Bluegene/L applications:: Parallelism on a massive scale

被引：2

作者：

De Supinski, Bronis R. ^{[1
]}

Schulz, Martin ^{[1
]}

Bulatov, Vasily V. ^{[1
]}

Cabot, William ^{[1
]}

Chan, Bor ^{[1
]}

Cook, Andrew W. ^{[1
]}

Draegerl, Erik W. ^{[1
]}

Glosli, James N. ^{[1
]}

Greenough, Jeffrey A. ^{[1
]}

Henderson, Keith ^{[1
]}

Kubota, Alison ^{[1
]}

Louis, Steve ^{[1
]}

Miller, Brian J. ^{[1
]}

Patel, Mehul V. ^{[1
]}

Spelce, Thomas E. ^{[1
]}

Streitz, Frederick H. ^{[1
]}

Williams, Peter L. ^{[1
]}

Yates, Robert K. ^{[1
]}

Yoo, Andy ^{[1
]}

Almasi, George

Bhanot, Gyan

Gara, Alan

Gunnels, John A.

Gupta, Manish

Moreira, Jose

Sexton, James

Walkup, Bob

Archer, Charles

Gygi, Francois ^{[2
]}

Germann, Timothy C. ^{[3
]}

Kadau, Kai ^{[3
]}

Lomdahl, Peter S. ^{[3
]}

Rendleman, Charles ^{[4
]}

Welcome, Michael L. ^{[4
]}

McLendon, William ^{[5
]}

Hendrickson, Bruce ^{[5
]}

Franchetti, Franz ^{[6
]}

Kraig, Stefan ^{[7
]}

Lorenz, Juergen ^{[7
]}

Ueberhuber, Christoph W. ^{[7
]}

Chow, Edmond

Catalyuerek, Uemit ^{[8
]}

机构：

[1] Lawrence Livermore Natl Lab, Livermore, CA 94551 USA

[2] Univ Calif Davis, Davis, CA 95616 USA

[3] Los Alamos Natl Lab, Los Alamos, NM USA

[4] Lawrence Berkeley Natl Lab, Berkeley, CA USA

[5] Sandia Natl Labs, Livermore, CA 94550 USA

[6] Carnegie Mellon Univ, Pittsburgh, PA 15213 USA

[7] Vienna Univ Technol, Vienna, Austria

[8] Ohio State Univ, Columbus, OH 43210 USA

来源：

INTERNATIONAL JOURNAL OF HIGH PERFORMANCE COMPUTING APPLICATIONS | 2008年 / 22卷 / 01期

关键词：

massively parallel architectures; BlueGene/L; application scalability; performance study and optimization;

D O I：

10.1177/1094342007085025

中图分类号：

TP3 [计算技术、计算机技术];

学科分类号：

0812 ;

摘要：

BlueGene/L (BG/L), developed through a partnership between IBM and Lawrence Livermore National Laboratory (LLNL), is currently the world's largest system both in terms of scale, with 131,072 processors, and absolute performance, with a peak rate of 367 Tflop/s. BG/L has led the last four Top500 lists with a Unpack rate of 280.6 Tflop/s for the full machine installed at LLNL and is expected to remain the fastest computer in the next few editions. However, the real value of a machine such as BG/L derives from the scientific breakthroughs that real applications can produce by successfully using its unprecedented scale and computational power. In this paper, we describe our experiences with eight large scale applications on BG/L from several application domains, ranging from molecular dynamics to dislocation dynamics and turbulence simulations to searches in semantic graphs. We also discuss the challenges we faced when scaling these codes and present several successful optimization techniques. All applications show excellent scaling behavior, even at very large processor counts, with one code even achieving a sustained performance of more than 100 Tflop/s, clearly demonstrating the real success of the BG/L design.

引用

页码：33 / 51

页数：19

共 50 条

[1] Massive -scale Streaming Analytics: Models, Parallelism, & Real -world Applications
Bader, David A.
[J]. SPAA'18: PROCEEDINGS OF THE 30TH ACM SYMPOSIUM ON PARALLELISM IN ALGORITHMS AND ARCHITECTURES, 2018, : 193 - 193
[2] The BlueGene/L supercomputer
Bhanot, G
Chen, D
Gara, A
Vranas, P
[J]. NUCLEAR PHYSICS B-PROCEEDINGS SUPPLEMENTS, 2003, 119 : 114 - 121
[3] A volumetric FFT for BlueGene/L
Eleftheriou, M
Moreira, JE
Fitch, BG
Germain, RS
[J]. HIGH PERFORMANCE COMPUTING - HIPC 2003, 2003, 2913 : 194 - 203
[4] The BlueGene/L project
Chen, D
[J]. COMPUTER SIMULATION STUDIES IN CONDENSED-MATTER PHYSICS XVI, 2006, 103 : 174 - 178
[5] CFD on the BlueGene/L supercomputer
Sugavanam, SA
[J]. PARALLEL COMPUTATIONAL FLUID DYNAMICS: ADVANCED NUMERICAL METHODS SOFTWARE AND APPLICATIONS, 2004, : 37 - 40
[6] QCD on the BlueGene/L supercomputer
Bhanot, G
Chen, D
Gara, A
Sexton, J
Vranas, P
[J]. NUCLEAR PHYSICS B-PROCEEDINGS SUPPLEMENTS, 2005, 140 : 823 - 825
[7] Implementing MPI on the BlueGene/L supercomputer
Almási, G
Archer, C
Castaños, JG
Erway, CC
Heidelberger, P
Martorell, X
Moreira, JE
Pinnow, K
Ratterman, J
Smeds, N
Steinmacher-burow, B
Gropp, W
Toonen, B
[J]. EURO-PAR 2004 PARALLEL PROCESSING, PROCEEDINGS, 2004, 3149 : 833 - 845
[8] Massive Atomics for Massive Parallelism on GPUs
Egielski, Ian
Huang, Jesse
Zhang, Eddy Z.
[J]. ACM SIGPLAN NOTICES, 2014, 49 (11) : 93 - 103
[9] ON THE HORIZON - MASSIVE PARALLELISM
RATTNER, J
[J]. ELECTRONIC DESIGN, 1990, 38 (16) : 58 - 59
[10] Job scheduling for the BlueGene/L system
Krevat, E
Castaños, JG
Moreira, JE
[J]. JOB SCHEDULING STRATEGIES FOR PARALLEL PROCESSING, 2002, 2537 : 38 - 54

← 1 2 3 4 5 →