Research On Distributed Web Service Discovery Technology Based On Grid

As a new Internet technology,the fundamental purpose of grid is to aggregate wide-area computing resources and share wide-area information resources.In the open grid service architecture,it is emphasized to encapsulate and provide various resources in the form of services.The strong dynamics of grids lead to the continuous creation and deletion of grid services.Mass temporary short-term services and service requests are focused on In the entire grid system,the difficulty of discovering services with high accuracy,high response and high efficiency is also increasing.The traditional grid technology often adopts a centralized registry or aggregate catalog to register and discover service resources.However,as time goes by,the radiation scene and scale of grid services have risen by one dimension.Centralized service discovery is in progress.The existing single point of failure and performance bottlenecks are imminent,and the high-performance distributed service discovery architecture has become a result-oriented.This article first combines the structure of the overlay network in the Chord protocol of the peer-to-peer computing model,reorganizes the centralized service registry into a distributed registry,constructs a multi-layer partitioned overlay network structure,and disassembles the centralized network load.A distributed grid service discovery model based on improved Chord is presented,which avoids single point of failure and performance bottlenecks.Secondly,it analyzes the distributed service discovery algorithm part of the Chord protocol in detail.For the lack of consideration of local naming conflicts and lack of semantics,a distributed service discovery algorithm based on improved Chord is presented as a response.When mapping,the semantic empowerment of the identifier corresponding to the service resource is carried out through the semantic dictionary WordNet and the similarity calculation,and the area header is added to solve the problem of local naming conflicts.Finally,the algorithm proposed in this paper is verified by experiments.The experimental results show that the algorithm has a higher recall rate and precision rate,which has advantages over other distributed service discovery algorithms.