The Research Of An Access And Scheduling Model Based On Subject-Oriented Data

With the development of web2.0, the information resource has blust and the data of the web increased rapidly. In this Condition, the index and search of the data drives all the internet company to develop their own search engine, but the engine is still designed for the indiscrimination data from the web.The concept of the verbal search engine makes the searh from the web much more convenient.The scope of our search is changed from all data in web to the data of assigned industry,by this way the search of data is become "accuracy,fast, brieve".But the verbal search engine of the same industry all around has not get a standard,so the same data was stored more than once and the resource has been wasted serverly.At the same time,the index of data is based the article,which makes the located of the data is blurry.The data fragment is raised to make our search much more accurate, much more fast, much brieve. Because of the data fragement we can index data based theire title, paragraph,index and abstract,then we can locate at the accuracy part of the article.But the data fragement is now being used when the article is called and the data is still store as a whole body.And we must frage them when we use it.This undoubtedly sharpen the burden of our server.So store the fragement is needed for the search engine.This article is based on a cooperation project with a research center in Beijing. We have processed the data fragement and the subject extracted.So In the paper we intend to do work in this list: 1) Analyse the disadvantage and problem in this system and raise the concept of fragement data and the advantage in why we use this;2) Design a system mode for the store and callable of the fragement data based on the requirement,after the establish of the model,the system can store the mining data in the rule of our distribution system.3) Design the scheduling algorithm of multi processor,and improve the efficient of them to process message,and the index mechanism can quickly find the relative subject to search.