Correctness and Parallelism in Composite Systems

In recent years, databases have started to be used as intelligent repositories for a variety of semantically-richer systems. A consequence of such architectures is that transaction scheduling takes place throughout composite systems consisting of layered subsystems. Such transaction architec-tures have been studied extensively. Existing theory, however , limits the degree of parallelism, and makes a number of simplifying assumptions which cannot be taken for granted in practice. This paper proposes a new model and correctness criterion , stack connict consistency, for composite transactional systems. The main contribution of the new model is to establish the correctness conditions under which higher degrees of parallelism can be achieved between operations of the same transaction, as well as between connicting operations of diierent transactions, in a uniform way. This possibility , although hinted at previously, has not yet been exploited in practical composite systems. Hence, we hope to improve the practical impact of many key results in this area. 1 Introduction In general, a composite system is one in which scheduling takes place simultaneously and independently at several levels. For such systems, the most relevant transaction models are those based on a multilevel approach, i.e., those in which each scheduling level is implemented as a function of the operations and objects provided by the immediately-lower level Wei91, WS92]. The use of several levels of scheduling allows implementation details to be abstracted out, with the degree of parallelism being determined by higher-level semantics. This conngura-tion has received quite some attention since it corresponds to the internal architecture of many database management systems GMB + 81, MHL + 92, Lom92, WH93]. One important aspect of such systems is the degree of indepen