Static scheduling techniques for dependent tasks on dynamically reconfigurable devices

Yang Qu (Corresponding Author), Juha-Pekka Soininen, Jari Nurmi

Research output: Contribution to journalArticleScientificpeer-review

25 Citations (Scopus)

Abstract

Dynamically reconfigurable hardware not only has high silicon reusability, but it can also deliver high performance for computation-intensive tasks. Advanced features such as run-time reconfiguration allow multiple tasks to be mapped onto the same device either simultaneously or multiplexed in time domain. These tasks need to be scheduled optimally or near optimally in order to efficiently utilize the device. It is a NP-hard problem, because task scheduling, allocation and configuration prefetching all need to be considered. In this paper, we target dependent task models and propose three static schedulers that use different problem solving strategies. The first is a heuristic approach developed from traditional list-based schedulers. It presents high efficiency but the least accuracy. The second is based on a full-domain search using constraint programming. It can guarantee to produce optimal solutions but requires significant searching effort. The last is a guided random search technique based on a genetic algorithm, which shows reasonable efficiency and much better accuracy than the heuristic approach.
Original languageEnglish
Pages (from-to)861-876
JournalJournal of Systems Architecture
Volume53
Issue number11
DOIs
Publication statusPublished - 2007
MoE publication typeA1 Journal article-refereed

Fingerprint

Scheduling
Reconfigurable hardware
Reusability
Computational complexity
Genetic algorithms
Silicon

Keywords

  • run-time reconfiguration
  • dynamic reconfigurable systems
  • dynamic reconfiguration
  • reconfiguration
  • dynamically reconfigurable hardware
  • task scheduling
  • genetic algorithm
  • algorithms

Cite this

@article{b421f0e241bd46e7826a685b0583ce2c,
title = "Static scheduling techniques for dependent tasks on dynamically reconfigurable devices",
abstract = "Dynamically reconfigurable hardware not only has high silicon reusability, but it can also deliver high performance for computation-intensive tasks. Advanced features such as run-time reconfiguration allow multiple tasks to be mapped onto the same device either simultaneously or multiplexed in time domain. These tasks need to be scheduled optimally or near optimally in order to efficiently utilize the device. It is a NP-hard problem, because task scheduling, allocation and configuration prefetching all need to be considered. In this paper, we target dependent task models and propose three static schedulers that use different problem solving strategies. The first is a heuristic approach developed from traditional list-based schedulers. It presents high efficiency but the least accuracy. The second is based on a full-domain search using constraint programming. It can guarantee to produce optimal solutions but requires significant searching effort. The last is a guided random search technique based on a genetic algorithm, which shows reasonable efficiency and much better accuracy than the heuristic approach.",
keywords = "run-time reconfiguration, dynamic reconfigurable systems, dynamic reconfiguration, reconfiguration, dynamically reconfigurable hardware, task scheduling, genetic algorithm, algorithms",
author = "Yang Qu and Juha-Pekka Soininen and Jari Nurmi",
year = "2007",
doi = "10.1016/j.sysarc.2007.02.004",
language = "English",
volume = "53",
pages = "861--876",
journal = "Journal of Systems Architecture",
issn = "1383-7621",
publisher = "Elsevier",
number = "11",

}

Static scheduling techniques for dependent tasks on dynamically reconfigurable devices. / Qu, Yang (Corresponding Author); Soininen, Juha-Pekka; Nurmi, Jari.

In: Journal of Systems Architecture, Vol. 53, No. 11, 2007, p. 861-876.

Research output: Contribution to journalArticleScientificpeer-review

TY - JOUR

T1 - Static scheduling techniques for dependent tasks on dynamically reconfigurable devices

AU - Qu, Yang

AU - Soininen, Juha-Pekka

AU - Nurmi, Jari

PY - 2007

Y1 - 2007

N2 - Dynamically reconfigurable hardware not only has high silicon reusability, but it can also deliver high performance for computation-intensive tasks. Advanced features such as run-time reconfiguration allow multiple tasks to be mapped onto the same device either simultaneously or multiplexed in time domain. These tasks need to be scheduled optimally or near optimally in order to efficiently utilize the device. It is a NP-hard problem, because task scheduling, allocation and configuration prefetching all need to be considered. In this paper, we target dependent task models and propose three static schedulers that use different problem solving strategies. The first is a heuristic approach developed from traditional list-based schedulers. It presents high efficiency but the least accuracy. The second is based on a full-domain search using constraint programming. It can guarantee to produce optimal solutions but requires significant searching effort. The last is a guided random search technique based on a genetic algorithm, which shows reasonable efficiency and much better accuracy than the heuristic approach.

AB - Dynamically reconfigurable hardware not only has high silicon reusability, but it can also deliver high performance for computation-intensive tasks. Advanced features such as run-time reconfiguration allow multiple tasks to be mapped onto the same device either simultaneously or multiplexed in time domain. These tasks need to be scheduled optimally or near optimally in order to efficiently utilize the device. It is a NP-hard problem, because task scheduling, allocation and configuration prefetching all need to be considered. In this paper, we target dependent task models and propose three static schedulers that use different problem solving strategies. The first is a heuristic approach developed from traditional list-based schedulers. It presents high efficiency but the least accuracy. The second is based on a full-domain search using constraint programming. It can guarantee to produce optimal solutions but requires significant searching effort. The last is a guided random search technique based on a genetic algorithm, which shows reasonable efficiency and much better accuracy than the heuristic approach.

KW - run-time reconfiguration

KW - dynamic reconfigurable systems

KW - dynamic reconfiguration

KW - reconfiguration

KW - dynamically reconfigurable hardware

KW - task scheduling

KW - genetic algorithm

KW - algorithms

U2 - 10.1016/j.sysarc.2007.02.004

DO - 10.1016/j.sysarc.2007.02.004

M3 - Article

VL - 53

SP - 861

EP - 876

JO - Journal of Systems Architecture

JF - Journal of Systems Architecture

SN - 1383-7621

IS - 11

ER -