A heuristic approach to schedule periodic real-time tasks on reconfigurable hardware

Abstract

This paper deals with scheduling periodic real-time tasks on reconfigurable hardware devices, such as FPGAs. Reconfigurable hardware devices are increasingly used in embedded systems. To utilize these devices also for systems with real-time constraints, predictable task scheduling is required. We formalize the periodic task scheduling problem and propose two preemptive scheduling algorithms. The first is an adaption of the well-known Earliest Deadline First (EDF) technique to the FPGA execution model. Although the algorithm reveals good scheduling performance, it lacks an efficient schedulability test and requires a high number of FPGA configurations. The second algorithm uses the concept of servers that reserve area and execution time for other tasks. Tasks are successively merged into servers, which are then scheduled sequentially. While this method is inferior to the EDF-based technique regarding schedulability, it comes with a fast schedulability test and greatly reduces the number of required FPGA configurations.

Partitioned scheduling of periodic real-time tasks onto reconfigurable hardware

“…We will work on heuristics and approximation algorithms in order to find partitioned schedules for larger task sets. Further, we plan to compare the partitioning approaches to other scheduling techniques, such as global EDF scheduling [7,6]. A more detailed modeling of the reconfiguration port and context save and restore mechanisms could possibly lead to improved scheduling guarantees.…”

“…In contrast, in a partitioned schedule all instances of a task execute on the same processor. While we studied non-partitioned scheduling in [7,6], this work applies the concept of a partitioned schedule to our reconfigurable hardware execution model:…”

Balancing perfectly periodic service schedules: An application from recycling and waste management

“…Danne and Platzner [7] showed that to analyze a system composed of nonpreemptive periodic tasks, it is enough to study its behavior for a time interval equal to the least common multiple. Kermia and Sorel [15] proposed a greedy heuristic to solve distributed real-time applications represented by systems of nonpreemptive and interdependent periodic tasks.…”

Improving utilization of reconfigurable resources using two-dimensional compaction

“…Here, we assume that tasks are independent, so we do not have onboard task communications. We have assumed independent tasks that don't require inter-task communication, this happens in many applications such as Discrete Wavelet Transform and MPEG Video Decoder [20].…”