期刊
IEEE TRANSACTIONS ON COMPUTERS
卷 70, 期 10, 页码 1747-1761出版社
IEEE COMPUTER SOC
DOI: 10.1109/TC.2020.3023169
关键词
Task analysis; Engines; Graphics processing units; Computational modeling; Computer architecture; Real-time systems; Hardware; Real-time systems; heterogeneous architectures; scheduling
This paper introduces the HPC-DAG model specifically designed for heterogeneous platforms in designing complex real-time applications. The model allows system designers to specify alternative implementations of software components and use conditional branches to model if-then-else statements. By proposing heuristic allocation algorithms, the approach aims to improve schedulability for latency-sensitive applications.
Recent commercial hardware platforms for embedded real-time systems feature heterogeneous processing units and computing accelerators on the same System-on-Chip. When designing complex real-time applications for such architectures, the designer is exposed to a number of difficult choices, like deciding on which compute engine to execute a certain task, or what degree of parallelism to adopt for a given function. To help the designer exploring the wide space of design choices and tune the scheduling parameters, we propose a novel real-time application model, called HPC-DAG (Heterogeneous Parallel Condition Directed Acyclic Graph Model), specifically conceived for heterogeneous platforms. An HPC-DAG allows the system designer to specify alternative implementations of a software component for different processing engines, as well as conditional branches to model if-then-else statements. We also propose a schedulability analysis for the HPC-DAG model and a set of heuristic allocation algorithms aimed at improving schedulability for latency sensitive applications. Our analysis takes into account the cost of preempting a task, which can be non-negligible on certain processors. We show the use of our approach on a realistic case study, and we demonstrate its effectiveness by comparing it with state-of-the-art algorithms previously proposed in literature.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据