Partition and place finite element model on wafer-scale engine

Jinwei Liu; Xiaopeng Zhang; Shiju Lin; Xinshi Zang; Jingsong Chen; Bentian Jiang; Martin D. F. Wong; Evangeline F. Y. Young

doi:10.1145/3489517.3530565

Partition and place finite element model on wafer-scale engine

Jinwei Liu, Xiaopeng Zhang, Shiju Lin, Xinshi Zang, Jingsong Chen, Bentian Jiang, Martin D. F. Wong, Evangeline F. Y. Young

Office of the Provost

Research output: Chapter in book/report/conference proceeding › Conference proceeding › peer-review

Abstract

The finite element method (FEM) is a well-known technique for approximately solving partial differential equations and it finds application in various engineering disciplines. The recently introduced wafer-scale engine (WSE) has shown the potential to accelerate FEM by up to 10,000×. However, accelerating FEM to the full potential of a WSE is non-trivial. Thus, in this work, we propose a partitioning algorithm to partition a 3D finite element model into tiles. The tiles can be thought of as a special netlist and are placed onto the 2D array of a WSE by our placement algorithm. Compared to the best-known approach, our partitioning has around 5% higher accuracy, and our placement algorithm can produce around 11% shorter wirelength (L1.5-normalized) on average.

Original language	English
Title of host publication	59th ACM/IEEE Design Automation Conference - Proceedings 2022
Publisher	Association for Computing Machinery (ACM)
Pages	631-636
Number of pages	6
ISBN (Print)	9781450391429
DOIs	https://doi.org/10.1145/3489517.3530565
Publication status	Published - 13 Jul 2022
Event	59th ACM/IEEE Design Automation Conference, DAC 2022 - San Francisco, United States Duration: 10 Jul 2022 → 14 Jul 2022 https://www.dac.com/About/Conference-Archive/59th-DAC-2022 (Conference website) https://www.dac.com/Portals/0/DAC%2059/59DAC%20Onsite%20Guide_v3.pdf?ver=GbBS5sBuhmEVJWVEz9CNIg%3d%3d (Conference programme) https://dl.acm.org/doi/proceedings/10.1145/3489517 (Conference proceedings)

Publication series

Name	ACM/IEEE Design Automation Conference - Proceedings
ISSN (Print)	0738-100X

Conference

Conference	59th ACM/IEEE Design Automation Conference, DAC 2022
Country/Territory	United States
City	San Francisco
Period	10/07/22 → 14/07/22
Internet address	https://www.dac.com/About/Conference-Archive/59th-DAC-2022 (Conference website) https://www.dac.com/Portals/0/DAC%2059/59DAC%20Onsite%20Guide_v3.pdf?ver=GbBS5sBuhmEVJWVEz9CNIg%3d%3d (Conference programme) https://dl.acm.org/doi/proceedings/10.1145/3489517 (Conference proceedings)

Scopus Subject Areas

Computer Science Applications
Control and Systems Engineering
Electrical and Electronic Engineering
Modelling and Simulation

Access to Document

10.1145/3489517.3530565

Cite this

Liu, J., Zhang, X., Lin, S., Zang, X., Chen, J., Jiang, B., Wong, M. D. F., & Young, E. F. Y. (2022). Partition and place finite element model on wafer-scale engine. In 59th ACM/IEEE Design Automation Conference - Proceedings 2022 (pp. 631-636). (ACM/IEEE Design Automation Conference - Proceedings). Association for Computing Machinery (ACM). https://doi.org/10.1145/3489517.3530565

@inproceedings{0afba56490ee442894b18e769f5c3d44,

title = "Partition and place finite element model on wafer-scale engine",

abstract = "The finite element method (FEM) is a well-known technique for approximately solving partial differential equations and it finds application in various engineering disciplines. The recently introduced wafer-scale engine (WSE) has shown the potential to accelerate FEM by up to 10,000×. However, accelerating FEM to the full potential of a WSE is non-trivial. Thus, in this work, we propose a partitioning algorithm to partition a 3D finite element model into tiles. The tiles can be thought of as a special netlist and are placed onto the 2D array of a WSE by our placement algorithm. Compared to the best-known approach, our partitioning has around 5% higher accuracy, and our placement algorithm can produce around 11% shorter wirelength (L1.5-normalized) on average.",

author = "Jinwei Liu and Xiaopeng Zhang and Shiju Lin and Xinshi Zang and Jingsong Chen and Bentian Jiang and Wong, {Martin D. F.} and Young, {Evangeline F. Y.}",

