子过程声明与输入输出类型已经与WSDL有着良好的对应关系，因此很容易将现有的WSDL绑定(binding)适用于OWL-S(例如SOAP绑定)；Since OWL-S is an XML-based language, and its atomic process declaration and input and output types have a good correspondence with WSDL, it is easy to apply the existing WSDL binding (binding) to OWL-S ( For example, SOAP binding);
值得一提的是现有的服务说明还不能满足基于QoS的服务组合要求，而且一般服务说明也没有包括QoS属性，所以必须在文档中嵌入服务的QoS属性值。It is worth mentioning that the existing service descriptions still cannot meet the requirements of QoS-based service combination, and the general service descriptions do not include QoS attributes, so the QoS attribute values of the services must be embedded in the document.
步骤3的具体步骤为：The specific steps of step 3 are:
步骤3.1)服务匹配：首先由于本发明只关心服务的输入参数集、输出参数集与QoS属性值，于是定义服务可以使用一个三元组表示：Step 3.1) Service matching: First, since the present invention only cares about the input parameter set, output parameter set and QoS attribute value of the service, the service definition can be represented by a triplet:
WS＝{wi，wo，qos}WS={wi,wo,qos}
其中w
i是输入参数集，其中w
o是输出参数集，而qos则是服务的本地QoS属性值。
Where w i is the input parameter set, w o is the output parameter set, and qos is the local QoS attribute value of the service.
服务WS
B与服务WS
A是否匹配取决于WS
B的w
o
B是否与WS
A的w
i
A匹配。服务参数类型的匹配分为Exact，Subsume，Relaxed，Fail四种类型；假设w
o
B的一个参数为O
B，而w
i
A的一个参数是I
A，则：
Service Service WS A and WS B matches depending on whether the w WS B WS A o B and the w i A match. The matching of service parameter types is divided into four types: Exact, Subsume, Relaxed, and Fail; assuming that a parameter of w o B is O B , and a parameter of w i A is I A , then:
(1)Exact：O
B与I
A的概念(concept)是等价的(equivalent)；
(1) Exact: The concept of O B and I A is equivalent;
(2)Subsume：O
B是I
A的子概念(sub-concept)；
(2) Subsume: O B I A is a sub-concepts (sub-concept);
(3)Relaxed：I
A是O
B的父概念(super-concept)；
(3) Relaxed: I A is the super-concept of O B;
(4)Fail：除了上面匹配，其他都为Fail；(4) Fail: Except for the above matches, all others are Fail;
在此定义O
B与I
A匹配为：O
B与I
A概念等价或者O
B是I
A的子概念。于是对于WS
B的w
o
B与WS
A的w
i
A匹配定义为：对于任意输入参数I
A∈w
i
A，都存在O
B∈w
o
B，使得O
B与I
A匹配。此处定义函数match(w
o
B，w
i
A)表示w
o
B与w
i
A匹配关系；故而只要WS
B的w
o
B与WS
A的w
i
A匹配则WS
B与WS
A匹配。
Here, we define that O B and I A match as: O B is equivalent to the concept of I A or O B is a sub-concept of I A. Thus for w WS B o w i A match is defined as WS A and B: For any input parameter I A ∈w i A, there are O B ∈w o B, B and O such that I A match. As defined herein function match (w o B, w i A) denotes i A matching relationship w o B and W; therefore as long as the WS w B B o w i A match is the WS and WS A and WS A B match.
