Scheduling Periodic Jobs that Allow Imprecise Results
IEEE Transactions on Computers
Algorithms for Scheduling Imprecise Computations
Computer - Special issue on real-time systems
Rate monotonic scheduling in hard real-time systems
Information Processing Letters
The Deferrable Server Algorithm for Enhanced Aperiodic Responsiveness in Hard Real-Time Environments
IEEE Transactions on Computers
Incremental Reconfiguration and Load Adjustment in Adaptive Real-Time Systems
IEEE Transactions on Computers
Minimizing Aperiodic Response Times in a Firm Real-Time Environment
IEEE Transactions on Software Engineering
Scheduling Algorithms for Multiprogramming in a Hard-Real-Time Environment
Journal of the ACM (JACM)
Optimal Reward-Based Scheduling for Periodic Real-Time Tasks
IEEE Transactions on Computers
Algorithms for Scheduling Imprecise Computations with Timing Constraints to Minimize Maximum Error
IEEE Transactions on Computers
IEEE Transactions on Computers
ECRTS '02 Proceedings of the 14th Euromicro Conference on Real-Time Systems
A least upper bound on the fault tolerance of real-time systems
Journal of Systems and Software
Journal of Systems and Software
Power saving and fault-tolerance in real-time critical embedded systems
Journal of Systems Architecture: the EUROMICRO Journal
Hi-index | 0.00 |
This paper addresses the problem of scheduling hard and non-hard real-time sets of tasks that share the processor. The notions of singularity and k-schedulability are introduced and methods based on them are proposed. The execution of hard tasks is postponed in such a way that hard deadlines are not missed but slack time is advanced to execute non-hard tasks. In a first application, two singularity methods are used to schedule mixed systems with hard deterministic sets and stochastic non-hard sets. They are compared to methods proposed by other authors (servers, slack stealing), background and M/M/1. The metric is the average response time in servicing non-hard tasks and the proposed methods show a good relative performance. In a second application, the previous methods, combined with two heuristics, are used for the on-line scheduling of real-time mandatory/reward-based optional systems with or without depreciation of the reward with time. The objective is to meet the mandatory time-constraints and maximize the reward accrued over the hyperperiod. To the best of the authors' knowledge, these are the only on-line methods proposed to address the problem and outperform Best Incremental Return, often used as a yardstick.