note = "Funding Information: The work described in this paper was partially supported by a grant from the Research Grants Council of the Hong Kong Special Administrative Region, China (Project No. CUHK 14209320). Publisher Copyright: {\textcopyright} 2022 ACM.; 59th ACM/IEEE Design Automation Conference, DAC 2022 ; Conference date: 10-07-2022 Through 14-07-2022",

year = "2022",

month = jul,

day = "13",

doi = "10.1145/3489517.3530565",

language = "English",

isbn = "9781450391429",

series = "ACM/IEEE Design Automation Conference - Proceedings",

publisher = "Association for Computing Machinery (ACM)",

pages = "631--636",

booktitle = "59th ACM/IEEE Design Automation Conference - Proceedings 2022",

address = "United States",

url = "https://www.dac.com/About/Conference-Archive/59th-DAC-2022, https://www.dac.com/Portals/0/DAC%2059/59DAC%20Onsite%20Guide_v3.pdf?ver=GbBS5sBuhmEVJWVEz9CNIg%3d%3d, https://dl.acm.org/doi/proceedings/10.1145/3489517",

}

Liu, J, Zhang, X, Lin, S, Zang, X, Chen, J, Jiang, B, Wong, MDF & Young, EFY 2022, Partition and place finite element model on wafer-scale engine. in 59th ACM/IEEE Design Automation Conference - Proceedings 2022. ACM/IEEE Design Automation Conference - Proceedings, Association for Computing Machinery (ACM), pp. 631-636, 59th ACM/IEEE Design Automation Conference, DAC 2022, San Francisco, California, United States, 10/07/22. https://doi.org/10.1145/3489517.3530565

Partition and place finite element model on wafer-scale engine. / Liu, Jinwei; Zhang, Xiaopeng; Lin, Shiju et al.
59th ACM/IEEE Design Automation Conference - Proceedings 2022. Association for Computing Machinery (ACM), 2022. p. 631-636 (ACM/IEEE Design Automation Conference - Proceedings).

Research output: Chapter in book/report/conference proceeding › Conference proceeding › peer-review

TY - GEN

T1 - Partition and place finite element model on wafer-scale engine

AU - Liu, Jinwei

AU - Zhang, Xiaopeng

AU - Lin, Shiju

AU - Zang, Xinshi

AU - Chen, Jingsong

AU - Jiang, Bentian

AU - Wong, Martin D. F.

AU - Young, Evangeline F. Y.

N1 - Funding Information: The work described in this paper was partially supported by a grant from the Research Grants Council of the Hong Kong Special Administrative Region, China (Project No. CUHK 14209320). Publisher Copyright: © 2022 ACM.

PY - 2022/7/13

Y1 - 2022/7/13

N2 - The finite element method (FEM) is a well-known technique for approximately solving partial differential equations and it finds application in various engineering disciplines. The recently introduced wafer-scale engine (WSE) has shown the potential to accelerate FEM by up to 10,000×. However, accelerating FEM to the full potential of a WSE is non-trivial. Thus, in this work, we propose a partitioning algorithm to partition a 3D finite element model into tiles. The tiles can be thought of as a special netlist and are placed onto the 2D array of a WSE by our placement algorithm. Compared to the best-known approach, our partitioning has around 5% higher accuracy, and our placement algorithm can produce around 11% shorter wirelength (L1.5-normalized) on average.

AB - The finite element method (FEM) is a well-known technique for approximately solving partial differential equations and it finds application in various engineering disciplines. The recently introduced wafer-scale engine (WSE) has shown the potential to accelerate FEM by up to 10,000×. However, accelerating FEM to the full potential of a WSE is non-trivial. Thus, in this work, we propose a partitioning algorithm to partition a 3D finite element model into tiles. The tiles can be thought of as a special netlist and are placed onto the 2D array of a WSE by our placement algorithm. Compared to the best-known approach, our partitioning has around 5% higher accuracy, and our placement algorithm can produce around 11% shorter wirelength (L1.5-normalized) on average.

UR - https://dl.acm.org/doi/10.1145/3489517.3530565

UR - http://www.scopus.com/inward/record.url?scp=85137501966&partnerID=8YFLogxK

U2 - 10.1145/3489517.3530565

DO - 10.1145/3489517.3530565

M3 - Conference proceeding

AN - SCOPUS:85137501966

SN - 9781450391429

T3 - ACM/IEEE Design Automation Conference - Proceedings

SP - 631

EP - 636

BT - 59th ACM/IEEE Design Automation Conference - Proceedings 2022

PB - Association for Computing Machinery (ACM)

T2 - 59th ACM/IEEE Design Automation Conference, DAC 2022

Y2 - 10 July 2022 through 14 July 2022

ER -

Partition and place finite element model on wafer-scale engine

Abstract

Publication series

Conference

Scopus Subject Areas

Access to Document

Other files and links

Fingerprint

Cite this