<span style="" class="text typewriter">POET</span> (v0.1): speedup of many-core parallel reactive transport simulations with fast DHT lookups

<p>Coupled reactive transport simulations are extremely demanding in terms of required computational power, which hampers their application and leads to coarsened and oversimplified domains. The chemical sub-process represents the major bottleneck: its acceleration is an urgent challenge which...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	M. De Lucia, M. Kühn, A. Lindemann, M. Lübke, B. Schnor
Formato:	article
Lenguaje:	EN
Publicado:	Copernicus Publications 2021
Materias:	Geology QE1-996.5
Acceso en línea:	https://doaj.org/article/12d3a7e280804863b5f8e3093f6363f9
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:12d3a7e280804863b5f8e3093f6363f9
record_format	dspace
spelling	oai:doaj.org-article:12d3a7e280804863b5f8e3093f6363f92021-12-01T14:30:07Z<span style="" class="text typewriter">POET</span> (v0.1): speedup of many-core parallel reactive transport simulations with fast DHT lookups10.5194/gmd-14-7391-20211991-959X1991-9603https://doaj.org/article/12d3a7e280804863b5f8e3093f6363f92021-12-01T00:00:00Zhttps://gmd.copernicus.org/articles/14/7391/2021/gmd-14-7391-2021.pdfhttps://doaj.org/toc/1991-959Xhttps://doaj.org/toc/1991-9603<p>Coupled reactive transport simulations are extremely demanding in terms of required computational power, which hampers their application and leads to coarsened and oversimplified domains. The chemical sub-process represents the major bottleneck: its acceleration is an urgent challenge which gathers increasing interdisciplinary interest along with pressing requirements for subsurface utilization such as spent nuclear fuel storage, geothermal energy and CO<span class="inline-formula"><sub>2</sub></span> storage. In this context we developed <code>POET</code> (POtsdam rEactive Transport), a research parallel reactive transport simulator integrating algorithmic improvements which decisively speed up coupled simulations. In particular, <code>POET</code> is designed with a master/worker architecture, which ensures computational efficiency in both multicore and cluster compute environments. <code>POET</code> does not rely on contiguous grid partitions for the parallelization of chemistry but forms work packages composed of grid cells distant from each other. Such scattering prevents particularly expensive geochemical simulations, usually concentrated in the vicinity of a reactive front, from generating load imbalance between the available CPUs (central processing units), as is often the case with classical partitions. Furthermore, <code>POET</code> leverages an original implementation of the distributed hash table (DHT) mechanism to cache the results of geochemical simulations for further reuse in subsequent time steps during the coupled simulation. The caching is hence particularly advantageous for initially chemically homogeneous simulations and for smooth reaction fronts. We tune the rounding employed in the DHT on a 2D benchmark to validate the caching approach, and we evaluate the performance gain of <code>POET</code>'s master/worker architecture and the DHT speedup on a 3D benchmark comprising around 650 000 grid elements. The runtime for 200 coupling iterations, corresponding to 960 simulation days, reduced from about 24 h on 11 workers to 29 min on 719 workers. Activating the DHT reduces the runtime further to 2 h and 8 min respectively. Only with these kinds of reduced hardware requirements and computational costs is it possible to realistically perform the long-term complex reactive transport simulations, as well as perform the uncertainty analyses required by pressing societal challenges connected with subsurface utilization.</p>M. De LuciaM. KühnM. KühnA. LindemannM. LübkeM. LübkeB. SchnorCopernicus PublicationsarticleGeologyQE1-996.5ENGeoscientific Model Development, Vol 14, Pp 7391-7409 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Geology QE1-996.5
spellingShingle	Geology QE1-996.5 M. De Lucia M. Kühn M. Kühn A. Lindemann M. Lübke M. Lübke B. Schnor <span style="" class="text typewriter">POET</span> (v0.1): speedup of many-core parallel reactive transport simulations with fast DHT lookups