步骤3.2)服务组合与选择问题模型：Step 3.2) Service composition and selection problem model:
描述服务组合选择问题如下：在给定服务集WS＝{WS
1，WS
2，…，WS
n}(n为服务总数)条件下寻找满足服务请求R＝{w
i
R，w
o
R，qos}的服务组合方案问题定义为寻找一个特定目标服务集WS
g＝{WS
g(1)，WS
g(2)，…，WS
g(k)}，且一旦该服务集确定，根据服务之间的匹配关系，服务集组合调用流程图也相应确定，并且此目标服务集满足以下条件：
Describe the service composition selection problem as follows: in a given service set WS = {WS 1 , WS 2 ,..., WS n } (n is the total number of services) to find a satisfactory service request R = {w i R , w o R , qos }’S service composition solution problem is defined as finding a specific target service set WS g = {WS g(1) , WS g(2) …, WS g(k) }, and once the service set is determined, according to the service set The matching relationship of the service set combination call flow chart is also determined accordingly, and the target service set meets the following conditions:
(1)match(w
i
R∪w
o
g(1)∪w
o
g(2)∪…∪w
o
g(i)，w
i
g(i+1))＝true；
(1)match(w i R ∪w o g(1) ∪w o g(2) ∪…∪w o g(i) , w i g(i+1) )=true;
(2)match(w
o
g(1)∪w
o
g(2)∪…∪w
o
g(k)，w
o
R)＝true；
(2)match(w o g(1) ∪w o g(2) ∪…∪w o g(k) ,w o R )=true;
(3)组合服务方案WS
g全局qos最优；
(3) The combined service plan WS g is the global qos optimal;
其中下标数组g＝{g(1)，g(2)，…,g(k)}中所有元素值均小于n，且k为下表数组大小；由上可知在搜索目标服务集过程中，不仅每步得出的中间服务集输出参数集并上请求输入参数集要与下一个可调用服务的输入参数集匹配，而且搜索完成后整个服务集的输出参数要与请求输出参数集匹配；不仅如此，在满足用户的功能要求之外，应尽可能使组合方案的全局QoS属性值最优。Wherein the subscript array g={g(1), g(2),...,g(k)} all element values are less than n, and k is the size of the array in the table below; from the above we can see that in the process of searching for the target service set , Not only the output parameter set of the intermediate service set obtained at each step and the request input parameter set must match the input parameter set of the next callable service, but also the output parameters of the entire service set must match the request output parameter set after the search is completed; Not only that, in addition to meeting the user's functional requirements, the global QoS attribute value of the combined scheme should be optimized as much as possible.
步骤4的具体步骤为：The specific steps of step 4 are:
上述服务组合问题可描述为在一个服务匹配图中寻找能满足服务请求并拥有最优QoS值的服务组合方案子图问题，称为SSOD问题；本发明在解决SSOD问题上提出了基于广度优先搜索的WSACA(Web Service Atomatic Composition Algorithm)算法。The above-mentioned service composition problem can be described as the problem of finding a service composition scheme subgraph that can satisfy the service request and have the optimal QoS value in a service matching graph, which is called the SSOD problem; the present invention proposes a breadth-based search for solving the SSOD problem. WSACA (Web Service Atomatic Composition Algorithm) algorithm.
步骤4.1)算法当中所使用到的数据结构和输出的哈希表：Step 4.1) The data structure and output hash table used in the algorithm:
表2.WSACA所用的数据结构Table 2. Data structure used by WSACA
表2所展示的是服务WSNode数据结构，在组合选择算法计算中所使用，也可以添加额外结构如关联的服务列表，这样在接下来产生服务组合OWL-S文档时会显得方便一些；Table 2 shows the data structure of the service WSNode, which is used in the calculation of the combination selection algorithm. You can also add additional structures such as the associated service list, so that it will be more convenient when the service combination OWL-S document is generated next;
表3.WSACA算法完成后输出的哈希表Table 3. Hash table output after WSACA algorithm is completed
此哈希表称为RefTable哈希表，用来存储至今为止提供特定输出参数类型并具有最好全局QoS的服务，即一个输出参数提供者的存储查找表。This hash table is called the RefTable hash table, and is used to store the service that provides a specific output parameter type and has the best global QoS so far, that is, a storage lookup table of an output parameter provider.
步骤4.2)服务自动组合选择算法WSACA：Step 4.2) Automatic service combination selection algorithm WSACA:
基本思想是基于广度优先搜索的，利用现有的服务输入查找能够调用的服务列表，然后将可调用服务的输出加入可得输入之中，更新服务的全局QoS及RefTable哈希表，并进行下一层次的搜索，其具体算法描述如下：The basic idea is based on breadth-first search, using the existing service input to find the list of services that can be called, then adding the output of the callable service to the available input, updating the global QoS and RefTable hash table of the service, and proceeding The specific algorithm of the first-level search is described as follows:
由上面描述的算法WSACA能够在用户给定输入并请求特定输出后，根据用户的功能请求与具有全局最优QoS值的目标驱动下自动寻找组合服务方案。The algorithm WSACA described above can automatically find a combined service plan based on the user's function request and the goal with the global optimal QoS value after the user has given input and requested a specific output.
