Research On Mobile Query Technology Based On Agent

The rapid development of mobile communication technology makes traditional network more flexible. Users can access database and receive their query results through all kinds of wireless connections. Because of the characteristic difference between mobile computing entironment and fixed computing entironment, traditional query technology does not fit for mobile computing entironment. So the research on mobile query technology takes an important position in the research on mobile database.First, this paper is intended to research on the usability of weakly connected query, mobile query model and disposal, etc. It introduces the focused technology-Agent and analyzes defects of system architecture based on Agent. Meanwhile it proposes an Agent-based mobile query layered structure, which supports disconnection and transference of information among layers. It can also balance work load and improve query efficiency with the collaboration of Agent. The whole process of query requests was shown based on it. Then, the clustering technology was proposed to alleviate swarm problems of query result.Next, this paper induces and compares methods of Agent security and communication technology, taking efficiency, disconnection, etc into account. This paper proposes the idea that using Mobile Agent(MA)'s intelligent clone technology can solve the problems of improving MA's query efficiency and not aggravating the load of remote host at the same time. This paper applies data sign and data encryption to treat query result comprehensively, designs a mobile communication way as well as communication model based on ZQTM model, and shows MA location, message transference, migration communication algorithm, whose characteristics are location transparency, adaptability and fault tolerance, etc.Finally, the proposed clusting technology is verified through imitating experiments and using LoadRunner that is a industrial standard load testing tool. By analyzing the result, it indicates that average transaction response time shortens and the throughput increases so that it proves that clusting technology is effctive.