Enhancing timeliness and saving power in real-time databases

In data-intensive real-time embedded applications, it is desirable to process data service requests in a timely manner using fresh data, spending less power. However, related work is relatively scarce despite the importance. In this paper, we present an effective approach to reduce both deadline misses and power expenditure in real-time databases with one or more processor by merging similar real-time queries to decrease repeated data accesses and processing, while doing dynamic power management. In a simulation study, our approach substantially decreases deadline misses and power consumption compared to state-of-the-art baselines.