MOE

A special-purpose parallel computer for high-speed, large-scale molecular orbital calculation

Koji Hashimoto, Hiroto Tomita, Inoue Koji, Katsuhiko Metsugi, Kazuaki Murakami, Shinjiro Inabata, So Yamada, Nobuaki Miyakawa, Hajime Takashima, Kunihiro Kitamura, Shigeru Obara, Takashi Amisaki, Kazutoshi Tanabe, Umpei Nagashima

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

3 Citations (Scopus)

Abstract

We are constructing a high-performance, special-purpose parallel machine for ab initio Molecular Orbital calculations, called MOE (Molecular Orbital calculation Engine). The sequential execution time is O(N4) where N is the number of basis functions, and most of time is spent to the calculations of electron repulsion integrals (ERIs). The calculation of ERIs have a lot of parallelism of O(N4), and therefore MOE tries to exploit the parallelism. This paper discuss the MOE architecture and examines important aspects of architecture design, which is required to calculate ERIs according to the "Obara method". We conclude that n-way parallelization is the most cost-effective, hence we designed the MOE prototype system with a host computer and many processing nodes. The processing node includes a 76 bit oating-point MULTIPLY-and-ADD unit and internal memory, etc., and it performs ERI computations efficiently. We estimate that the prototype system with 100 processing nodes calculate the energy of proteins in a few days.

Original languageEnglish
Title of host publicationACM/IEEE SC 1999 Conference, SC 1999
PublisherInstitute of Electrical and Electronics Engineers Inc.
Number of pages1
ISBN (Electronic)1581130910, 9781581130911
DOIs
Publication statusPublished - Jan 1 1999
Event1999 ACM/IEEE Conference on Supercomputing, SC 1999 - Portland, United States
Duration: Nov 13 1999Nov 19 1999

Publication series

NameACM/IEEE SC 1999 Conference, SC 1999

Other

Other1999 ACM/IEEE Conference on Supercomputing, SC 1999
CountryUnited States
CityPortland
Period11/13/9911/19/99

Fingerprint

Orbital calculations
Molecular orbitals
Engines
Electrons
Processing
Proteins
Data storage equipment
Costs

All Science Journal Classification (ASJC) codes

  • Hardware and Architecture

Cite this

Hashimoto, K., Tomita, H., Koji, I., Metsugi, K., Murakami, K., Inabata, S., ... Nagashima, U. (1999). MOE: A special-purpose parallel computer for high-speed, large-scale molecular orbital calculation. In ACM/IEEE SC 1999 Conference, SC 1999 [1592701] (ACM/IEEE SC 1999 Conference, SC 1999). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/SC.1999.10000

MOE : A special-purpose parallel computer for high-speed, large-scale molecular orbital calculation. / Hashimoto, Koji; Tomita, Hiroto; Koji, Inoue; Metsugi, Katsuhiko; Murakami, Kazuaki; Inabata, Shinjiro; Yamada, So; Miyakawa, Nobuaki; Takashima, Hajime; Kitamura, Kunihiro; Obara, Shigeru; Amisaki, Takashi; Tanabe, Kazutoshi; Nagashima, Umpei.

ACM/IEEE SC 1999 Conference, SC 1999. Institute of Electrical and Electronics Engineers Inc., 1999. 1592701 (ACM/IEEE SC 1999 Conference, SC 1999).