步骤5的具体步骤为：The specific steps of step 5 are:
由WSACA算法得到的RefTable并没有得出最后的组合方案表示结果，只是找出了算法认为提供特定类型并具有最优全局QoS的服务，间接的求出了一个服务组合流程图；必须将上面服务的组合形式生成OWL-S文件以方便将来的调用或者组合服务发布，所以必须有一个算法能自动生成组合服务的OWL-S文档说明。The RefTable obtained by the WSACA algorithm does not give the final combination plan expression result, but finds out the service that the algorithm believes to provide a specific type and has the best global QoS, and indirectly finds a service composition flowchart; the above services must be OWL-S file is generated in the combined form of OWL-S to facilitate future calls or composite service release, so there must be an algorithm that can automatically generate OWL-S document descriptions for composite services.
步骤5.1)为了从服务组合流程图中得到关键路径生成树，使用一个HashMap存储关键路径树(注意只存储其中链接关系)，关键字是WSNode类型，值的类型是List<WSNode>服务列表类型，其中关键字表示的是父亲，而值是其连接的孩子序列，因此只需遍历一次流程图将关键路径生成树上的各个连接关系存储下来即可，具体生成关键路径树的算法如下：Step 5.1) In order to obtain the critical path spanning tree from the service composition flowchart, use a HashMap to store the critical path tree (note that only the link relationship is stored), the key is the WSNode type, and the value type is the List<WSNode> service list type. The keyword represents the father, and the value is the sequence of its connected children, so you only need to traverse the flowchart once to store each connection relationship on the critical path spanning tree. The specific algorithm for generating the critical path tree is as follows:
由于响应时间的限制，在构造了关键路径生成树之后产生的控制流程能保证服务被调用时与之有数据相关的服务输出都已经具备了，这就是构造关键路径生成树的必要性。在服务中添加了一个输入关联服务哈希表，即存储了每个服务的输入与提供此输入数据对应的服务输出，从而实现服务间数据关联绑定。Due to the limitation of response time, the control flow generated after the critical path spanning tree is constructed can ensure that the service output related to the data is available when the service is called. This is the necessity of constructing the critical path spanning tree. An input correlation service hash table is added to the service, that is, the input of each service and the service output corresponding to the input data are stored, so as to realize the data association and binding between services.
步骤5.2)实例说明此生成算法：Step 5.2) An example illustrates this generation algorithm:
由图3可以看出，用户给定输入A，请求输出E，组合服务的全局QoS值(即组合服务的响应时间)为65。组合服务的调用时间长短取决于流程图中的关键调用路径，在此为了解决问题的方便，定义服务流程图的关键路径生成树为：1)它是组合服务流程图的一棵生成树；2)树上的每个服务的全局QoS与父亲服务的全局QoS值相减之后得到的值与服务的本地QoS值相等。It can be seen from Figure 3 that the user is given input A and requests output E, and the global QoS value of the combined service (ie, the response time of the combined service) is 65. The call time of the composite service depends on the critical call path in the flowchart. Here, for the convenience of solving the problem, the critical path spanning tree of the service flowchart is defined as: 1) It is a spanning tree of the composite service flowchart; 2 ) The value obtained by subtracting the global QoS value of each service on the tree from the global QoS value of the parent service is equal to the local QoS value of the service.
在此可证明组合服务关键路径生成树存在：首先从起点(用户提供输入集)到服务的最长路径(即关键路径)显然存在(可将服务的本地QoS映射到边上)，取起点到每个服务关键路径中的一条，所组成的便是一个联通无回路的子图，这便是生成的一棵关键路径树。Here, it can be proved that the spanning tree of the critical path of the composite service exists: first, the longest path from the starting point (the input set provided by the user) to the service (the critical path) obviously exists (the local QoS of the service can be mapped to the edge), and the starting point One of the critical paths of each service is composed of a sub-graph of China Unicom without loops, which is the generated critical path tree.
