Power conscious fixed priority scheduling for hard real-time systems
Proceedings of the 36th annual ACM/IEEE Design Automation Conference
Task scheduling and voltage selection for energy minimization
Proceedings of the 39th annual Design Automation Conference
Energy Aware Scheduling for Distributed Real-Time Systems
IPDPS '03 Proceedings of the 17th International Symposium on Parallel and Distributed Processing
Power-Aware Scheduling for Periodic Real-Time Tasks
IEEE Transactions on Computers
Dynamic slack reclamation with procrastination scheduling in real-time embedded systems
Proceedings of the 42nd annual Design Automation Conference
Dynamic Algorithms for Energy Minimization on Parallel Machines
PDP '08 Proceedings of the 16th Euromicro Conference on Parallel, Distributed and Network-Based Processing (PDP 2008)
Hi-index | 0.00 |
Energy consumption is a critical issue in parallel and distributedsystems. Workflows consist of a number of tasks that need to be executed tocomplete an application. These tasks typically have precedence relationshipsthat have to be observed during execution for correctness. DAGs (DirectedAcyclic Graphs) can be used to represent many such workflows. The staticalgorithms to schedule for energy minimization under the deadline constraintsare based on estimating worst case execution time for each task to guaranteethat the application completes by a given deadline. During execution, manytasks may complete earlier than expected during the actual execution. Thisallows for adjusting the schedule for the tasks that have not yet begun executionto incorporate the extra slack. This has to be done with the dual goal ofreducing the energy requirements while still meeting the deadline constraints. Inthis paper, we present a novel dynamic algorithm for remapping tasks forenergy efficient scheduling of DAG based applications for DVS enabledsystems. Our experimental results show that the combination of our dynamicassignment and dynamic slack allocation leads to significantly better energyminimization compared to not changing the static schedule and/or onlyperforming dynamic slack allocation. Furthermore, its execution timerequirements are small enough to be useful for a large number of applications.