The MARUTI hard real-time operating system
ACM SIGOPS Operating Systems Review
The design of the spring kernel
Tutorial: hard real-time systems
Optimization in non-preemptive scheduling for aperiodic tasks
Optimization in non-preemptive scheduling for aperiodic tasks
A decomposition approach to non-preemptive real-time scheduling
Real-Time Systems
Mission-oriented replication of periodic tasks in real-time distributed systems
Mission-oriented replication of periodic tasks in real-time distributed systems
Scheduling an overloaded real-time system
Scheduling an overloaded real-time system
Scheduling Algorithms for Multiprogramming in a Hard-Real-Time Environment
Journal of the ACM (JACM)
Computers and Intractability: A Guide to the Theory of NP-Completeness
Computers and Intractability: A Guide to the Theory of NP-Completeness
Priority Inheritance Protocols: An Approach to Real-Time Synchronization
IEEE Transactions on Computers
Hi-index | 0.00 |
Real-timecomputer systems are essential for many applications, such asrobot control, avionics, medical instrumentation, manufacturing,etc. The correctness of the system depends on the temporal correctnessas well as the functional correctness of the task executions.In order to assure temporal correctness it is necessary thatthe resources be scheduled to meet the temporal requirementsof applications. When we consider the problem of nonpreemptivescheduling of a set of tasks in a processor for which no feasiblesolution exists, some tasks may have to be rejected so that aschedule can be generated for the rest. In this paper, we considerthe problem of generating an optimal schedule such that the numberof rejected tasks is minimized, and then the finish time is minimizedfor the accepted tasks. We propose to use an analytic approachto solve this problem. We first discuss the super sequence basedtechnique which was originally proposed for reducing the searchspace in testing the feasibility of a task set. Then we showby the Conformation theorem that the super sequence constructedfrom the task set also provides a valid and reduced search spacefor the optimization problem. While the complexity of our schedulingalgorithm in the worst case remains exponential, our simulationresults show that the cost is reasonable for the average case.