Scheduling with bus access optimization for distributed embedded systems
read more
Citations
Energy-aware communication and task scheduling for network-on-chip architectures under real-time constraints
Timing analysis of the FlexRay communication protocol
Design Optimization of Time-and Cost-Constrained Fault-Tolerant Distributed Embedded Systems
Holistic scheduling and analysis of mixed time/event-triggered distributed embedded systems
References
Scheduling algorithms for multiprogramming in a hard real-time environment
Synthesis and optimization of digital circuits
Real-Time Systems: Design Principles for Distributed Embedded Applications
NP-complete scheduling problems
Dynamic critical-path scheduling: an effective technique for allocating task graphs to multiprocessors
Related Papers (5)
Frequently Asked Questions (10)
Q2. What have the authors stated for future works in "Scheduling with bus access optimization for distributed embedded systems" ?
Considering a TTP-based system architecture, the authors have shown that the general scheduling algorithm for conditional process graphs can be successfully applied if the strategy for message planning is adapted to the requirements of the TDMA protocol. There the authors also considered the possibility of messages being split over several successive frames. The authors have shown that important performance gains can be obtained, without any additional cost, by optimizing the bus access scheme. The authors neither insist here on the relatively simple procedure for postprocessing of the schedule table, during which the table can be simplified for certain situations in which identical activation times are scheduled for a given process on different columns.
Q3. What is the important application of the scheduling algorithm?
One of the very important applications of their scheduling algorithm is for performance estimation during design space exploration.
Q4. How does TTP provide a global time-base without overhead?
By applying a clock synchronization algorithm, TTP provides a global time-base of known precision, without any overhead on the communication.
Q5. How many buses are used for broadcasting condition values?
For broadcasting of condition values, only buses are considered to which all processors are connected, and the authors assume that at least one such bus exists.
Q6. What is the way to calculate the execution times of a process?
For many applications, actual execution times of processes are depending on the current data and/or the internal state of the system.
Q7. What is the heuristic for scheduling a process graph?
List scheduling heuristics [29] are based on ordered lists from which processes are extracted to be scheduled at certain moments.
Q8. What is the mapping of processes to processors and buses?
The mapping of processes to processors and buses is given by a function , where is the set of processing elements. , whereis the set of programmable processors, is the set of dedicated hardware components, and is the set of allocated buses.
Q9. What is the architecture for a system with static nonpreemptive scheduling?
TTP is also perfectly suited for systems implemented with static nonpreemptive scheduling, and thus represents an ideal target architecture for the scheduling approach presented in the previous sections.
Q10. What is the algorithm for scheduling of process graphs?
The algorithm presented in Fig. 4 is able to schedule, based on a certain priority function, process graphs without conditional control dependencies.