Research And Implementation Of Cloud Computing Services Composition In Freight System

Cloud computing is characterized by the application of hardware and software will be published as a service. And it is emphasizing service-oriented storage and calculated manner. The goal is to build lightweight software client service model. The application software is provided to users as services. And the storage and computation procedure are placed in the cloud. In theory it proposed the establishment of tiered service model, independent of hardware service and software services to publish. So that the different needs of users can get to meet the functional requirements.In our research, we firstly do the research on the cloud hierarchy proposed semantic-based service composition method. Using the semantic description method unified the original relatively independent three-tier service from the perspective of data integration point and service calls. Call the point of unity in the whole prototype system using all the data described in Resource Description Framework (RDF) and Web Ontology Language (OWL) format. All the user interfaces are Representational State Transfer (REST)-style Web services applications to encapsulate data and operation from a semantic point of view is given.We proposed using ontology as a cloud platform service model. And use the ontology technology describing the cloud computing platform state. We make reasoning according to the new tasks and the task queue situation. Automatically set the corresponding cluster computing services, take advantage of the hardware resources to improve the efficiency of task execution.In the study made use of RDF data format freight invoice data, changed the traditional textual storage means. Freight invoice data structured operational and verified by experiments storage and query efficiency greatly improved.Made use Mashup cloud computing technology and other data service for the synthesis of final results. And the newly add data will not only not break the semantic of the original data but also improve on the original data, while adding new semantic information to improve the applications.Through experiments in the study we developed a corresponding virtual machine cluster partition strategy. And to verify the use of RDF data stored HBase than the use of relational database queries are stored in the memory speed and the speed has greatly improved. And this discusses how the final release synthesis service with the semantic features of the full freight invoice.