Unsteady turbomachinery computations using massively parallel platforms

Edwin van der Weide; Georgi Kalitzin; Jörg Schlüter; Juan J. Alonso

doi:10.2514/6.2006-421

Unsteady turbomachinery computations using massively parallel platforms

Edwin van der Weide^*, Georgi Kalitzin, Jörg Schlüter, Juan J. Alonso

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

124 Citations (Scopus)

Abstract

The CPU power offered by the latest generation of supercomputers is such that the unsteady simulation of the full wheel of a compressor or turbine is now within reach. This CPU power is mainly obtained by increasing the number of processors beyond several thousands, e.g. the BlueGene/L computer of Lawrence Livermore National Laboratory has approximately 130,000 processors. Consequently extreme care must be taken when the simulation codes are ported to these platforms. This paper discusses the computer scientific aspects of simulating unsteady turbomachinery flows on massively parallel systems when multi-block structured grids are used. Load balance, parallel IO and search algorithms are addressed, especially for the case where the number of processors is larger than the number of blocks, i.e. when blocks must be split during runtime. Preliminary results for cases with more than 200 million nodes running on 1,800 processors are presented.

Original language	English
Title of host publication	44th AIAA Aerospace Sciences Meeting, Reno, Nevada, 9-12 January 2006
Subtitle of host publication	Collection of Technical Papers
Publisher	American Institute of Aeronautics and Astronautics
Pages	4904-4919
Number of pages	16
Volume	7
ISBN (Electronic)	978-1-62410-039-0
ISBN (Print)	978-1-56347-807-9
DOIs	https://doi.org/10.2514/6.2006-421
Publication status	Published - 1 Dec 2006
Externally published	Yes
Event	44th AIAA Aerospace Sciences Meeting 2006 - Reno, United States Duration: 9 Jan 2006 → 12 Jan 2006 Conference number: 44

Conference

Conference	44th AIAA Aerospace Sciences Meeting 2006
Country/Territory	United States
City	Reno
Period	9/01/06 → 12/01/06

Access to Document

10.2514/6.2006-421

Cite this

@inproceedings{15ac3467b1d54a9e8539e58ae6ae6fdd,

title = "Unsteady turbomachinery computations using massively parallel platforms",

abstract = "The CPU power offered by the latest generation of supercomputers is such that the unsteady simulation of the full wheel of a compressor or turbine is now within reach. This CPU power is mainly obtained by increasing the number of processors beyond several thousands, e.g. the BlueGene/L computer of Lawrence Livermore National Laboratory has approximately 130,000 processors. Consequently extreme care must be taken when the simulation codes are ported to these platforms. This paper discusses the computer scientific aspects of simulating unsteady turbomachinery flows on massively parallel systems when multi-block structured grids are used. Load balance, parallel IO and search algorithms are addressed, especially for the case where the number of processors is larger than the number of blocks, i.e. when blocks must be split during runtime. Preliminary results for cases with more than 200 million nodes running on 1,800 processors are presented.",

author = "{van der Weide}, Edwin and Georgi Kalitzin and J{\"o}rg Schl{\"u}ter and Alonso, {Juan J.}",

year = "2006",

month = dec,

day = "1",

doi = "10.2514/6.2006-421",

language = "English",

isbn = "978-1-56347-807-9",

volume = "7",

pages = "4904--4919",

booktitle = "44th AIAA Aerospace Sciences Meeting, Reno, Nevada, 9-12 January 2006",

publisher = "American Institute of Aeronautics and Astronautics",

address = "United States",

note = "44th AIAA Aerospace Sciences Meeting 2006 ; Conference date: 09-01-2006 Through 12-01-2006",

}

van der Weide, E, Kalitzin, G, Schlüter, J & Alonso, JJ 2006, Unsteady turbomachinery computations using massively parallel platforms. in 44th AIAA Aerospace Sciences Meeting, Reno, Nevada, 9-12 January 2006: Collection of Technical Papers. vol. 7, American Institute of Aeronautics and Astronautics, pp. 4904-4919, 44th AIAA Aerospace Sciences Meeting 2006, Reno, Nevada, United States, 9/01/06. https://doi.org/10.2514/6.2006-421

Unsteady turbomachinery computations using massively parallel platforms. / van der Weide, Edwin; Kalitzin, Georgi; Schlüter, Jörg et al.
44th AIAA Aerospace Sciences Meeting, Reno, Nevada, 9-12 January 2006: Collection of Technical Papers. Vol. 7 American Institute of Aeronautics and Astronautics, 2006. p. 4904-4919.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Unsteady turbomachinery computations using massively parallel platforms

AU - van der Weide, Edwin

AU - Kalitzin, Georgi

AU - Schlüter, Jörg

AU - Alonso, Juan J.

N1 - Conference code: 44

PY - 2006/12/1

Y1 - 2006/12/1

N2 - The CPU power offered by the latest generation of supercomputers is such that the unsteady simulation of the full wheel of a compressor or turbine is now within reach. This CPU power is mainly obtained by increasing the number of processors beyond several thousands, e.g. the BlueGene/L computer of Lawrence Livermore National Laboratory has approximately 130,000 processors. Consequently extreme care must be taken when the simulation codes are ported to these platforms. This paper discusses the computer scientific aspects of simulating unsteady turbomachinery flows on massively parallel systems when multi-block structured grids are used. Load balance, parallel IO and search algorithms are addressed, especially for the case where the number of processors is larger than the number of blocks, i.e. when blocks must be split during runtime. Preliminary results for cases with more than 200 million nodes running on 1,800 processors are presented.

AB - The CPU power offered by the latest generation of supercomputers is such that the unsteady simulation of the full wheel of a compressor or turbine is now within reach. This CPU power is mainly obtained by increasing the number of processors beyond several thousands, e.g. the BlueGene/L computer of Lawrence Livermore National Laboratory has approximately 130,000 processors. Consequently extreme care must be taken when the simulation codes are ported to these platforms. This paper discusses the computer scientific aspects of simulating unsteady turbomachinery flows on massively parallel systems when multi-block structured grids are used. Load balance, parallel IO and search algorithms are addressed, especially for the case where the number of processors is larger than the number of blocks, i.e. when blocks must be split during runtime. Preliminary results for cases with more than 200 million nodes running on 1,800 processors are presented.

UR - http://www.scopus.com/inward/record.url?scp=34250899811&partnerID=8YFLogxK

U2 - 10.2514/6.2006-421

DO - 10.2514/6.2006-421

M3 - Conference contribution

AN - SCOPUS:34250899811

SN - 978-1-56347-807-9

VL - 7

SP - 4904

EP - 4919

BT - 44th AIAA Aerospace Sciences Meeting, Reno, Nevada, 9-12 January 2006

PB - American Institute of Aeronautics and Astronautics

T2 - 44th AIAA Aerospace Sciences Meeting 2006

Y2 - 9 January 2006 through 12 January 2006

ER -

Unsteady turbomachinery computations using massively parallel platforms

Abstract

Conference

Access to Document

Other files and links

Fingerprint

Cite this