Contribution to a conference proceedings/Contribution to a book

FZJ-2020-01507

http://join2-wiki.gsi.de/foswiki/pub/Main/Artwork/join2_logo100x88.png Sub-Grid Scale Modelling at Scale with Deep Learning and up to 60 Billion Degrees of Freedom

Bode, M.RWTH* ; Denker, D.RWTH* ; Jitsev, J. (Corresponding author)FZJ* ; Pitsch, H.RWTH*

2020
Forschungszentrum Jülich GmbH Zentralbibliothek, Verlag Jülich

Please use a persistent id in citations: http://hdl.handle.net/2128/24572

Abstract: This work presents fully resolved direct numerical simulations (DNSs) of a turbulent reactive planar temporally non-premixed jet configuration with up to 60 billion degrees of freedom. As scalar mixing is of utmost importance for this kind of configuration, a novel deep learning (DL) approach in the context of large-eddy simulation is presented which results in predictive mixing statistics on underresolved grids. The usability of the mixing model is approved by applying it to the DNS data. Furthermore, node performance measurements for the training of the DL networks are shown for different computing clusters.

---

Contributing Institute(s):

1. John von Neumann - Institut für Computing (NIC)
2. Jülich Supercomputing Center (JSC)

Research Program(s):

1. 512 - Data-Intensive Science and Federated Computing (POF3-512) (POF3-512)

Appears in the scientific report 2020

---

Database coverage:
;

---

---

Linked articles:

http://join2-wiki.gsi.de/foswiki/pub/Main/Artwork/join2_logo100x88.png Book/Proceedings Müller, M. (Editor)Extern* ; Binder, K. (Editor) ; Trautmann, A. (Editor)FZJ*
NIC Symposium 2020: proceedings
2020NIC Symposium, JülichJülich, Germany, 27 Feb 2020 - 28 Feb 20202020-02-272020-02-28 Jülich : Forschungszentrum Jülich GmbH Zentralbibliothek, Verlag, NIC Series 50, v, 424 S. (2020)2020   Files  Fulltext by OpenAccess repository BibTeX | EndNote: XML, Text | RIS

---