Research And Implementation Of The Distributed Computing Based On COM/DCOM

With the rapid development of computer technology, computing power and computing speed has made great improvement. The computing models are converted from host based centralized computing to network connection based distributed computing. The application of computer is also developed from numerical calculation to data processing, transaction processing, engineering control and information management. Moreover, with the rapid development of network technology and popularization, distributed technology has also been promoted and widespread.Currently, the major distributed applications are based on three or more layers of the distributed architecture, and multi-layer architecture is expanding the three layers architecture. Therefore, three layers distributed architecture not only can be helped to understand and master the architecture and principles of major distributed technology, but also lay the foundation for the further research of distributed technology. Since introduction and application of the object-oriented concept, object-oriented distributed technology has become the core technology of distributed application systems. By studying the distributed architecture, principles, communication mechanisms of COM/DCOM technology, the principles and implementation mechanisms of distributed technology can be understood in-depth.First, the article introduces the principle of COM/DCOM technology, studies the RPC communication mechanism and analyses its principles and mechanisms. Secondly, by analyzing the differences, advantages and disadvantages of scheduling algorithms in distributed systems, the article focuses on centralized control scheduling algorithm which uses periodic scanning strategy. For the defects of periodic scanning strategy, this article improves the centralized control scheduling algorithm and applies it to simulation platform. Again, this article applies the COM/DCOM technology to modeling and simulation platform, and combines with improved centralized control scheduling algorithm. It achieves good results. Finally, by analyzing the experimental data, the article verifies the feasibility of the system and good performance of scheduling algorithm. The article also summarizes the research results of the system, existing problems and the next step work, and looks forward to the works which need to study and explore in-depth.