根据上述算法将图3服务组合图生成组合服务关键路径树如图4所示：According to the above algorithm, the composite service critical path tree is generated from the service composition diagram in Figure 3 as shown in Figure 4:
组合服务关键路径生成树生成之后，从服务开端开始，每个服务对应一个Perform，而整个组合服务流程对应一个Sequence，并且从服务开端之后，其接下来的每棵子树对应一个Sequence，而开端之后的Sequence控制结构添加一个SplitJoin结构，这个SplitJoin结构再将子树对应的Sequence一一加入本身，之后子树同样用此方法来递归构造服务控制结构，这便形成了整个组合服务的控制构造，图4生成关键路径树之后，利用上述原理得到的控制构造如图5所示：After the composite service critical path spanning tree is generated, starting from the beginning of the service, each service corresponds to a Perform, and the entire composite service process corresponds to a Sequence, and from the beginning of the service, each subsequent subtree corresponds to a Sequence, and after the beginning Add a SplitJoin structure to the Sequence control structure. This SplitJoin structure adds the Sequence corresponding to the subtree one by one to itself, and then the subtree also uses this method to recursively construct the service control structure, which forms the control structure of the entire composite service. 4 After generating the critical path tree, the control structure obtained by using the above principles is shown in Figure 5:
由上面组合服务控制可知，组合服务默认由一个主Sequence构成，在主Sequence中只有一个SplitJoin结构，SplitJoin中添加了Sequence A0与Sequence A1，在SplitJoin结构中的Sequence之间的服务都可并行调用例如上述的服务W1与W2，但是Sequence之间的服务可能有数据关联绑定，如上图服务W2的输出要提供给W3的输入，但是由于响应时间的限制，在构造了关键路径生成树之后产生的控制流程能保证服务被调用时与之有数据相关的服务输出都已经具备了，这就是构造关键路径生成树的必要性。From the above composite service control, it can be seen that the composite service is composed of a main Sequence by default. There is only one SplitJoin structure in the main Sequence. The Sequence A0 and Sequence A1 are added to the SplitJoin. The services between the Sequences in the SplitJoin structure can be called in parallel. For example The above services W1 and W2, but the service between Sequences may have data association binding, as shown in the above figure, the output of service W2 should be provided to the input of W3, but due to the limitation of response time, it is generated after the critical path spanning tree is constructed The control flow can ensure that the service output related to the data is available when the service is called. This is the necessity of constructing the critical path spanning tree.
由于响应时间的限制，在构造了关键路径生成树之后产生的控制流程能保证服务被调用时与之有数据相关的服务输出都已经具备了，这就是构造关键路径生成树的必要性。在服务中添加了一个输入关联服务哈希表，即存储了每个服务的输入与提供此输入数据对应的服务输出，从而实现服务间的数据关联绑定。Due to the limitation of response time, the control flow generated after the critical path spanning tree is constructed can ensure that the service output related to the data is available when the service is called. This is the necessity of constructing the critical path spanning tree. An input association service hash table is added to the service, that is, the input of each service and the service output corresponding to the input data are stored, so as to realize the data association binding between services.
以上所述，仅为本发明中的具体实施方式，但本发明的保护范围并不局限于此，任何熟悉该技术的人在本发明所揭露的技术范围内，可理解想到的变换或替换，都应涵盖在本发明的包含范围之内，因此，本发明的保护范围应该以权利要求书的保护范围为准。The above are only specific implementations of the present invention, but the protection scope of the present invention is not limited to this. Anyone familiar with the technology can understand conceivable changes or substitutions within the technical scope disclosed in the present invention. All should be covered within the scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.
