Distributed Parallel Computing System Based On DCOM

This paper mainly studies distributed parallel computing theory and how to build a distributed system based on DCOM. With the development of computer and network ability, people's requirement for computing velocity, reliability of system and actual effect of cost must extremely stimulates the interesting in distributed parallel computing system. Distributed system can be used at a lot of fields now and could be used everywhere in the future. The paper firstly studies some basic concepts such as task distributing, remote calling, loading balance, processing synchronization and communication, distributed computing system model and parallel executing time connected to distributed parallel computing system. Secondly, comparing DCOM to CORBA and Java_RMI, which are three distributed computing object people are familiar to, we describe the theory of COM/DCOM. Finally, we build a distributed parallel computing system based on DCOM and use it to compute multiplication of two matrixs. Then we analyze the system's ability by using distributed parallel computing theory. The experimental results indicate that distributed parallel computing on multiple computers is faster than sequent computing on a single computer for complex problem on computing. In addition, there are some distributed parallel computing system's abilities that a single machine cannot own. The study on distributed parallel computing is significant on theory and in practice.