Article Preview
Top1. Introduction
A Web service is a piece of software that is used to transmit information between two parties. It may be financial transactions, or text transactions, or data transactions or, media transactions, or Business-to-Business (B2B) interaction. But ordinarily, in the current world, a single service could not able to satisfy the heterogeneous B2B interaction process. This arises a concept called Semantic Web Service Composition (SWSC) process. Thus, there is need for a standard formats and organizing the heterogeneous components for enabling B2B interaction to compose the services in a semantic order for seamless execution of services, which brings about a concept called service-oriented-Architecture (SOA). According to (Anji Reddy et al 2012), the technical implementation of SOA is facilitated by a model termed Service-Oriented Computing (SOC). Real-world examples for the SWSC process are ticket booking, holiday-trip booking, e-book purchasing, e-shopping, and e-insurance system, etc. Usually, the SWSC process is attained by four stages: They are: i) Service Discovery – to retrieve functionally similar services, ii) Service Selection- to pick best services based on Quality of Service (QoS) values, iii) service composition – generate various possible plans in the ordering of execution of services, and iv) Service Execution – execute on optimal service.
In the foremost service discovery task, from the given user query request, the discovery process should semantically understand and extracts what are inputs (I) are given by the query, what are the outputs (O) expected by the user, what are the preconditions (P) of the user, and what is the effect (E) of a user (IOPE), then, using matchmaking algorithms like bipartite matching, etc, to retrieves the matched services which reside in the centralized web service repository (UDDI registry) (Amirthasaravanan et al., 2016). Then the best services are picked from the discovered list in terms of service-constraint QoS values and user-demanded QoS values (Meysam Ahmadi Oskooei & Salwani Mohd Daud, 2014), (Yaswanth, 2016), and (Maheswari & Karpagam, 2015). Then the top-ranked services based on the weightage given for specific QoS attributes are retrieved and considered for the composition process (Aram AlSedrani & Ameur Touir, 2016).
Figure 1. Different phases of semantic web service composition
Accordingly, several possible composition plans are generated by combining various selected individual services using composition methods like colored Petri-nets, Sword, OWL-X plan planner, etc. In some cases, the service composition order may be random. The services are executed in parallel if there is no dependency between services. If services are dependent then the output of one service is given as the input for another service. After that, optimal plans are generated by considering dependencies. Working on different web service composition phases is illustrated in Figure 1.
However, there is a chance for service providers to provide counterfeit information about their service functionality or instant shutdown of services, or removing any mock service information from the registry. Meanwhile, the users do not have the ability to check the background and origin of web services. The commercial purpose web services are grievous could contain counterfeit provider's information that can even expose to data breaches. The vast majority of these platforms are proprietary in nature, giving no negotiating power for users to fully trust the platform. The objective of the paper is to propose a smart contract enabled services provider agreement (smart contract instead of SLA agreement) and QoS aware service composition in order to enhance trust, clear communication, efficiency, and secure execution of web services.
The Ethereum Blockchain is considered a second-generation cryptocurrency that uses the proof-of-stake (PoS) consensus mechanism (Schwab, 2017). Rather than leveraging the blockchain platform solely for monetary transactions, the Ethereum Blockchain platform views itself as a network for general purpose decentralized, distributed, and trustworthy platform for developing real-world E-applications scenarios, which is technically denied in the concept of current financial transaction blockchain platforms (Z Zheng et al., 2017).