There is growing concern among users of data base management systems regarding the cost and performance of such systems. As is typical of highly complex, generalized software, data management systems can be expensive to run, particularly in terms of processing time. A possible solution to this problem is to offload the data management functions into an inexpensive, dedicated minicomputer, called a dedicated data management computer, or DDM. This "backend" approach to the problem has been investigated by Bell Telephone Laboratories [1], using a large DBTG-based system. Their results indicated that such an approach is feasible, though no performance figures have been published. In an effort to obtain more detailed data on the performance implications of a DDM system, two experiments were performed by the Sperry Corporate Research Center. The first experiment was designed to obtain estimates of the percentage of CP time which could be offloaded from the host to a DDM. The second experiment, a prototype implementation of a DDM configuration using a standard minicomputer, was designed to demonstrate the credibility of such a system.

TWO time-sharing models are described.One is the conventional round-robin model in which each customer receives at most q seconds of service at a time.If this completes his service requirement, he leaves the system; otherwise he joins the end of the queue to await his next turn.The second model is a modification of the round-robin system in which the amount of service per pass depends on the rate at which programs arrive in the system.The models are analyzed under the assumption of constant, nonzero overhead when the processor swaps one program for another.Expressions are derived for the mean waiting time in queue as a function of service requirement and for the mean system cost due to waiting time in queue.