Introduction There is a great need for building research prototypes of real-time database systems, because existing techniques are often ad hoc or concealed in internal solutions, and few, if any, commercial offerings exist. As real-time applications inherently use reactive mechanisms, and a wealth of results now exists on active database management systems [5], we decided to build DeeDS, a [D]istributed Activ[e], Real-Tim[e] [D]atabase [S]ystem prototype. This research effort is challenging due to the unique combination of reactive mechanisms, distributed technology, dynamic scheduling, and integrated monitoring, under a mixture of soft and hard real-time constraints. To make all this feasible with a limited amount of man-power, we have taken a deliberate approach of building with existing components where possible, and including only those features absolutely necessary to obtain the benefit of each of the architectural dimensions. At the same time, features a

This document describes a technology transfer activity of distributed active real-time database management systems in general, and the DeeDS prototype in particular, from University of Skovde to ENEA Data AB and ENEA OSE Systems AB, Taby, Sweden. This activity is planned to be carried out during Fall 1996, in the 6 month period immediately following the completion of the DeeDS prototype.

University of Skovde Technology Transfer: DeeDS Summary Beating complexity Complexity Scheduling Reactive mechanisms Replication Date: 1st October 1996 Slide: 4 University of Skovde Technology Transfer: DeeDS Summary 4 Notes With these three mechanisms we try to break up the complexity so it will be possible to study the environment and the control system in each of these perspectives before looking at the total perspective. The reason is the well-known divide-andconquer principle. However, the holistic perspective must not be forgotten. It is important to note that the DeeDS architecture is complex for reasons that it is meant to be used in solving highly complex problems. Date: 1st October 1996 Slide: 4-1 University of Skovde Technology Transfer: DeeDS Summary Scheduling Complexity Scheduling ffl Provides -- easy off-line schedulability analysis of hard tasks<F4.

Slide: 3-1 University of Skovde Technology Transfer: DeeDS Summary Beating complexity Complexity Scheduling Reactive mechanisms Replication Date: 1st October 1996 Slide: 4 University of Skovde Technology Transfer: DeeDS Summary 4 Notes With these three mechanisms we try to break up the complexity so it will be possible to study the environment and the control system in each of these perspectives before looking at the total perspective. The reason is the well-known divide-andconquer principle. However, the holistic perspective must not be forgotten. It is important to note that the DeeDS architecture is complex for reasons that it is meant to be used in solving highly complex problems. Date: 1st October 1996 Slide: 4-1 University of Skovde Technology Transfer: DeeDS Summary Scheduling Complexity Scheduling ffl Provides -- easy off-line schedulability analysis o