Transaction Scheduling Heuristics in Mobile Distributed Real Time Database System

Background: Today's widely used small and portable mobile database technologies have geared toward the mobile distributed real-time database systems (MDRTDBS).Various real time applications like mobile devices, missile systems, navigation control systems, satellites and many others are some examples of MDRTDBS. In the new era of technology, a large domain of applications are based on MDRTDBS, meanwhile different intrinsic limitation like disconnection and mobility typically effect on its correct execution. Mobile distributed real-time systems have different wireless constrained such like energy, processing capacity, memory storage facilities and variable network communication channels. In last few years, different applications run on different mobile nodes needed a suitable transaction mechanism to complete their service without failing its deadline. In recent years researchers focused on MDRTDBS, to develop a suitable concurrency control, commit control method. Replication, check pointing, security, caching and query processing are some other hot research topics in the field of MDRTDBS. Objective: it is needed to maintain data consistency and correct results in mobile distributed real time database system. In our review we have identified key issues which might be considered for development of various transaction executing protocols. We have Introduce a taxonomy of different CC, commit, replication and security issues, which could be advantageous for design, and development of transaction protocols. Method: In the review we have discussed various concurrency, commit, replication methods. Apart from these we have discussed various check pointing, caching and query techniques which is developed in database system. A comparison among various concurrency and commit protocols has been done in the review. The role of different key methods which can affect and help the transaction execution in wireless environment is discussed separately in the paper. Results: Analytical results are not mentioned in the review paper. However the role and affect on the transaction execution are mentioned clearly. Issues and their advantages of different concurrency and commit protocols are mentioned. Conclusion: It is found that transaction processing is still a challenging area of research. A number of issues has been discussed and reviewed various approaches to control concurrency control and atomicity methods. We have presented a detailed survey and classification of various issues based on commit, concurrency, and replication methods for MTDRTDBS. However, in the paper different security, caching and query processing and check pointing issues has been also discussed which should be considered for future work. Database researchers have needed to integrate these issues with their work and develop a suitable protocol.