Neural network based algorithm for the scheduling problem in high-level synthesis

Mehrdad Nourani, Christos Papachristou, Yoshiyasu Takefuji

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

4 Citations (Scopus)

Abstract

This paper presents a new scheduling approach for high-level synthesis based on a deterministic modified Hopfield model. Our model uses a four dimensional neural network architecture to schedule the operations of a data flow graph (DFG) and maps them to specific functional units. Neural Network-based Scheduling (NNS) is achieved by formulating the scheduling problem in terms of an energy function and by using the motion equation corresponding to the variation of energy. The algorithm searches the scheduling space in parallel and finds the optimal schedule. The main contribution of this work is an efficient parallel scheduling algorithm under time and resource constraints appropriate for implementing on a parallel machine. The algorithm is based on moves in the scheduling space, which correspond to moves towards the equilibrium point (lowest energy state) in the dynamic system space. Neurons' motion equation is the core of this guided movement mechanism and guarantees that the state of the system always converges to the lowest energy state.

Original languageEnglish
Title of host publicationEuropean Design Automation Conference
PublisherPubl by IEEE
Pages341-346
Number of pages6
ISBN (Print)0818627808
Publication statusPublished - 1992
Externally publishedYes
EventEuropean Design Automation Conference -EURO-VHDL '92 - Hamburg, Ger
Duration: 1992 Sep 71992 Sep 10

Other

OtherEuropean Design Automation Conference -EURO-VHDL '92
CityHamburg, Ger
Period92/9/792/9/10

Fingerprint

Scheduling
Neural networks
Electron energy levels
Equations of motion
Data flow graphs
Scheduling algorithms
Network architecture
Parallel algorithms
Neurons
Dynamical systems
High level synthesis

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Nourani, M., Papachristou, C., & Takefuji, Y. (1992). Neural network based algorithm for the scheduling problem in high-level synthesis. In European Design Automation Conference (pp. 341-346). Publ by IEEE.

Neural network based algorithm for the scheduling problem in high-level synthesis. / Nourani, Mehrdad; Papachristou, Christos; Takefuji, Yoshiyasu.

European Design Automation Conference. Publ by IEEE, 1992. p. 341-346.

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

Nourani, M, Papachristou, C & Takefuji, Y 1992, Neural network based algorithm for the scheduling problem in high-level synthesis. in European Design Automation Conference. Publ by IEEE, pp. 341-346, European Design Automation Conference -EURO-VHDL '92, Hamburg, Ger, 92/9/7.
Nourani M, Papachristou C, Takefuji Y. Neural network based algorithm for the scheduling problem in high-level synthesis. In European Design Automation Conference. Publ by IEEE. 1992. p. 341-346
Nourani, Mehrdad ; Papachristou, Christos ; Takefuji, Yoshiyasu. / Neural network based algorithm for the scheduling problem in high-level synthesis. European Design Automation Conference. Publ by IEEE, 1992. pp. 341-346
@inproceedings{d606baccbe48411088b31c48e7ed402b,
title = "Neural network based algorithm for the scheduling problem in high-level synthesis",
abstract = "This paper presents a new scheduling approach for high-level synthesis based on a deterministic modified Hopfield model. Our model uses a four dimensional neural network architecture to schedule the operations of a data flow graph (DFG) and maps them to specific functional units. Neural Network-based Scheduling (NNS) is achieved by formulating the scheduling problem in terms of an energy function and by using the motion equation corresponding to the variation of energy. The algorithm searches the scheduling space in parallel and finds the optimal schedule. The main contribution of this work is an efficient parallel scheduling algorithm under time and resource constraints appropriate for implementing on a parallel machine. The algorithm is based on moves in the scheduling space, which correspond to moves towards the equilibrium point (lowest energy state) in the dynamic system space. Neurons' motion equation is the core of this guided movement mechanism and guarantees that the state of the system always converges to the lowest energy state.",
author = "Mehrdad Nourani and Christos Papachristou and Yoshiyasu Takefuji",
year = "1992",
language = "English",
isbn = "0818627808",
pages = "341--346",
booktitle = "European Design Automation Conference",
publisher = "Publ by IEEE",

}

TY - GEN

T1 - Neural network based algorithm for the scheduling problem in high-level synthesis

AU - Nourani, Mehrdad

AU - Papachristou, Christos

AU - Takefuji, Yoshiyasu

PY - 1992

Y1 - 1992

N2 - This paper presents a new scheduling approach for high-level synthesis based on a deterministic modified Hopfield model. Our model uses a four dimensional neural network architecture to schedule the operations of a data flow graph (DFG) and maps them to specific functional units. Neural Network-based Scheduling (NNS) is achieved by formulating the scheduling problem in terms of an energy function and by using the motion equation corresponding to the variation of energy. The algorithm searches the scheduling space in parallel and finds the optimal schedule. The main contribution of this work is an efficient parallel scheduling algorithm under time and resource constraints appropriate for implementing on a parallel machine. The algorithm is based on moves in the scheduling space, which correspond to moves towards the equilibrium point (lowest energy state) in the dynamic system space. Neurons' motion equation is the core of this guided movement mechanism and guarantees that the state of the system always converges to the lowest energy state.

AB - This paper presents a new scheduling approach for high-level synthesis based on a deterministic modified Hopfield model. Our model uses a four dimensional neural network architecture to schedule the operations of a data flow graph (DFG) and maps them to specific functional units. Neural Network-based Scheduling (NNS) is achieved by formulating the scheduling problem in terms of an energy function and by using the motion equation corresponding to the variation of energy. The algorithm searches the scheduling space in parallel and finds the optimal schedule. The main contribution of this work is an efficient parallel scheduling algorithm under time and resource constraints appropriate for implementing on a parallel machine. The algorithm is based on moves in the scheduling space, which correspond to moves towards the equilibrium point (lowest energy state) in the dynamic system space. Neurons' motion equation is the core of this guided movement mechanism and guarantees that the state of the system always converges to the lowest energy state.

UR - http://www.scopus.com/inward/record.url?scp=0027005103&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0027005103&partnerID=8YFLogxK

M3 - Conference contribution

SN - 0818627808

SP - 341

EP - 346

BT - European Design Automation Conference

PB - Publ by IEEE

ER -