Network simplex method based multiple voltage scheduling in power-efficient high-level synthesis

Cong Hao, Song Chen, Takeshi Yoshimura

Research output: Chapter in Book/Report/Conference proceedingConference contribution

4 Citations (Scopus)

Abstract

In this work, we focus on the problem of latency-constrained scheduling with consideration of multiple voltage technologies in High-level synthesis. Without the resource concern, we propose an Integer Linear Programming (ILP) formulation and further relax it to a piecewise Linear Programming (LP) problem, which is optimally solved using the efficient piecewise-linear extended network simplex method(PLNSM). The experimental results showed 80X+ speedup compared to the general LP formulation. Considering the resource usage, we propose a two-stage heuristic Network Simplex Method based Power-efficient Multiple Voltage Scheduling(NPMVS) method. Firstly, the above relaxed LP formulation is modified to perform mobility allocation and delay assignment for the operations so as to minimize the power and the differences between the allocated operation mobilities and the predefined target mobilities. The modified formulation is solved using the PLNSM and iteratively performed to minimize power and resource density variation in control steps by gradually updating the predefined target mobilities. Secondly, with the allocated operation mobilities, we apply dependency-free operation scheduling with the objective of minimizing the resource usage. Experimental results show that the proposed method can produce optimum solutions for all 6 benchmarks with 14 groups of data in a maximum time of 0.25 second.

Original languageEnglish
Title of host publicationProceedings of the Asia and South Pacific Design Automation Conference, ASP-DAC
Pages237-242
Number of pages6
DOIs
Publication statusPublished - 2013
Event2013 18th Asia and South Pacific Design Automation Conference, ASP-DAC 2013 - Yokohama
Duration: 2013 Jan 222013 Jan 25

Other

Other2013 18th Asia and South Pacific Design Automation Conference, ASP-DAC 2013
CityYokohama
Period13/1/2213/1/25

Fingerprint

Linear programming
Scheduling
Electric potential
High level synthesis

Keywords

  • High-level Synthesis
  • Scheduling

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Computer Science Applications
  • Computer Graphics and Computer-Aided Design

Cite this

Hao, C., Chen, S., & Yoshimura, T. (2013). Network simplex method based multiple voltage scheduling in power-efficient high-level synthesis. In Proceedings of the Asia and South Pacific Design Automation Conference, ASP-DAC (pp. 237-242). [6509602] https://doi.org/10.1109/ASPDAC.2013.6509602

Network simplex method based multiple voltage scheduling in power-efficient high-level synthesis. / Hao, Cong; Chen, Song; Yoshimura, Takeshi.

Proceedings of the Asia and South Pacific Design Automation Conference, ASP-DAC. 2013. p. 237-242 6509602.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Hao, C, Chen, S & Yoshimura, T 2013, Network simplex method based multiple voltage scheduling in power-efficient high-level synthesis. in Proceedings of the Asia and South Pacific Design Automation Conference, ASP-DAC., 6509602, pp. 237-242, 2013 18th Asia and South Pacific Design Automation Conference, ASP-DAC 2013, Yokohama, 13/1/22. https://doi.org/10.1109/ASPDAC.2013.6509602
Hao C, Chen S, Yoshimura T. Network simplex method based multiple voltage scheduling in power-efficient high-level synthesis. In Proceedings of the Asia and South Pacific Design Automation Conference, ASP-DAC. 2013. p. 237-242. 6509602 https://doi.org/10.1109/ASPDAC.2013.6509602
Hao, Cong ; Chen, Song ; Yoshimura, Takeshi. / Network simplex method based multiple voltage scheduling in power-efficient high-level synthesis. Proceedings of the Asia and South Pacific Design Automation Conference, ASP-DAC. 2013. pp. 237-242
@inproceedings{c24592f4891c4e2094366dbf2ae7ab16,
title = "Network simplex method based multiple voltage scheduling in power-efficient high-level synthesis",
abstract = "In this work, we focus on the problem of latency-constrained scheduling with consideration of multiple voltage technologies in High-level synthesis. Without the resource concern, we propose an Integer Linear Programming (ILP) formulation and further relax it to a piecewise Linear Programming (LP) problem, which is optimally solved using the efficient piecewise-linear extended network simplex method(PLNSM). The experimental results showed 80X+ speedup compared to the general LP formulation. Considering the resource usage, we propose a two-stage heuristic Network Simplex Method based Power-efficient Multiple Voltage Scheduling(NPMVS) method. Firstly, the above relaxed LP formulation is modified to perform mobility allocation and delay assignment for the operations so as to minimize the power and the differences between the allocated operation mobilities and the predefined target mobilities. The modified formulation is solved using the PLNSM and iteratively performed to minimize power and resource density variation in control steps by gradually updating the predefined target mobilities. Secondly, with the allocated operation mobilities, we apply dependency-free operation scheduling with the objective of minimizing the resource usage. Experimental results show that the proposed method can produce optimum solutions for all 6 benchmarks with 14 groups of data in a maximum time of 0.25 second.",
keywords = "High-level Synthesis, Scheduling",
author = "Cong Hao and Song Chen and Takeshi Yoshimura",
year = "2013",
doi = "10.1109/ASPDAC.2013.6509602",
language = "English",
isbn = "9781467330299",
pages = "237--242",
booktitle = "Proceedings of the Asia and South Pacific Design Automation Conference, ASP-DAC",

