26th IEEE International Real-Time Systems Symposium (RTSS'05) (2005)
Dec. 5, 2005 to Dec. 8, 2005
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/RTSS.2005.5
Ming Xiong , Bell Labs, Lucent Technologies
Song Han , City University of Hong Kong
Kam-Yiu Lam , City University of Hong Kong
Periodic update transaction model has been used to maintain freshness (or temporal validity) of real-time data. Period and deadline assignment has been the main focus in the past studies such as theMore-Less scheme  in which update transactions are guaranteed by the Deadline Monotonic scheduling algorithm  to complete by their deadlines. In this paper, we propose a novel algorithm, namely deferrable scheduling, for minimizing imposed workload while maintaining temporal validity of real-time data. In contrast to previous work, update transactions scheduled by the deferrable scheduling algorithm follow a sporadic task model. The deferrable scheduling algorithm exploits the semantics of temporal validity constraint of real-time data by judiciously deferring the sampling times of update transaction jobs as late as possible. We present a theoretical analysis of its processor utilization, which is verified in our experiments. Our experimental results also demonstrate that the deferrable scheduling algorithm is a very effective approach, and it significantly outperforms the More-Less scheme in terms of reducing processor workload.
M. Xiong, S. Han and K. Lam, "A Deferrable Scheduling Algorithm for Real-Time Transactions Maintaining Data Freshness," 26th IEEE International Real-Time Systems Symposium (RTSS'05)(RTSS), Miami, Florida, 2005, pp. 27-37.