description	<p>Coupled reactive transport simulations are extremely demanding in terms of required computational power, which hampers their application and leads to coarsened and oversimplified domains. The chemical sub-process represents the major bottleneck: its acceleration is an urgent challenge which gathers increasing interdisciplinary interest along with pressing requirements for subsurface utilization such as spent nuclear fuel storage, geothermal energy and CO<span class="inline-formula"><sub>2</sub></span> storage. In this context we developed <code>POET</code> (POtsdam rEactive Transport), a research parallel reactive transport simulator integrating algorithmic improvements which decisively speed up coupled simulations. In particular, <code>POET</code> is designed with a master/worker architecture, which ensures computational efficiency in both multicore and cluster compute environments. <code>POET</code> does not rely on contiguous grid partitions for the parallelization of chemistry but forms work packages composed of grid cells distant from each other. Such scattering prevents particularly expensive geochemical simulations, usually concentrated in the vicinity of a reactive front, from generating load imbalance between the available CPUs (central processing units), as is often the case with classical partitions. Furthermore, <code>POET</code> leverages an original implementation of the distributed hash table (DHT) mechanism to cache the results of geochemical simulations for further reuse in subsequent time steps during the coupled simulation. The caching is hence particularly advantageous for initially chemically homogeneous simulations and for smooth reaction fronts. We tune the rounding employed in the DHT on a 2D benchmark to validate the caching approach, and we evaluate the performance gain of <code>POET</code>'s master/worker architecture and the DHT speedup on a 3D benchmark comprising around 650 000 grid elements. The runtime for 200 coupling iterations, corresponding to 960 simulation days, reduced from about 24 h on 11 workers to 29 min on 719 workers. Activating the DHT reduces the runtime further to 2 h and 8 min respectively. Only with these kinds of reduced hardware requirements and computational costs is it possible to realistically perform the long-term complex reactive transport simulations, as well as perform the uncertainty analyses required by pressing societal challenges connected with subsurface utilization.</p>
format	article
author	M. De Lucia M. Kühn M. Kühn A. Lindemann M. Lübke M. Lübke B. Schnor
author_facet	M. De Lucia M. Kühn M. Kühn A. Lindemann M. Lübke M. Lübke B. Schnor
author_sort	M. De Lucia
title	<span style="" class="text typewriter">POET</span> (v0.1): speedup of many-core parallel reactive transport simulations with fast DHT lookups
title_short	<span style="" class="text typewriter">POET</span> (v0.1): speedup of many-core parallel reactive transport simulations with fast DHT lookups
title_full	<span style="" class="text typewriter">POET</span> (v0.1): speedup of many-core parallel reactive transport simulations with fast DHT lookups
title_fullStr	<span style="" class="text typewriter">POET</span> (v0.1): speedup of many-core parallel reactive transport simulations with fast DHT lookups
title_full_unstemmed	<span style="" class="text typewriter">POET</span> (v0.1): speedup of many-core parallel reactive transport simulations with fast DHT lookups
title_sort	<span style="" class="text typewriter">poet</span> (v0.1): speedup of many-core parallel reactive transport simulations with fast dht lookups
publisher	Copernicus Publications
publishDate	2021
url	https://doaj.org/article/12d3a7e280804863b5f8e3093f6363f9
work_keys_str_mv	AT mdelucia spanstyleclasstexttypewriterpoetspanv01speedupofmanycoreparallelreactivetransportsimulationswithfastdhtlookups AT mkuhn spanstyleclasstexttypewriterpoetspanv01speedupofmanycoreparallelreactivetransportsimulationswithfastdhtlookups AT mkuhn spanstyleclasstexttypewriterpoetspanv01speedupofmanycoreparallelreactivetransportsimulationswithfastdhtlookups AT alindemann spanstyleclasstexttypewriterpoetspanv01speedupofmanycoreparallelreactivetransportsimulationswithfastdhtlookups AT mlubke spanstyleclasstexttypewriterpoetspanv01speedupofmanycoreparallelreactivetransportsimulationswithfastdhtlookups AT mlubke spanstyleclasstexttypewriterpoetspanv01speedupofmanycoreparallelreactivetransportsimulationswithfastdhtlookups AT bschnor spanstyleclasstexttypewriterpoetspanv01speedupofmanycoreparallelreactivetransportsimulationswithfastdhtlookups
_version_	1718405041077878784

<span style="" class="text typewriter">POET</span> (v0.1): speedup of many-core parallel reactive transport simulations with fast DHT lookups

Ejemplares similares