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

Hashimoto, K, Tomita, H, Koji, I, Metsugi, K, Murakami, K, Inabata, S, Yamada, S, Miyakawa, N, Takashima, H, Kitamura, K, Obara, S, Amisaki, T, Tanabe, K & Nagashima, U 1999, MOE: A special-purpose parallel computer for high-speed, large-scale molecular orbital calculation. in ACM/IEEE SC 1999 Conference, SC 1999., 1592701, ACM/IEEE SC 1999 Conference, SC 1999, Institute of Electrical and Electronics Engineers Inc., 1999 ACM/IEEE Conference on Supercomputing, SC 1999, Portland, United States, 11/13/99. https://doi.org/10.1109/SC.1999.10000
Hashimoto K, Tomita H, Koji I, Metsugi K, Murakami K, Inabata S et al. MOE: A special-purpose parallel computer for high-speed, large-scale molecular orbital calculation. In ACM/IEEE SC 1999 Conference, SC 1999. Institute of Electrical and Electronics Engineers Inc. 1999. 1592701. (ACM/IEEE SC 1999 Conference, SC 1999). https://doi.org/10.1109/SC.1999.10000
Hashimoto, Koji ; Tomita, Hiroto ; Koji, Inoue ; Metsugi, Katsuhiko ; Murakami, Kazuaki ; Inabata, Shinjiro ; Yamada, So ; Miyakawa, Nobuaki ; Takashima, Hajime ; Kitamura, Kunihiro ; Obara, Shigeru ; Amisaki, Takashi ; Tanabe, Kazutoshi ; Nagashima, Umpei. / MOE : A special-purpose parallel computer for high-speed, large-scale molecular orbital calculation. ACM/IEEE SC 1999 Conference, SC 1999. Institute of Electrical and Electronics Engineers Inc., 1999. (ACM/IEEE SC 1999 Conference, SC 1999).
@inproceedings{99960eea188a4d28a44d442ec007a6ac,
title = "MOE: A special-purpose parallel computer for high-speed, large-scale molecular orbital calculation",
abstract = "We are constructing a high-performance, special-purpose parallel machine for ab initio Molecular Orbital calculations, called MOE (Molecular Orbital calculation Engine). The sequential execution time is O(N4) where N is the number of basis functions, and most of time is spent to the calculations of electron repulsion integrals (ERIs). The calculation of ERIs have a lot of parallelism of O(N4), and therefore MOE tries to exploit the parallelism. This paper discuss the MOE architecture and examines important aspects of architecture design, which is required to calculate ERIs according to the {"}Obara method{"}. We conclude that n-way parallelization is the most cost-effective, hence we designed the MOE prototype system with a host computer and many processing nodes. The processing node includes a 76 bit oating-point MULTIPLY-and-ADD unit and internal memory, etc., and it performs ERI computations efficiently. We estimate that the prototype system with 100 processing nodes calculate the energy of proteins in a few days.",
author = "Koji Hashimoto and Hiroto Tomita and Inoue Koji and Katsuhiko Metsugi and Kazuaki Murakami and Shinjiro Inabata and So Yamada and Nobuaki Miyakawa and Hajime Takashima and Kunihiro Kitamura and Shigeru Obara and Takashi Amisaki and Kazutoshi Tanabe and Umpei Nagashima",
year = "1999",
month = "1",
day = "1",
doi = "10.1109/SC.1999.10000",
language = "English",
series = "ACM/IEEE SC 1999 Conference, SC 1999",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
booktitle = "ACM/IEEE SC 1999 Conference, SC 1999",
address = "United States",

}

TY - GEN

T1 - MOE

T2 - A special-purpose parallel computer for high-speed, large-scale molecular orbital calculation

AU - Hashimoto, Koji

AU - Tomita, Hiroto

AU - Koji, Inoue

AU - Metsugi, Katsuhiko

AU - Murakami, Kazuaki

AU - Inabata, Shinjiro

AU - Yamada, So

AU - Miyakawa, Nobuaki

AU - Takashima, Hajime

AU - Kitamura, Kunihiro

AU - Obara, Shigeru

AU - Amisaki, Takashi

AU - Tanabe, Kazutoshi

AU - Nagashima, Umpei

PY - 1999/1/1

Y1 - 1999/1/1

N2 - We are constructing a high-performance, special-purpose parallel machine for ab initio Molecular Orbital calculations, called MOE (Molecular Orbital calculation Engine). The sequential execution time is O(N4) where N is the number of basis functions, and most of time is spent to the calculations of electron repulsion integrals (ERIs). The calculation of ERIs have a lot of parallelism of O(N4), and therefore MOE tries to exploit the parallelism. This paper discuss the MOE architecture and examines important aspects of architecture design, which is required to calculate ERIs according to the "Obara method". We conclude that n-way parallelization is the most cost-effective, hence we designed the MOE prototype system with a host computer and many processing nodes. The processing node includes a 76 bit oating-point MULTIPLY-and-ADD unit and internal memory, etc., and it performs ERI computations efficiently. We estimate that the prototype system with 100 processing nodes calculate the energy of proteins in a few days.

AB - We are constructing a high-performance, special-purpose parallel machine for ab initio Molecular Orbital calculations, called MOE (Molecular Orbital calculation Engine). The sequential execution time is O(N4) where N is the number of basis functions, and most of time is spent to the calculations of electron repulsion integrals (ERIs). The calculation of ERIs have a lot of parallelism of O(N4), and therefore MOE tries to exploit the parallelism. This paper discuss the MOE architecture and examines important aspects of architecture design, which is required to calculate ERIs according to the "Obara method". We conclude that n-way parallelization is the most cost-effective, hence we designed the MOE prototype system with a host computer and many processing nodes. The processing node includes a 76 bit oating-point MULTIPLY-and-ADD unit and internal memory, etc., and it performs ERI computations efficiently. We estimate that the prototype system with 100 processing nodes calculate the energy of proteins in a few days.

UR - http://www.scopus.com/inward/record.url?scp=35348915106&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=35348915106&partnerID=8YFLogxK

U2 - 10.1109/SC.1999.10000

DO - 10.1109/SC.1999.10000

M3 - Conference contribution

T3 - ACM/IEEE SC 1999 Conference, SC 1999

BT - ACM/IEEE SC 1999 Conference, SC 1999

PB - Institute of Electrical and Electronics Engineers Inc.

ER -