Processing Distributed Compoud-Data Streams

In the environment of distribute data stream systems, the available communication bandwidth is a bottleneck resource. It is significant to reduce the communication overhead as possible for improving the availability of communication bandwidth with the constraint of the precision of queries. In this paper, we propose a new method for transferring data streams in distributed data stream systems, named as compound-data streams. The idea is that raw data streams are grouped and merged into compound-data streams, and then compound-data streams, instead of raw data streams, are transferred to the central processor node. By this way, the communication overhead can be reduced greatly.