The Construction And Application Research Of Measurement System For Service Ecosystem

With the widespread adoption of Service-Oriented Architecture(SOA),consumers are accustomed to the ubiquitous and easily accessible services with offline and online.In the service-oriented environment,the complex linkages between services,service providers and consumers connect the service industry into a highly relevant value co-creation network,which is called service ecosystem.Compared with traditional products,services are more closely related to human activities.Due to the uncertainty and variability of human activities,there are many dynamic factors within the service ecosystem.The dynamic factors in the service ecosystem are spread through the links between the various elements of the service ecosystem and affect the entire ecosystem.Threrfore,researching the complex,open,dynamic and adaptive service ecosystem to ensure the robust and diversified development of the entire service ecosystem is becoming a research issue which is worthy of attention.Although a large number of studies have discussed the evolution of service ecosystems and their management and control,there is a lack of the metrics of service ecosystems.The service ecosystem includes seven elements,consumers,services,service providers,mashups,mashup providers,platform operators,and the environment.This thesis proposes the network model based on the relationships among elements and makes construction of ecosystem measurement system.The current researches on service ecosystem measurement don't take into account the relationship between service providers and consumers and services.Based on these limitations,this thesis proposes measurement indicators for service ecosystems.According to the metrics,through the experimental results,it can be seen that the Programmable Web service ecosystem has experienced a rapid growth in the number of Web services and the number of Mashup during the period from 2005 to 2016.The number of registered consumers has reached 66,741,and the number of service providers has reached 2,640.of The service ecosystem has the features of the diversity.Web service is the indispensable part of the service ecosystem.The performance of services can greatly affect the service ecosystem.Scholars have conducted a lot of research work on the performance of services.The most popular and widely used Web services are SOAP and REST.In order to measure the service performance of the Programmable Web service ecosystem in multiple dimensions,this thesis deploys experiments in three different geographic locations and selects different data subsets for comparison between SOAP and REST services.After the deployment of three sets of experiments,it can be known that REST have better permances than SOAP in terms of the average response time,the throughput,and the packet size.