gpu activities at fi muni and their results
play

GPU activities at FI MUNI and their results Ji Matela, Ji Filipovi - PowerPoint PPT Presentation

Dec 09, 2022 •460 likes •579 views

GPU activities at FI MUNI and their results Ji Matela, Ji Filipovi <matela@ics.muni.cz> , <fila@ics.muni.cz> Laboratory of Advanced Network Technologies MetaCentrum CESNET Grid Computing Seminar 2010 Praha,

  1. GPU activities at FI MUNI and their results Jiří Matela, Jiří Filipovič <matela@ics.muni.cz> , <fila@ics.muni.cz> Laboratory of Advanced Network Technologies MetaCentrum CESNET Grid Computing Seminar 2010 Praha, 2010–10–15 1/11

  2. AutoGrid • Potential maps generation for molecular docking • The most computationally expensive parts accelerated on GPU • CPU part analyzed and modified 2/11

  3. AutoGrid – Speedup Accelerated design shows speedup of up to 400 × 450 disntance-dependent constant 400 350 300 speedup 250 200 150 100 50 0 0 50 100 150 200 250 300 350 grid size 3/11

  4. Discrete Wavelet Transform (DWT) • Digital signal processing technique • Application in diverse areas • digital speech recognition • multi-resolution video processing • data compression 4/11

  5. DWT – Speedup Our GPU implementation shows 68 × speedup 5/11

  6. Real-Time Video and Fast Large-Scale Image Compression • Ongoing project • Real-time compression and transmission of video in HD post-HD resolutions • Fast compression of pathological images of resolutions in order of gigapixels • GPU acceleration of JPEG2000 6/11

  7. JPEG2000 compression process Data compression Context Arithmetic DWT Modeling Encoding 7/11

  8. Context-Modeling in JPEG2000 • Serial algorithm • Redesign to fit specifics of GPUs • 12 × faster compared to JasPer CPU implementation 8/11

  9. Soft tissues simulations • Haptic surgical simulators • Simulations modelled using Finite Element Method (FEM) • FEM discretizes the modeled object as a mesh of elements • Per element computation and system of equations solving • Per element computation is complex problem so that it needs to be decomposed into several GPU functions • Not easy to choose decomposition granularity • Manual development of as small functions as possible • Automatic fusion into lager functions 9/11

  10. Preliminary performance gain of fusion 80% gain compared to non fused approach GPU SMEM/GMEM GPU GMEM 10000 CPU 1 core 8000 thousands elements/s 6000 4000 2000 0 0 10000 20000 30000 40000 50000 60000 # elements 10/11

  11. Thank you for you attention! Q?/A! <matela@ics.muni.cz> , <fila@ics.muni.cz> http://www.sitola.cz/ 11/11

Download Presentation
Download Policy: The content available on the website is offered to you 'AS IS' for your personal information and use only. It cannot be commercialized, licensed, or distributed on other websites without prior consent from the author. To download a presentation, simply click this link. If you encounter any difficulties during the download process, it's possible that the publisher has removed the file from their server.

Recommend

Summary Summary of of UPV UPV activities activities and and results results Mara

Summary Summary of of UPV UPV activities activities and and results results Mara

ICT for EU-India Cross Cultural Dissemination Summary Summary of of UPV UPV activities activities and and results results Mara Alpuente OUTLINE General profile of our node: ELP and RFIA Overview of ELP activities

664 views • 50 slides
}w !&quot;#$%&amp;'()+,-./012345&lt;yA| Illustraons by Ji Franek. Results Comparison

}w !&quot;#$%&amp;'()+,-./012345&lt;yA| Illustraons by Ji Franek. Results Comparison

Math Indexer and Searcher under the Hood: History and Development of a Winning Strategy Michal Rika, Petr Sojka, Marn Lka Masaryk University, Faculty of Informacs, Brno, Czech Republic mruzicka@mail.muni.cz, sojka@fi.muni.cz,

711 views • 16 slides
Bond Issuance, Muni A Bond Issuance, Muni Advisor R visor Rule and le and Continuing Disclosure

Bond Issuance, Muni A Bond Issuance, Muni Advisor R visor Rule and le and Continuing Disclosure

Bond Issuance, Muni A Bond Issuance, Muni Advisor R visor Rule and le and Continuing Disclosure Continuing Disclosure Presenters: Lawrence Flynn Walter Goldsmith Topics for Discussion Continuing Disclosure - Background and post-MCDC

364 views • 20 slides
Muni Service Equity Strategy Muni Service Equity Strategy Update Sean Kennedy, Acting Deputy

Muni Service Equity Strategy Muni Service Equity Strategy Update Sean Kennedy, Acting Deputy

Muni Service Equity Strategy Muni Service Equity Strategy Update Sean Kennedy, Acting Deputy Director, Operations Support, SFMTA Transit Division SFCTA Board of Directors, July 9, 2019 SFMTA Transit Planning Priorities Equity Strategy lines

273 views • 15 slides
Timely Feedback in Unstructured Cybersecurity Exercises Jan Vykopal, Radek Olejek, Karolna

