DATABASES WITH DEADLINE AND CONTINGENCY CONSTRAINTS

Real-time database systems associate the concept of deadlines with tra nsaction executions. Previous approaches use ''best effort'' technique s to schedule a given set of transactions to meet the deadlines as wel l as to ensure the consistency of the database. However, such approach es are inadequate for target applications which have ''hard'' real-tim e deadlines that need to be met in the event of crisis situations. In such cases, it is important to obtain contingency plans that may be in voked with guaranteed execution time characteristics. This paper prese nts an alternative model for real-time database systems in which deadl ines are associated with ''contingency'' constraints rather than direc tly with transactions. Our approach leads to a predicate-based model t hat intrinsically incorporates both triggering and relative timing con straints regarding the transaction executions. We exhibit that selecti ng contingency plans with respect to various optimality criteria has i nherent computational inefficiencies. We study the issues in schedulin g of the selected plans with the focus on the contention among the tra nsactions for data resources. Our results exhibit that the data conten tion, by itself, has a severe adverse impact on the schedulability of the deadline-constrained transactions. We discuss some of the practica l implications of our results, and we suggest some counter-measures to handle the computational complexities.