During the past few years, several studies have been made on the performance of real-time database systems with respect to the number of transactions that miss their deadlines. All of these studies have used simulation models or database testbeds as their performance evaluation tools. We present, in this paper, a preliminary analytical performance study of real-time transaction processing. Using a series of approximations, we derive simple closed-form solutions to reduced real-time database models. By virtue of their simplicity, these solutions provide considerable insight into the observed performance. Although quantitatively approximate, the solutions accurately capture system sensitivity to workload parameters and yield performance bounds. Our results indicate that increased transaction slack times degrade performance under heavy loads for the real-time database systems considered in this study. The analysis also shows that the absolute sizes of transaction data sets, independent of their relationship to the database size, have a significant impact on performance. Interestingly, our approximation techniques for real-time database models are applicable to classical single- server real-time models as well, resulting in simple approximations that closely match complex exact solutions presented in the literature.