Distributed real time database systems: background and literature review
Distributed and Parallel Databases
A data-entity approach for component-based real-time embedded systems development
ETFA'09 Proceedings of the 14th IEEE international conference on Emerging technologies & factory automation
| Hi-index | 0.00 |
Many real-time systems need to maintain fresh viewswhich are derived from shared data that are distributedamong multiple sites. When a base data item changes,all derived views that are based on it need to be recomputed.There are two major derived data recomputationstrategies - immediate update and on-demandupdate. However, they both have their advantagesand limitations. In this paper, we study the performanceof derived data update using immediate andon-demand strategies in distributed real-time databasesand identify several criteria for choosing proper updatepolicies. Based on these criteria, we propose a deriveddata update algorithm. In our algorithm, the updatepolicy of a particular derived data item is determineddynamically by its access frequency, current transactionmiss ratio and the system utilization. A thoroughsimulation study shows that our algorithm outperformsimmediate and on-demand update in most cases.