Parallel Scheduling of Multiple SDF Graphs Onto Heterogeneous Processors

Abstract

Parallel scheduling of multiple real-time applications onto heterogeneous processors is needed in the emerging embedded systems such as self-driving cars, smart cameras, and smartphones. Assuming that an embedded application is specified as a synchronous dataflow (SDF) graph or its extension, we propose a novel parallel scheduling methodology based on an evolutionary algorithm where the mapping of tasks onto processors is evolved to optimize a given objective function in an iterative fashion. In each iteration, we use an existing worst-case response time (WCRT) analysis tool to check if all applications satisfy their real-time requirements by translating each SDF graph into a directed acyclic graph (DAG) that is assumed in the WCRT analysis tool. Since the WCRT analysis must be performed in each iteration of evolution, we propose a clustering technique to reduce drastically the analysis time that depends on the number of nodes and their dependency. We formally prove that the proposed clustering technique does not change the estimated WCRT of each application. The effectiveness of the proposed scheduling methodology with the clustering technique is verified with extensive experiments using real-life benchmarks, randomly generated graphs, and the comparison with the existing technique.

A Novel Technique to Support Deep Learning Applications in a Model-Based Embedded Software Design Methodology

“…Recently, Jeong et al [25] proposed a methodology based on the genetic algorithm to explore the mappings for multiple dataflow graphs on CPU-GPU heterogeneous system with a deep learning application. In their work, they assume that the mapping for the deep learning application is fixed and find the mapping of dataflow graphs, aiming to reduce the worst-case response time for each dataflow application.…”

“…Since there is no previous work that tackles the same mapping problem, we devise a GA-based scheme by merging two recent previous works [14] and [25] and take it as a baseline technique to compare, denoted by Baseline in the experimental results. The former work is used to map multiple DL applications on the target board first, and the latter is used to map dataflow tasks after DL application mapping is completed.…”

Improved Scheduling Algorithm for Synchronous Data Flow Graphs on a Homogeneous Multi-Core Systems

“…Scheduling methods for SDFG on multi-core systems are classified into static scheduling [23,24] and self-time scheduling [25][26][27][28][29]. The self-time scheduling method studies SDFG, which completes scheduling based on actor readiness and completion times, and is a scheduling strategy with many real-time requirements.…”