Claims (5)
- 一种基于关键路径生成树的自动服务组合方法，其特征在于，包括以下步骤：An automatic service composition method based on a critical path spanning tree is characterized in that it comprises the following steps:步骤1)计算组合服务的QoS模型；Step 1) Calculate the QoS model of the combined service;步骤2)构建服务组合与选择的问题模型；Step 2) Construct a problem model of service composition and selection;步骤3)创建服务自动组合与选择算法WSACA；Step 3) Create automatic service combination and selection algorithm WSACA;步骤4)构设服务组合关键路径生成树算法。Step 4) Construct a spanning tree algorithm for the critical path of service composition.
- 根据权利要求1所述的一种基于关键路径生成树的自动服务组合方法，其特征在于，步骤1)具体步骤为：An automatic service composition method based on a critical path spanning tree according to claim 1, wherein the specific steps of step 1) are:步骤1-1)提出响应时间的全局QoS计算方法，公式如下：Step 1-1) Propose a global QoS calculation method for response time, the formula is as follows:
其中，服务与服务前后关系有三种：序列(Sequence)，分支(Split)和汇合(Join)，假设getLocQos(S)为得到服务S的本地响应时间(即服务的单独调用响应时间)，而getGlbQos(S)为获取服务S的全局响应时间(在组合服务过程中运行完此服务所需的响应时间)，glbQos(S)表示得到服务S的全局响应时间为qos，并能改变它的值；Among them, there are three types of service and service context: Sequence, Split and Join. Assume that getLocQos(S) is the local response time of service S (that is, the response time of a single service call), and getGlbQos (S) To obtain the global response time of service S (the response time required to run this service in the process of combining services), glbQos(S) indicates that the global response time of service S is qos, and its value can be changed;步骤1-2)通过对可靠性与可得性的取对数与归一化操作使得其全局QoS计算方法与响应时间一样；归一化公式如下：Step 1-2) Through the logarithm and normalization of reliability and availability, the global QoS calculation method is the same as the response time; the normalization formula is as follows:
其中qos*为服务的原始响应时间值(真实响应时间值)，而qos max为所有服务当中的最大响应时间，qos min为所有服务当中的最小响应时间，qos为归一化之后的响应时间； Where qos* is the original response time value of the service (real response time value), and qos max is the maximum response time among all services, qos min is the minimum response time among all services, and qos is the normalized response time;步骤1-3)通过用户权重分配从而完成各项属性QoS值的聚合。Step 1-3) Through user weight distribution, the aggregation of the QoS values of various attributes is completed. - 根据权利要求2所述的基于关键路径生成树的自动服务组合方法，其特征在于，步骤2)具体步骤为：The method for automatic service composition based on critical path spanning tree according to claim 2, wherein the specific steps of step 2) are:步骤2-1)服务匹配，基于服务输入参数集、输出参数集与QoS属性值给出两个服务匹配的判断规则，用以辅助服务组合与选择问题模型的构建；Step 2-1) Service matching, based on the service input parameter set, output parameter set and QoS attribute value, the judgment rule of two service matching is given to assist the construction of service combination and selection problem model;步骤2-2)服务组合与选择问题模型，在给定服务集内寻找满足服务请求的服务组合方案问题定义为寻找一个特定目标服务集，且一旦该服务集确定，根据服务之间的匹配关系，服务集组合调用流程图也相应确定。Step 2-2) Service composition and selection problem model, find a service composition solution that satisfies the service request in a given service set. The problem is defined as finding a specific target service set, and once the service set is determined, according to the matching relationship between services , The service set combination call flow chart is also determined accordingly.
