GPU performance analysis for viscoacoustic wave equations using fast stencil computation from the symbolic specification

Seismic forward modeling is a computationally and data-intensive stage in the seismic processing workflow. By profiling the kernels of seismic forward modeling algorithms, it was observed that they need to access a wide variety of memory locations, in addition to the computational cost of performing...

Full description

Saved in:
Bibliographic Details
Published in:The Journal of supercomputing 2023-08, Vol.79 (12), p.12853-12868
Main Authors: Jesus, Lauê, Nogueira, Peterson, Speglich, João, Boratto, Murilo
Format: Article
Language:English
Subjects:
Algorithms
Chips (memory devices)
Compilers
Computer memory
Computer Science
Environment models
Floating point arithmetic
Interpreters
Kernels
Mathematical models
Performance enhancement
Processor Architectures
Programming Languages
Wave equations
Workflow
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by
cites cdi_FETCH-LOGICAL-c270t-9bce3686dc36fe22ce5539c0cd7fa810d9c7ccab7a2c4841b79de2abe370c91f3
container_end_page 12868
container_issue 12
container_start_page 12853
container_title The Journal of supercomputing
container_volume 79
creator Jesus, Lauê
Nogueira, Peterson
Speglich, João
Boratto, Murilo
description Seismic forward modeling is a computationally and data-intensive stage in the seismic processing workflow. By profiling the kernels of seismic forward modeling algorithms, it was observed that they need to access a wide variety of memory locations, in addition to the computational cost of performing floating-point operations for the numerical solution of wave equations. In this context, the Roofline model was used to analyze six representative computing kernels in seismic modeling on GPU environment to indicate bottlenecks in the performance and suggest improvements of these wave equation propagators. Based on this, six viscoacoustic equations were implemented using the Devito tool. Experimental data have shown that optimizations in increasing data reuse and decreasing off-chip memory traffic can significantly improve performance.
doi_str_mv 10.1007/s11227-023-05178-3
format article
fullrecord <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_journals_2827962952</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2827962952</sourcerecordid><originalsourceid>FETCH-LOGICAL-c270t-9bce3686dc36fe22ce5539c0cd7fa810d9c7ccab7a2c4841b79de2abe370c91f3</originalsourceid><addsrcrecordid>eNp9kM1OwzAQhC0EEqXwApwscQ74p4mTI6qgICHBgZ4tZ2sXV0mcepNC3x7TInHjtNqdmZXmI-Sas1vOmLpDzoVQGRMyYzlXZSZPyITnKq2zcnZKJqwSLCvzmTgnF4gbxthMKjkhX4u3Je1tdCG2pgNLTWeaPXqk6UJ3HiEYCCMOHuin2Vlqt6MZfOiQjui7NXUGB4qD7cA3FELbj8NBpy6Glg4fluK-rUOT8thb8M7DQb8kZ840aK9-55QsHx_e50_Zy-vieX7_koFQbMiqGqwsymIFsnBWCLB5LitgsFLOlJytKlAAplZGQGrKa1WtrDC1lYpBxZ2ckpvj3z6G7Whx0JswxlQStSiFqgpR5SK5xNEFMSBG63QffWviXnOmfwjrI2GdCOsDYS1TSB5DmMzd2sa_1_-kvgG_NYJZ</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2827962952</pqid></control><display><type>article</type><title>GPU performance analysis for viscoacoustic wave equations using fast stencil computation from the symbolic specification</title><source>Springer Nature</source><creator>Jesus, Lauê ; Nogueira, Peterson ; Speglich, João ; Boratto, Murilo</creator><creatorcontrib>Jesus, Lauê ; Nogueira, Peterson ; Speglich, João ; Boratto, Murilo</creatorcontrib><description>Seismic forward modeling is a computationally and data-intensive stage in the seismic processing workflow. By profiling the kernels of seismic forward modeling algorithms, it was observed that they need to access a wide variety of memory locations, in addition to the computational cost of performing floating-point operations for the numerical solution of wave equations. In this context, the Roofline model was used to analyze six representative computing kernels in seismic modeling on GPU environment to indicate bottlenecks in the performance and suggest improvements of these wave equation propagators. Based on this, six viscoacoustic equations were implemented using the Devito tool. Experimental data have shown that optimizations in increasing data reuse and decreasing off-chip memory traffic can significantly improve performance.