}

TY - GEN

T1 - Network simplex method based multiple voltage scheduling in power-efficient high-level synthesis

AU - Hao, Cong

AU - Chen, Song

AU - Yoshimura, Takeshi

PY - 2013

Y1 - 2013

N2 - In this work, we focus on the problem of latency-constrained scheduling with consideration of multiple voltage technologies in High-level synthesis. Without the resource concern, we propose an Integer Linear Programming (ILP) formulation and further relax it to a piecewise Linear Programming (LP) problem, which is optimally solved using the efficient piecewise-linear extended network simplex method(PLNSM). The experimental results showed 80X+ speedup compared to the general LP formulation. Considering the resource usage, we propose a two-stage heuristic Network Simplex Method based Power-efficient Multiple Voltage Scheduling(NPMVS) method. Firstly, the above relaxed LP formulation is modified to perform mobility allocation and delay assignment for the operations so as to minimize the power and the differences between the allocated operation mobilities and the predefined target mobilities. The modified formulation is solved using the PLNSM and iteratively performed to minimize power and resource density variation in control steps by gradually updating the predefined target mobilities. Secondly, with the allocated operation mobilities, we apply dependency-free operation scheduling with the objective of minimizing the resource usage. Experimental results show that the proposed method can produce optimum solutions for all 6 benchmarks with 14 groups of data in a maximum time of 0.25 second.

AB - In this work, we focus on the problem of latency-constrained scheduling with consideration of multiple voltage technologies in High-level synthesis. Without the resource concern, we propose an Integer Linear Programming (ILP) formulation and further relax it to a piecewise Linear Programming (LP) problem, which is optimally solved using the efficient piecewise-linear extended network simplex method(PLNSM). The experimental results showed 80X+ speedup compared to the general LP formulation. Considering the resource usage, we propose a two-stage heuristic Network Simplex Method based Power-efficient Multiple Voltage Scheduling(NPMVS) method. Firstly, the above relaxed LP formulation is modified to perform mobility allocation and delay assignment for the operations so as to minimize the power and the differences between the allocated operation mobilities and the predefined target mobilities. The modified formulation is solved using the PLNSM and iteratively performed to minimize power and resource density variation in control steps by gradually updating the predefined target mobilities. Secondly, with the allocated operation mobilities, we apply dependency-free operation scheduling with the objective of minimizing the resource usage. Experimental results show that the proposed method can produce optimum solutions for all 6 benchmarks with 14 groups of data in a maximum time of 0.25 second.

KW - High-level Synthesis

KW - Scheduling

UR - http://www.scopus.com/inward/record.url?scp=84877742260&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84877742260&partnerID=8YFLogxK

U2 - 10.1109/ASPDAC.2013.6509602

DO - 10.1109/ASPDAC.2013.6509602

M3 - Conference contribution

SN - 9781467330299

SP - 237

EP - 242

BT - Proceedings of the Asia and South Pacific Design Automation Conference, ASP-DAC

ER -