A fast online task placement algorithm on 3D partially reconfigurable devices

Tingyu Zhou; Tieyuan Pan; Takahiro Watanabe

doi:10.1109/TENCON.2017.8227902

A fast online task placement algorithm on 3D partially reconfigurable devices

Tingyu Zhou, Tieyuan Pan, Takahiro Watanabe

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

2 Citations (Scopus)

Abstract

Three dimensional (3D) integration technology provides a great opportunity for reconfigurable device to improve performance, while the literatures for 3D reconfigurable device are lack of an efficient task scheduling and placement algorithm. In this paper, a Maximal Empty Cuboids (MECs) list is proposed to manage free space on 3D reconfigurable device. When a task is assigned or removed, a series of MECs are selected and cut into multiple Maximal Empty Rectangles (MERs) or MECs in the horizontal direction and stored in different layers. By updating the MERs in each layer and extending them from bottom to up layer, the updated MECs can be generated quickly. The experimental results verify the efficiency of proposed algorithm.

Original language	English
Title of host publication	TENCON 2017 - 2017 IEEE Region 10 Conference
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	427-432
Number of pages	6
Volume	2017-December
ISBN (Electronic)	9781509011339
DOIs	https://doi.org/10.1109/TENCON.2017.8227902
Publication status	Published - 2017 Dec 19
Event	2017 IEEE Region 10 Conference, TENCON 2017 - Penang, Malaysia Duration: 2017 Nov 5 → 2017 Nov 8

Other

Other	2017 IEEE Region 10 Conference, TENCON 2017
Country	Malaysia
City	Penang
Period	17/11/5 → 17/11/8

Fingerprint

ASJC Scopus subject areas

Computer Science Applications
Electrical and Electronic Engineering

Cite this

Zhou, T., Pan, T., & Watanabe, T. (2017). A fast online task placement algorithm on 3D partially reconfigurable devices. In TENCON 2017 - 2017 IEEE Region 10 Conference (Vol. 2017-December, pp. 427-432). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/TENCON.2017.8227902

@inproceedings{f662a9a6fcc94f7f91e4cb20d87095d7,

title = "A fast online task placement algorithm on 3D partially reconfigurable devices",

abstract = "Three dimensional (3D) integration technology provides a great opportunity for reconfigurable device to improve performance, while the literatures for 3D reconfigurable device are lack of an efficient task scheduling and placement algorithm. In this paper, a Maximal Empty Cuboids (MECs) list is proposed to manage free space on 3D reconfigurable device. When a task is assigned or removed, a series of MECs are selected and cut into multiple Maximal Empty Rectangles (MERs) or MECs in the horizontal direction and stored in different layers. By updating the MERs in each layer and extending them from bottom to up layer, the updated MECs can be generated quickly. The experimental results verify the efficiency of proposed algorithm.",

author = "Tingyu Zhou and Tieyuan Pan and Takahiro Watanabe",

year = "2017",

month = dec

day = "19",

doi = "10.1109/TENCON.2017.8227902",

language = "English",

volume = "2017-December",

pages = "427--432",

booktitle = "TENCON 2017 - 2017 IEEE Region 10 Conference",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

note = "2017 IEEE Region 10 Conference, TENCON 2017 ; Conference date: 05-11-2017 Through 08-11-2017",

}

TY - GEN

T1 - A fast online task placement algorithm on 3D partially reconfigurable devices

AU - Zhou, Tingyu

AU - Pan, Tieyuan

AU - Watanabe, Takahiro

PY - 2017/12/19

Y1 - 2017/12/19

N2 - Three dimensional (3D) integration technology provides a great opportunity for reconfigurable device to improve performance, while the literatures for 3D reconfigurable device are lack of an efficient task scheduling and placement algorithm. In this paper, a Maximal Empty Cuboids (MECs) list is proposed to manage free space on 3D reconfigurable device. When a task is assigned or removed, a series of MECs are selected and cut into multiple Maximal Empty Rectangles (MERs) or MECs in the horizontal direction and stored in different layers. By updating the MERs in each layer and extending them from bottom to up layer, the updated MECs can be generated quickly. The experimental results verify the efficiency of proposed algorithm.

AB - Three dimensional (3D) integration technology provides a great opportunity for reconfigurable device to improve performance, while the literatures for 3D reconfigurable device are lack of an efficient task scheduling and placement algorithm. In this paper, a Maximal Empty Cuboids (MECs) list is proposed to manage free space on 3D reconfigurable device. When a task is assigned or removed, a series of MECs are selected and cut into multiple Maximal Empty Rectangles (MERs) or MECs in the horizontal direction and stored in different layers. By updating the MERs in each layer and extending them from bottom to up layer, the updated MECs can be generated quickly. The experimental results verify the efficiency of proposed algorithm.

UR - http://www.scopus.com/inward/record.url?scp=85044228793&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85044228793&partnerID=8YFLogxK

U2 - 10.1109/TENCON.2017.8227902

DO - 10.1109/TENCON.2017.8227902

M3 - Conference contribution

AN - SCOPUS:85044228793

VL - 2017-December

SP - 427

EP - 432

BT - TENCON 2017 - 2017 IEEE Region 10 Conference

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2017 IEEE Region 10 Conference, TENCON 2017

Y2 - 5 November 2017 through 8 November 2017

ER -

Access to Document

10.1109/TENCON.2017.8227902