A real-time database system (RTDBS) is designed to provide timely response to the transactions of data-intensive applications. Each transaction processed in an RTDBS is associated with a timing constraint in the form of a deadline. Efficient transaction-scheduling algorithms are required to minimize the number of missed transaction deadlines. In thls paper a performance evaluation model is provided to enable distributed RTDBS designers to analyze transaction scheduling algorithms. The model is

more »

... eveloped progressing from a simple mathematical analysis to complicated simulations. The performance is expressed in terms of the fraction of satisfied transaction deadlines. The paper also provides an example simulation experiment implemented using the model presented. P; ,aubmit;: Probability that the ith operation of the transaction does not access site j. P; ,ubmit• = ' J Pr* Pread; + Pw * Pwrite,(i = 1, 2, ... , access_mean) (13) Pr: Probability that the operation is a read (see (1) ). Pread;: Probability that the data is not read from site j. Pw: Probability that the operation is a write (see (2) ). Pwrite;: Probability that the data is not written at site j. Pread; = 1 -Pread; Pread;: Probability that the data item is read from site j; from (10): 1 Pread· = -' 2n 1 Pread· = 1 --2 , n Pwrite· = 1 -Pwrite · J J