Qpace

Power-efficient parallel architecture based on IBM PowerXCell 8i

H. Baier, H. Boettiger, M. Drochner, N. Eicker, U. Fischer, Z. Fodor, A. Frommer, C. Gomez, G. Goldrian, S. Heybrock, D. Hierl, M. Hüsken, T. Huth, B. Krill, J. Lauritsen, T. Lippert, T. Maurer, B. Mendl, N. Meyer, A. Nobile & 15 others I. Ouda, M. Pivanti, D. Pleiter, M. Ries, A. Schäfer, H. Schick, F. Schifano, H. Simma, S. Solbrig, T. Streuer, K. H. Sulanke, R. Tripiccione, J. S. Vogt, T. Wettig, F. Winter

Research output: Contribution to journalArticle

18 Citations (Scopus)

Abstract

QPACE is a novel massively parallel architecture optimized for lattice QCD simulations. Each node comprises an IBM PowerXCell 8i processor. The nodes are interconnected by a custom 3-dimensional torus network implemented on an FPGA. The architecture was systematically optimized with respect to power consumption. This put QPACE in the number one spot on the Green500 List published in November 2009. In this paper we give an overview of the architecture and highlight the steps taken to improve power efficiency.

Original languageEnglish
Pages (from-to)149-154
Number of pages6
JournalComputer Science - Research and Development
Volume25
Issue number3-4
DOIs
Publication statusPublished - Sep 2010

Fingerprint

Parallel architectures
Field programmable gate arrays (FPGA)
Electric power utilization

Keywords

  • Parallel architectures
  • Special-purpose and application-based systems

ASJC Scopus subject areas

  • Computer Science(all)

Cite this

Qpace : Power-efficient parallel architecture based on IBM PowerXCell 8i. / Baier, H.; Boettiger, H.; Drochner, M.; Eicker, N.; Fischer, U.; Fodor, Z.; Frommer, A.; Gomez, C.; Goldrian, G.; Heybrock, S.; Hierl, D.; Hüsken, M.; Huth, T.; Krill, B.; Lauritsen, J.; Lippert, T.; Maurer, T.; Mendl, B.; Meyer, N.; Nobile, A.; Ouda, I.; Pivanti, M.; Pleiter, D.; Ries, M.; Schäfer, A.; Schick, H.; Schifano, F.; Simma, H.; Solbrig, S.; Streuer, T.; Sulanke, K. H.; Tripiccione, R.; Vogt, J. S.; Wettig, T.; Winter, F.

In: Computer Science - Research and Development, Vol. 25, No. 3-4, 09.2010, p. 149-154.

