Dynamic Allocation of Abundant Data Along Update Sub-Cycles To Support Update Transactions In Wireless Broadcasting

Supporting transactions processing over wireless broadcasting environment has attracted a considerable amount of research in a mobile computing system. To allow more than one conflicting transactions to be committed within the same broadcast cycle, the main broadcast cycle need to be decompose into a sub cycles. This decomposition contains both the original data to be broadcast in Rcast cycle and the updates come from the committed transactions on these data items called Ucast cycle. Allocation of updated data items along Ucast cycles is highly affecting the concurrency among conflicting transactions. Given the conflicting degree of data items, one can design proper data allocation along Ucast Cycles to increase the concurrency among conflicting transactions. We explore in this paper the problem of adjusting abundant data allocations to respond in effective way to the changes of data conflicting probability, and develop an efficient algorithm ADDUcast to solve this problem. Performance of our adjustment algorithms is analyzed and a system simulator is developed to validate our results. It is shown by our results that the ADDUcast is highly increased the average number of committed transactions within the same broadcast cycle. Keywords—data broadcast; Dynamic Allocation; concurrency control; cycle decomposition; Abundant Data