</description><identifier>ISSN: 0920-8542</identifier><identifier>EISSN: 1573-0484</identifier><identifier>DOI: 10.1007/s11227-023-05178-3</identifier><language>eng</language><publisher>New York: Springer US</publisher><subject>Algorithms ; Chips (memory devices) ; Compilers ; Computer memory ; Computer Science ; Environment models ; Floating point arithmetic ; Interpreters ; Kernels ; Mathematical models ; Performance enhancement ; Processor Architectures ; Programming Languages ; Wave equations ; Workflow</subject><ispartof>The Journal of supercomputing, 2023-08, Vol.79 (12), p.12853-12868</ispartof><rights>The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2023. Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><cites>FETCH-LOGICAL-c270t-9bce3686dc36fe22ce5539c0cd7fa810d9c7ccab7a2c4841b79de2abe370c91f3</cites></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>314,780,784,27923,27924</link.rule.ids></links><search><creatorcontrib>Jesus, Lauê</creatorcontrib><creatorcontrib>Nogueira, Peterson</creatorcontrib><creatorcontrib>Speglich, João</creatorcontrib><creatorcontrib>Boratto, Murilo</creatorcontrib><title>GPU performance analysis for viscoacoustic wave equations using fast stencil computation from the symbolic specification</title><title>The Journal of supercomputing</title><addtitle>J Supercomput</addtitle><description>Seismic forward modeling is a computationally and data-intensive stage in the seismic processing workflow. By profiling the kernels of seismic forward modeling algorithms, it was observed that they need to access a wide variety of memory locations, in addition to the computational cost of performing floating-point operations for the numerical solution of wave equations. In this context, the Roofline model was used to analyze six representative computing kernels in seismic modeling on GPU environment to indicate bottlenecks in the performance and suggest improvements of these wave equation propagators. Based on this, six viscoacoustic equations were implemented using the Devito tool. Experimental data have shown that optimizations in increasing data reuse and decreasing off-chip memory traffic can significantly improve performance.</description><subject>Algorithms</subject><subject>Chips (memory devices)</subject><subject>Compilers</subject><subject>Computer memory</subject><subject>Computer Science</subject><subject>Environment models</subject><subject>Floating point arithmetic</subject><subject>Interpreters</subject><subject>Kernels</subject><subject>Mathematical models</subject><subject>Performance enhancement</subject><subject>Processor Architectures</subject><subject>Programming Languages</subject><subject>Wave equations</subject><subject>Workflow</subject><issn>0920-8542</issn><issn>1573-0484</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><recordid>eNp9kM1OwzAQhC0EEqXwApwscQ74p4mTI6qgICHBgZ4tZ2sXV0mcepNC3x7TInHjtNqdmZXmI-Sas1vOmLpDzoVQGRMyYzlXZSZPyITnKq2zcnZKJqwSLCvzmTgnF4gbxthMKjkhX4u3Je1tdCG2pgNLTWeaPXqk6UJ3HiEYCCMOHuin2Vlqt6MZfOiQjui7NXUGB4qD7cA3FELbj8NBpy6Glg4fluK-rUOT8thb8M7DQb8kZ840aK9-55QsHx_e50_Zy-vieX7_koFQbMiqGqwsymIFsnBWCLB5LitgsFLOlJytKlAAplZGQGrKa1WtrDC1lYpBxZ2ckpvj3z6G7Whx0JswxlQStSiFqgpR5SK5xNEFMSBG63QffWviXnOmfwjrI2GdCOsDYS1TSB5DmMzd2sa_1_-kvgG_NYJZ</recordid><startdate>20230801</startdate><enddate>20230801</enddate><creator>Jesus, Lauê</creator><creator>Nogueira, Peterson</creator><creator>Speglich, João</creator><creator>Boratto, Murilo</creator><general>Springer US</general><general>Springer Nature B.V</general><scope>AAYXX</scope><scope>CITATION</scope></search><sort><creationdate>20230801</creationdate><title>GPU performance analysis for viscoacoustic wave equations using fast stencil computation from the symbolic specification</title><author>Jesus, Lauê ; Nogueira, Peterson ; Speglich, João ; Boratto, Murilo</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c270t-9bce3686dc36fe22ce5539c0cd7fa810d9c7ccab7a2c4841b79de2abe370c91f3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><topic>Algorithms</topic><topic>Chips (memory devices)</topic><topic>Compilers</topic><topic>Computer memory</topic><topic>Computer Science</topic><topic>Environment models</topic><topic>Floating point arithmetic</topic><topic>Interpreters</topic><topic>Kernels</topic><topic>Mathematical models</topic><topic>Performance enhancement</topic><topic>Processor Architectures</topic><topic>Programming Languages</topic><topic>Wave equations</topic><topic>Workflow</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Jesus, Lauê</creatorcontrib><creatorcontrib>Nogueira, Peterson</creatorcontrib><creatorcontrib>Speglich, João</creatorcontrib><creatorcontrib>Boratto, Murilo</creatorcontrib><collection>CrossRef</collection><jtitle>The Journal of supercomputing</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Jesus, Lauê</au><au>Nogueira, Peterson</au><au>Speglich, João</au><au>Boratto, Murilo</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>GPU performance analysis for viscoacoustic wave equations using fast stencil computation from the symbolic specification</atitle><jtitle>The Journal of supercomputing</jtitle><stitle>J Supercomput</stitle><date>2023-08-01</date><risdate>2023</risdate><volume>79</volume><issue>12</issue><spage>12853</spage><epage>12868</epage><pages>12853-12868</pages><issn>0920-8542</issn><eissn>1573-0484</eissn><abstract>Seismic forward modeling is a computationally and data-intensive stage in the seismic processing workflow. By profiling the kernels of seismic forward modeling algorithms, it was observed that they need to access a wide variety of memory locations, in addition to the computational cost of performing floating-point operations for the numerical solution of wave equations. In this context, the Roofline model was used to analyze six representative computing kernels in seismic modeling on GPU environment to indicate bottlenecks in the performance and suggest improvements of these wave equation propagators. Based on this, six viscoacoustic equations were implemented using the Devito tool. Experimental data have shown that optimizations in increasing data reuse and decreasing off-chip memory traffic can significantly improve performance.</abstract><cop>New York</cop><pub>Springer US</pub><doi>10.1007/s11227-023-05178-3</doi><tpages>16</tpages></addata></record>
fulltext fulltext
identifier ISSN: 0920-8542
ispartof The Journal of supercomputing, 2023-08, Vol.79 (12), p.12853-12868
issn 0920-8542
1573-0484
language eng
recordid cdi_proquest_journals_2827962952
source Springer Nature
subjects Algorithms
Chips (memory devices)
Compilers
Computer memory
Computer Science
Environment models
Floating point arithmetic
Interpreters
Kernels
Mathematical models
Performance enhancement
Processor Architectures
Programming Languages
Wave equations
Workflow
title GPU performance analysis for viscoacoustic wave equations using fast stencil computation from the symbolic specification
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-09T06%3A32%3A27IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=GPU%20performance%20analysis%20for%20viscoacoustic%20wave%20equations%20using%20fast%20stencil%20computation%20from%20the%20symbolic%20specification&rft.jtitle=The%20Journal%20of%20supercomputing&rft.au=Jesus,%20Lau%C3%AA&rft.date=2023-08-01&rft.volume=79&rft.issue=12&rft.spage=12853&rft.epage=12868&rft.pages=12853-12868&rft.issn=0920-8542&rft.eissn=1573-0484&rft_id=info:doi/10.1007/s11227-023-05178-3&rft_dat=%3Cproquest_cross%3E2827962952%3C/proquest_cross%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-c270t-9bce3686dc36fe22ce5539c0cd7fa810d9c7ccab7a2c4841b79de2abe370c91f3%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=2827962952&rft_id=info:pmid/&rfr_iscdi=true