Automatic parallelization of multirate block diagrams of control systems on multicore platforms

Cumhur Erkan Tuncali, Georgios Fainekos, Yann-Hang Lee

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

This article addresses the problem of parallelizing model block diagrams for real-time embedded applications on multicore architectures. We describe a Mixed Integer Linear Programming formulation for finding a feasible mapping of the blocks to different CPU cores. For single-rate models, we use an objective function that minimizes the overall worst-case execution time. We introduce a set of heuristics to solve the problem for large models in a reasonable time. For multirate models, we solve the feasibility problem for finding a valid mapping. We study the scalability and efficiency of our approach with synthetic benchmarks and an engine controller from Toyota.

Original languageEnglish (US)
Article number15
JournalACM Transactions on Embedded Computing Systems
Volume16
Issue number1
DOIs
StatePublished - Oct 1 2016

Fingerprint

Control systems
Linear programming
Program processors
Scalability
Engines
Controllers

Keywords

  • Embedded control systems
  • Model-based development
  • Multicore platforms
  • Multirate
  • Optimization
  • Scheduling
  • Simulink
  • Task allocation

ASJC Scopus subject areas

  • Software
  • Hardware and Architecture

Cite this

@article{88e2743b6c3146c89bc152452e8b3c76,
title = "Automatic parallelization of multirate block diagrams of control systems on multicore platforms",
abstract = "This article addresses the problem of parallelizing model block diagrams for real-time embedded applications on multicore architectures. We describe a Mixed Integer Linear Programming formulation for finding a feasible mapping of the blocks to different CPU cores. For single-rate models, we use an objective function that minimizes the overall worst-case execution time. We introduce a set of heuristics to solve the problem for large models in a reasonable time. For multirate models, we solve the feasibility problem for finding a valid mapping. We study the scalability and efficiency of our approach with synthetic benchmarks and an engine controller from Toyota.",
keywords = "Embedded control systems, Model-based development, Multicore platforms, Multirate, Optimization, Scheduling, Simulink, Task allocation",
author = "Tuncali, {Cumhur Erkan} and Georgios Fainekos and Yann-Hang Lee",
year = "2016",
month = "10",
day = "1",
doi = "10.1145/2950055",
language = "English (US)",
volume = "16",
journal = "ACM Transactions on Embedded Computing Systems",
issn = "1539-9087",
publisher = "Association for Computing Machinery (ACM)",
number = "1",

}

TY - JOUR

T1 - Automatic parallelization of multirate block diagrams of control systems on multicore platforms

AU - Tuncali, Cumhur Erkan

AU - Fainekos, Georgios

AU - Lee, Yann-Hang

PY - 2016/10/1

Y1 - 2016/10/1

N2 - This article addresses the problem of parallelizing model block diagrams for real-time embedded applications on multicore architectures. We describe a Mixed Integer Linear Programming formulation for finding a feasible mapping of the blocks to different CPU cores. For single-rate models, we use an objective function that minimizes the overall worst-case execution time. We introduce a set of heuristics to solve the problem for large models in a reasonable time. For multirate models, we solve the feasibility problem for finding a valid mapping. We study the scalability and efficiency of our approach with synthetic benchmarks and an engine controller from Toyota.

AB - This article addresses the problem of parallelizing model block diagrams for real-time embedded applications on multicore architectures. We describe a Mixed Integer Linear Programming formulation for finding a feasible mapping of the blocks to different CPU cores. For single-rate models, we use an objective function that minimizes the overall worst-case execution time. We introduce a set of heuristics to solve the problem for large models in a reasonable time. For multirate models, we solve the feasibility problem for finding a valid mapping. We study the scalability and efficiency of our approach with synthetic benchmarks and an engine controller from Toyota.

KW - Embedded control systems

KW - Model-based development

KW - Multicore platforms

KW - Multirate

KW - Optimization

KW - Scheduling

KW - Simulink

KW - Task allocation

UR - http://www.scopus.com/inward/record.url?scp=84992178569&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84992178569&partnerID=8YFLogxK

U2 - 10.1145/2950055

DO - 10.1145/2950055

M3 - Article

VL - 16

JO - ACM Transactions on Embedded Computing Systems

JF - ACM Transactions on Embedded Computing Systems

SN - 1539-9087

IS - 1

M1 - 15

ER -