Research output: Contribution to journalArticle

Baier, H, Boettiger, H, Drochner, M, Eicker, N, Fischer, U, Fodor, Z, Frommer, A, Gomez, C, Goldrian, G, Heybrock, S, Hierl, D, Hüsken, M, Huth, T, Krill, B, Lauritsen, J, Lippert, T, Maurer, T, Mendl, B, Meyer, N, Nobile, A, Ouda, I, Pivanti, M, Pleiter, D, Ries, M, Schäfer, A, Schick, H, Schifano, F, Simma, H, Solbrig, S, Streuer, T, Sulanke, KH, Tripiccione, R, Vogt, JS, Wettig, T & Winter, F 2010, 'Qpace: Power-efficient parallel architecture based on IBM PowerXCell 8i', Computer Science - Research and Development, vol. 25, no. 3-4, pp. 149-154. https://doi.org/10.1007/s00450-010-0122-4
Baier, H. ; Boettiger, H. ; Drochner, M. ; Eicker, N. ; Fischer, U. ; Fodor, Z. ; Frommer, A. ; Gomez, C. ; Goldrian, G. ; Heybrock, S. ; Hierl, D. ; Hüsken, M. ; Huth, T. ; Krill, B. ; Lauritsen, J. ; Lippert, T. ; Maurer, T. ; Mendl, B. ; Meyer, N. ; Nobile, A. ; Ouda, I. ; Pivanti, M. ; Pleiter, D. ; Ries, M. ; Schäfer, A. ; Schick, H. ; Schifano, F. ; Simma, H. ; Solbrig, S. ; Streuer, T. ; Sulanke, K. H. ; Tripiccione, R. ; Vogt, J. S. ; Wettig, T. ; Winter, F. / Qpace : Power-efficient parallel architecture based on IBM PowerXCell 8i. In: Computer Science - Research and Development. 2010 ; Vol. 25, No. 3-4. pp. 149-154.
@article{9b12f9d620014ffdb5ab2fc0d29c9f10,
title = "Qpace: Power-efficient parallel architecture based on IBM PowerXCell 8i",
abstract = "QPACE is a novel massively parallel architecture optimized for lattice QCD simulations. Each node comprises an IBM PowerXCell 8i processor. The nodes are interconnected by a custom 3-dimensional torus network implemented on an FPGA. The architecture was systematically optimized with respect to power consumption. This put QPACE in the number one spot on the Green500 List published in November 2009. In this paper we give an overview of the architecture and highlight the steps taken to improve power efficiency.",
keywords = "Parallel architectures, Special-purpose and application-based systems",
author = "H. Baier and H. Boettiger and M. Drochner and N. Eicker and U. Fischer and Z. Fodor and A. Frommer and C. Gomez and G. Goldrian and S. Heybrock and D. Hierl and M. H{\"u}sken and T. Huth and B. Krill and J. Lauritsen and T. Lippert and T. Maurer and B. Mendl and N. Meyer and A. Nobile and I. Ouda and M. Pivanti and D. Pleiter and M. Ries and A. Sch{\"a}fer and H. Schick and F. Schifano and H. Simma and S. Solbrig and T. Streuer and Sulanke, {K. H.} and R. Tripiccione and Vogt, {J. S.} and T. Wettig and F. Winter",
year = "2010",
month = "9",
doi = "10.1007/s00450-010-0122-4",
language = "English",
volume = "25",
pages = "149--154",
journal = "Computer Science - Research and Development",
issn = "1865-2034",
publisher = "Springer Verlag",
number = "3-4",

}

TY - JOUR

T1 - Qpace

T2 - Power-efficient parallel architecture based on IBM PowerXCell 8i

AU - Baier, H.

AU - Boettiger, H.

AU - Drochner, M.

AU - Eicker, N.

AU - Fischer, U.

AU - Fodor, Z.

AU - Frommer, A.

AU - Gomez, C.

AU - Goldrian, G.

AU - Heybrock, S.

AU - Hierl, D.

AU - Hüsken, M.

AU - Huth, T.

AU - Krill, B.

AU - Lauritsen, J.

AU - Lippert, T.

AU - Maurer, T.

AU - Mendl, B.

AU - Meyer, N.

AU - Nobile, A.

AU - Ouda, I.

AU - Pivanti, M.

AU - Pleiter, D.

AU - Ries, M.

AU - Schäfer, A.

AU - Schick, H.

AU - Schifano, F.

AU - Simma, H.

AU - Solbrig, S.

AU - Streuer, T.

AU - Sulanke, K. H.

AU - Tripiccione, R.

AU - Vogt, J. S.

AU - Wettig, T.

AU - Winter, F.

PY - 2010/9

Y1 - 2010/9

N2 - QPACE is a novel massively parallel architecture optimized for lattice QCD simulations. Each node comprises an IBM PowerXCell 8i processor. The nodes are interconnected by a custom 3-dimensional torus network implemented on an FPGA. The architecture was systematically optimized with respect to power consumption. This put QPACE in the number one spot on the Green500 List published in November 2009. In this paper we give an overview of the architecture and highlight the steps taken to improve power efficiency.

AB - QPACE is a novel massively parallel architecture optimized for lattice QCD simulations. Each node comprises an IBM PowerXCell 8i processor. The nodes are interconnected by a custom 3-dimensional torus network implemented on an FPGA. The architecture was systematically optimized with respect to power consumption. This put QPACE in the number one spot on the Green500 List published in November 2009. In this paper we give an overview of the architecture and highlight the steps taken to improve power efficiency.

KW - Parallel architectures

KW - Special-purpose and application-based systems

UR - http://www.scopus.com/inward/record.url?scp=77958463950&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77958463950&partnerID=8YFLogxK

U2 - 10.1007/s00450-010-0122-4

DO - 10.1007/s00450-010-0122-4

M3 - Article

VL - 25

SP - 149

EP - 154

JO - Computer Science - Research and Development

JF - Computer Science - Research and Development

SN - 1865-2034

IS - 3-4

ER -