Using Real-Time Serializability and Optimistic Concurrency Control in Firm Real-Time Databases

In a real-time database system, it is difficult to meet all timing constraints due to the consistency requirements of the underlying database. Real-time database transaction scheduling requires the development of efficient concurrency control protocols that try to maximize the number of transactions satisfying their real-time constraints. However, when the transactions in the system are heterogeneous, they are not of the same importance some are of greater importance than others. Therefore, transaction importance should be taken into account in concurrency control. Although an optimistic approach has been shown to be better than locking protocols in real-time database systems, optimistic methods have the problems of unnecessary restarts and heavy restart overhead. In this paper, we propose a real-time serializability and a new optimistic concurrency control protocol called OCC-RTDATI (Optimistic Concurrency Control with Real-Time Serializability and Dynamic Adjustment of Serialization Order) which uses real-time serializability in the conflict resolution. In OCC-RTDATI the number of transaction restarts is minimized by dynamically adjusting the serialization order of the conflicting transactions. The need for dynamic serialization order adjustment is checked and the serialization order is updated in the validation phase. This provides more freedom to adjust the serialization order of conflicting transactions. These features are important and increase the probability that transactions of high importance will be completed before their deadlines.