CONTROL VERSUS DATA-FLOW IN PARALLEL DATABASE MACHINES

The execution of a query in a parallel database machine can be control led in either a control flow way, or in a data flow way. In the former case a single system node controls the entire query execution. In the latter case the processes that execute the query, although possibly r unning on different nodes of the system, trigger each other. Lately, m any database research projects focus on data Bow control since it shou ld enhance response times and throughput. In this paper, we study cont rol versus data flow with regard to controlling the execution of datab ase queries. An analytical model is used to compare control and data f low in order to gain insights into the question which mechanism is bet ter under which circumstances. Also, some systems using data how techn iques are described, and we will investigate to which degree they are really data flow. Our results show that for particular types of querie s data flow is very attractive, since it reduces the number of control messages and balances these messages over the nodes.