In this paper, we will focus on the contribution an efficient and an effective approach for a distributed discovery of Semantic Web services (SWs), which address some aspects, related to problems the time complexity of collaboration in the process of automatic discovery for SWs in order to achieve adaptability in a highly-dynamic distributed computing. For this purpose, our approach is based on P2P computing that proved to be scalable, efficient and robust solutions for distributed discovery of SWs. Indeed, we propose a matching technique of ontology OWL-S process model, in order to efficiently and effectively discover appropriate Web services distributed among all peers in a large-size P2P network. For improving performance and effectiveness, our scalable approach is mapped on unstructured P2P networks by exploiting Pareto efficiency to reduce considerably the execution time and message exchanged for disseminating request about the unstructured P2P network structure among the peers, to simplify the management of the P2P network and to ensure the quality of service. The experimental results show that the proposed approach enhances the network scalability while providing good overall performances. Also, we show that our approach can perform more efficient and effective distributed discovery of SWs with low computational cost in terms of completeness and accuracy in P2P computing, when compared to the other approaches.