- 根据权利要求3所述的基于关键路径生成树的自动服务组合方法，其特征在于， 步骤3)具体步骤为：The method for automatic service composition based on a critical path spanning tree according to claim 3, wherein the specific steps of step 3) are:步骤3-1)建立算法当中所用到的数据结构和输出的哈希表；Step 3-1) Establish the data structure and output hash table used in the algorithm;步骤3-2)构设服务自动组合选择算法WSACA，利用现有的服务输入查找能够调用的服务列表，然后将可调用服务的输出加入可得输入之中，更新服务的全局QoS及RefTable哈希表，并进行下一层次的搜索，具体算法描述如下：Step 3-2) Construct the service automatic combination selection algorithm WSACA, use the existing service input to find the list of services that can be called, and then add the output of the callable service to the available input, and update the global QoS and RefTable hash of the service Table, and perform the next level of search, the specific algorithm is described as follows:
由上面描述的算法WSACA能够在用户给定输入并请求特定输出后，根据用户的功能请求与具有全局最优QoS值的目标驱动下自动寻找组合服务方案。The algorithm WSACA described above can automatically find a combined service plan based on the user's function request and the goal with the global optimal QoS value after the user has given input and requested a specific output. - 根据权利要求4所述的基于关键路径生成树的自动服务组合方法，其特征在于，步骤4)的具体步骤为：The automatic service composition method based on critical path spanning tree according to claim 4, wherein the specific steps of step 4) are:提出服务流程图的关键路径生成树算法，选择步骤3中搜索出的服务用以组合；在服务中添加了一个输入关联服务哈希表，即存储了每个服务的输入与提供此输入数据对应的服务输出，从而实现服务间数据关联绑定；随后采用关键路径树生成算法，同时生成组合服务的OWL-S文件以方便调用或者发布；具体生成关键路径树的算法如下：The key path spanning tree algorithm of the service flowchart is proposed, and the services searched in step 3 are selected for combination; an input correlation service hash table is added to the service, that is, the input of each service is stored corresponding to the input data provided To realize the data association and binding between services; then use the critical path tree generation algorithm, and generate the OWL-S file of the combined service to facilitate the call or release; the specific algorithm for generating the critical path tree is as follows:
由于响应时间的限制，在构造了关键路径生成树之后产生的控制流程能保证服务被调用时与之有数据相关的服务输出都已经具备了，这就是构造关键路径生成树的必要性。在服务中添加了一个输入关联服务哈希表，即存储了每个服务的输入与提供此输入数据对应的服务输出，从而实现服务间数据关联绑定。Due to the limitation of response time, the control flow generated after the critical path spanning tree is constructed can ensure that the service output related to the data is available when the service is called. This is the necessity of constructing the critical path spanning tree. An input correlation service hash table is added to the service, that is, the input of each service and the service output corresponding to the input data are stored, so as to realize the data association and binding between services.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910967241.6A CN110780862B (en) | 2019-10-12 | 2019-10-12 | Automatic service combination method based on key path spanning tree |
| CN201910967241.6 | 2019-10-12 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2021068581A1 true WO2021068581A1 (en) | 2021-04-15 |
Family
ID=69385159
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2020/101390 WO2021068581A1 (en) | 2019-10-12 | 2020-07-10 | Automatic service composition method based on critical path spanning tree |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN110780862B (en) |
| WO (1) | WO2021068581A1 (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110780862B (en) * | 2019-10-12 | 2021-05-11 | 南京邮电大学 | Automatic service combination method based on key path spanning tree |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101175090A (en) * | 2007-11-21 | 2008-05-07 | 华为技术有限公司 | Service synthesizing system and method thereof based on service characteristic |
| US20080201277A1 (en) * | 2007-02-16 | 2008-08-21 | Matsushita Electric Industrial Co., Ltd. | System architecture and process for automating intelligent surveillance center operation |
| CN101605141A (en) * | 2008-08-05 | 2009-12-16 | 天津大学 | Web service relational network system based on semanteme |
| CN101719932A (en) * | 2009-11-20 | 2010-06-02 | 中国科学院计算技术研究所 | System and method for automatic service combination |
| CN103106269A (en) * | 2013-02-04 | 2013-05-15 | 南京大学 | Web service combination method based on searching space reduction |
| CN110780862A (en) * | 2019-10-12 | 2020-02-11 | 南京邮电大学 | Automatic service combination method based on key path spanning tree |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100106767A1 (en) * | 2008-10-24 | 2010-04-29 | Microsoft Corporation | Automatically securing distributed applications |
| CN103544547B (en) * | 2013-11-02 | 2016-08-17 | 南京航空航天大学 | A kind of trusted service combined method based on discretization overall situation distribution search |
