Exploration of schedule space by random walk

Liangwei Ge, Song Chen, Takeshi Yoshimura

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Scheduling, an important step in high-level synthesis, is essentially a searching process in the solution space. Due to the vastness of the solution space and the complexity of the imposed constraints, it is usually difficult to explore the solution space efficiently. In this paper, we present a random walk based perturbation method to explore the schedule space. The method works by limiting the search within a specifically defined sub-solution space (SSS), where schedules in the SSS can be found in polynomial time. Then, the SSS is repeatedly perturbed by using an N-dimension random walk so that better schedules can be searched in the new SSS. To improve the search efficiency, a guided perturbation strategy is presented that leads the random walk toward promising directions. Experiments on well-known benchmarks show that by controlling the number of perturbations, our method conveniently makes tradeoff between schedule quality and runtime. In reasonable runtime, the proposed method finds schedules of better quality than existing methods.

Original languageEnglish
Pages (from-to)30-42
Number of pages13
JournalIPSJ Transactions on System LSI Design Methodology
Volume2
DOIs
Publication statusPublished - 2009

Fingerprint

Scheduling
Polynomials
Experiments
High level synthesis

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Computer Science Applications

Cite this

Exploration of schedule space by random walk. / Ge, Liangwei; Chen, Song; Yoshimura, Takeshi.

In: IPSJ Transactions on System LSI Design Methodology, Vol. 2, 2009, p. 30-42.

Research output: Contribution to journalArticle

Ge, Liangwei ; Chen, Song ; Yoshimura, Takeshi. / Exploration of schedule space by random walk. In: IPSJ Transactions on System LSI Design Methodology. 2009 ; Vol. 2. pp. 30-42.
@article{204497ee1d20441b9edddf40dc8d53a9,
title = "Exploration of schedule space by random walk",
abstract = "Scheduling, an important step in high-level synthesis, is essentially a searching process in the solution space. Due to the vastness of the solution space and the complexity of the imposed constraints, it is usually difficult to explore the solution space efficiently. In this paper, we present a random walk based perturbation method to explore the schedule space. The method works by limiting the search within a specifically defined sub-solution space (SSS), where schedules in the SSS can be found in polynomial time. Then, the SSS is repeatedly perturbed by using an N-dimension random walk so that better schedules can be searched in the new SSS. To improve the search efficiency, a guided perturbation strategy is presented that leads the random walk toward promising directions. Experiments on well-known benchmarks show that by controlling the number of perturbations, our method conveniently makes tradeoff between schedule quality and runtime. In reasonable runtime, the proposed method finds schedules of better quality than existing methods.",
author = "Liangwei Ge and Song Chen and Takeshi Yoshimura",
year = "2009",
doi = "10.2197/ipsjtsldm.2.30",
language = "English",
volume = "2",
pages = "30--42",
journal = "IPSJ Transactions on System LSI Design Methodology",
issn = "1882-6687",
publisher = "Information Processing Society of Japan",

}

TY - JOUR

T1 - Exploration of schedule space by random walk

AU - Ge, Liangwei

AU - Chen, Song

AU - Yoshimura, Takeshi

PY - 2009

Y1 - 2009

N2 - Scheduling, an important step in high-level synthesis, is essentially a searching process in the solution space. Due to the vastness of the solution space and the complexity of the imposed constraints, it is usually difficult to explore the solution space efficiently. In this paper, we present a random walk based perturbation method to explore the schedule space. The method works by limiting the search within a specifically defined sub-solution space (SSS), where schedules in the SSS can be found in polynomial time. Then, the SSS is repeatedly perturbed by using an N-dimension random walk so that better schedules can be searched in the new SSS. To improve the search efficiency, a guided perturbation strategy is presented that leads the random walk toward promising directions. Experiments on well-known benchmarks show that by controlling the number of perturbations, our method conveniently makes tradeoff between schedule quality and runtime. In reasonable runtime, the proposed method finds schedules of better quality than existing methods.

AB - Scheduling, an important step in high-level synthesis, is essentially a searching process in the solution space. Due to the vastness of the solution space and the complexity of the imposed constraints, it is usually difficult to explore the solution space efficiently. In this paper, we present a random walk based perturbation method to explore the schedule space. The method works by limiting the search within a specifically defined sub-solution space (SSS), where schedules in the SSS can be found in polynomial time. Then, the SSS is repeatedly perturbed by using an N-dimension random walk so that better schedules can be searched in the new SSS. To improve the search efficiency, a guided perturbation strategy is presented that leads the random walk toward promising directions. Experiments on well-known benchmarks show that by controlling the number of perturbations, our method conveniently makes tradeoff between schedule quality and runtime. In reasonable runtime, the proposed method finds schedules of better quality than existing methods.

UR - http://www.scopus.com/inward/record.url?scp=79954543433&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79954543433&partnerID=8YFLogxK

U2 - 10.2197/ipsjtsldm.2.30

DO - 10.2197/ipsjtsldm.2.30

M3 - Article

VL - 2

SP - 30

EP - 42

JO - IPSJ Transactions on System LSI Design Methodology

JF - IPSJ Transactions on System LSI Design Methodology

SN - 1882-6687

ER -