期刊名称：International Journal of Engineering and Computer Science
印刷版ISSN：2319-7242
出版年度：2015
卷号：4
期号：5
页码：11916-11921
出版社：IJECS
摘要：Transactions in real-time databases should be scheduled considering both data consistency and timingconstraints. In addition, a real-time database must adapt to changes in the operating environment andguarantee the completion of critical tasks. The effects of scheduling decisions and concurrency controlmechanisms for real-time databasesystems have typically been demonstrated in a simulated environment. Many time-critical applications datamay be distributed among multiple sites. For example, such applications include command and control,industrial automation, aerospace and defense systems, telecommunications, banking, etc. In suchapplications, it imperative that the data be available to the requesting transactions at the time it is needed. Ina typical distributed database, the transaction is required to access the remote data directly, at the risk ofmissing its deadline. Another problem can occur in such a scenario when the requesting transaction accessesthe data, but it is not temporally valid. That is, its value is “out-of-date” because the transaction did not readfrom the most recent update. A replication algorithm creates replication transaction based on client’s datarequirements in a distributed real time databases. These replication transaction copy data objects to the siteon which they are needed just in time for the deadline to occur. The algorithm carefully computes theparameters of the replication transactions so that we can guarantee that any requests that read data, in fact,read temporally valid data. This algorithm is designed to work in a static environment in which all objectlocations, and client data requirements are known a priori