| US20150149362A1 (en) * | 2015-02-04 | 2015-05-28 | vitaTrackr, Inc. | Encryption and Distribution of Health-related Data |
| CN108170404B (en) * | 2017-12-25 | 2021-04-06 | 南京航空航天大学 | Web service combination verification method based on parameterized model |
| CN108764805A (en) * | 2018-06-11 | 2018-11-06 | 河南理工大学 | A kind of multi-model self-adapting recommendation method and system of collaborative logistics Services Composition |
| CN110120888B (en) * | 2019-04-26 | 2022-03-18 | 井冈山大学 | Cloud crowdsourcing platform mass service resource combination optimization method and system |
-
2019
- 2019-10-12 CN CN201910967241.6A patent/CN110780862B/en active Active
-
2020
- 2020-07-10 WO PCT/CN2020/101390 patent/WO2021068581A1/en active Application Filing
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080201277A1 (en) * | 2007-02-16 | 2008-08-21 | Matsushita Electric Industrial Co., Ltd. | System architecture and process for automating intelligent surveillance center operation |
| CN101175090A (en) * | 2007-11-21 | 2008-05-07 | 华为技术有限公司 | Service synthesizing system and method thereof based on service characteristic |
| CN101605141A (en) * | 2008-08-05 | 2009-12-16 | 天津大学 | Web service relational network system based on semanteme |
| CN101719932A (en) * | 2009-11-20 | 2010-06-02 | 中国科学院计算技术研究所 | System and method for automatic service combination |
| CN103106269A (en) * | 2013-02-04 | 2013-05-15 | 南京大学 | Web service combination method based on searching space reduction |
| CN110780862A (en) * | 2019-10-12 | 2020-02-11 | 南京邮电大学 | Automatic service combination method based on key path spanning tree |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110780862B (en) | 2021-05-11 |
| CN110780862A (en) | 2020-02-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Li et al. | An efficient and reliable approach for quality-of-service-aware service composition | |
| US20130124708A1 (en) | Method and system for adaptive composite service path management | |
| Di Martino et al. | Towards an Ontology-Based Intercloud Resource Catalogue--The IEEE P2302 Intercloud Approach for a Semantic Resource Exchange | |
| Luo et al. | A novel heuristic algorithm for QoS-aware end-to-end service composition | |
| JP2009531760A (en) | Service registry and related systems and methods | |
| US20170031986A1 (en) | Method and System for Providing Integrated Virtualized Database Management and Software Defined Network Topology | |
| Liao et al. | Accurate sub-swarms particle swarm optimization algorithm for service composition | |
| Jia et al. | An IoT service aggregation method based on dynamic planning for QoE restraints | |
| Miao et al. | Multicast virtual network mapping for supporting multiple description coding-based video applications | |
| WO2021068581A1 (en) | Automatic service composition method based on critical path spanning tree | |
| Dong et al. | Semantic service retrieval and QoS measurement in the digital ecosystem environment | |
| Chen et al. | Weighted overlay design for topic-based publish/subscribe systems on geo-distributed data centers | |
| Wang et al. | QoS-based and network-aware web service composition across cloud datacenters | |
| Feng et al. | Load shedding and distributed resource control of stream processing networks | |
| Frankova et al. | Service and protection level agreements for business processes | |
| Huang et al. | A novel Bi-Ant colony optimization algorithm for solving multi-objective service selection problem | |
| Wang et al. | Low-cost web service discovery based on distributed decision tree in P2P environments | |
| Liu et al. | Node-Resource-and User-Demand-Aware resource allocation in NFV-enabled elastic optical networks | |
| Kim et al. | A trust type based model for managing QoS in web services composition | |
| Thißen et al. | Considering qos aspects in web service composition | |
| Toma et al. | An evaluation of discouvery approaches in Grid and Web services environments | |
| Zhang et al. | A policy-driven service composition method for adaptation in pervasive computing environment | |
| Zahoor et al. | Rule-based semi automatic web services composition | |
| CN107360210B (en) | Virtual machine allocation method for cloud computing data center considering energy consumption and access delay | |
| Parhi et al. | Discovery of sensor web registry services for WSN with Multi-layered SOA framework |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 20874453 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 20874453 Country of ref document: EP Kind code of ref document: A1 |