Work

Optimal scheduling algorithms for software-defined radio pipelined and replicated task chains on multicore architectures

Year: 2025

Type: article

Abstract: Software-Defined Radio (SDR) represents a move from dedicated hardwareto software implementations of digital communication standards.This approach offers flexibility, shorter time to market,maintainability, and lower costs, but it requires an optimizeddistribution tasks in order to meet performance requirements.Thus, we study the problem of scheduling SDR linear task chains of sateless and stateful tasks for streaming processing.We model this problem as a pipelined workflow scheduling problem basedon pipelined and replicated parallelism on homogeneous resources.We propose an optimal dynamic programming solution and an optimalgreedy algorithm named OTAC for maximizing throughput whilealso minimizing resource utilization.Moreover, the optimality of the proposed scheduling algorithm is proved.We evaluate our solutions and compare their execution times andschedules to other algorithms using synthetic task chains and animplementation of the DVB-S2 communication standard on the AFF3CT SDR Domain Specific Language.Our results demonstrate how OTAC quickly finds optimalschedules, leading consistently to better results than otheralgorithms, or equivalent results with fewer resources.

Source: Journal of Parallel and Distributed Computing

DOI: 10.1016/j.jpdc.2025.105106