Timely Feedback in Unstructured Cybersecurity Exercises Jan Vykopal, Radek Olejek, Karolna

Timely Feedback in Unstructured Cybersecurity Exercises Jan Vykopal, Radek Olejek, Karolna Bursk, Kristna Zkopanov vykopal@ics.muni.cz, oslejsek@fi.muni.cz, { burska|zakopcanova } @mail.muni.cz February 22, 2018 Outline Cyber

284 views • 11 slides
HELCOM Muni report (2013) findings and recommendations Claus Boettcher Member of ad hoc working

HELCOM Muni report (2013) findings and recommendations Claus Boettcher Member of ad hoc working

HELCOM Muni report (2013) findings and recommendations Claus Boettcher Member of ad hoc working group HELCOM Muni, 2009 2013 The starting point app. 300.000 metric tons Chemical Munitions were confiscated in Germany in 1945

362 views • 15 slides
}w !&quot;#$%&amp;'()+,-./012345&lt;yA| Illustraons by Ji Franek. Movaon Math

}w !&quot;#$%&amp;'()+,-./012345&lt;yA| Illustraons by Ji Franek. Movaon Math

Maths Informaon Retrieval for Digital Libraries Michal Rika Masaryk University, Faculty of Informacs, Brno, Czech Republic mruzicka@mail.muni.cz https://mir.fi.muni.cz/ }w !&quot;#$%&amp;'()+,-./012345&lt;yA| Illustraons by

720 views • 55 slides
28 19 th Avenue Rapid Capital Project July 7, 2015 SFMTA Board of Directors ABOUT MUNI FORWARD

28 19 th Avenue Rapid Capital Project July 7, 2015 SFMTA Board of Directors ABOUT MUNI FORWARD

28 19 th Avenue Rapid Capital Project July 7, 2015 SFMTA Board of Directors ABOUT MUNI FORWARD Muni Forward is the SFMTAs initiative to make transit safer and more reliable. 2 EXISTING CONDITIONS Pedestrian safety Transit

398 views • 10 slides
Bi Biol oleachi eaching of of fly fly ash ash and and bot bottom om ash ash fr from om Muni

Bi Biol oleachi eaching of of fly fly ash ash and and bot bottom om ash ash fr from om Muni

Bi Biol oleachi eaching of of fly fly ash ash and and bot bottom om ash ash fr from om Muni Munici cipal pal Solid lid Wa Waste In Incin cineratio tion fo for met metal re reco cove very V. Funari, H. I. Gomes , M. Cappelletti, S.

161 views • 14 slides
Landscape Partnership Activities &amp; Results Quabbin to Cardigan Initiative: Massachusetts

Landscape Partnership Activities &amp; Results Quabbin to Cardigan Initiative: Massachusetts

Landscape Partnership Activities &amp; Results Quabbin to Cardigan Initiative: Massachusetts Focus Area Picture of MA Q2C Overview of Strategy and Activities Utilize some time-tested outreach events with a new sensitivity to the

270 views • 11 slides
Survey on collaborative TB-HIV activities in European countries Preliminary results Presented by

Survey on collaborative TB-HIV activities in European countries Preliminary results Presented by

Survey on collaborative TB-HIV activities in European countries Preliminary results Presented by Gerard de Vries, on behalf of the Wolfheze Working Group on Collaborative TB/HIV activities 31 May 2017 Powered by 42 (=76%) Total number of

886 views • 21 slides
Muni Service Equity Strategy Policy SFMTA Board of Directors May 6, 2014 Why develop a service

Muni Service Equity Strategy Policy SFMTA Board of Directors May 6, 2014 Why develop a service

Muni Service Equity Strategy Policy SFMTA Board of Directors May 6, 2014 Why develop a service equity strategy? SFMTA is committed to continually improving Muni service quality across SF and ensuring that service performs equally across

309 views • 15 slides
THE AUSTRIAN CASE STUDY E-Mobility in Austria and Results of the COMPETT project activities 11

THE AUSTRIAN CASE STUDY E-Mobility in Austria and Results of the COMPETT project activities 11

THE AUSTRIAN CASE STUDY E-Mobility in Austria and Results of the COMPETT project activities 11 June 2015 // Reinhard Jellinek INHALTSVERZEICHNIS 01 OVERVIEW ON E-MOBILITY IN AUSTRIA 02 COMPETT PROJECT ACTIVITIES SELECTED RESULTS 2 AUSTRIA

480 views • 28 slides
EUREKA ACTIVITIES DEPARTMENT Jason Green- Activities Director Cindy Hirsch- Activities

EUREKA ACTIVITIES DEPARTMENT Jason Green- Activities Director Cindy Hirsch- Activities

EUREKA ACTIVITIES DEPARTMENT Jason Green- Activities Director Cindy Hirsch- Activities Secretary Phone- 636-733-3137 Activities Standards MSHSAA- Missouri State High School Activities Association Sets minimum requirements for

587 views • 12 slides
Interest rates stochastic models Ioane Muni Toke Ecole Centrale Paris Option Math ematiques

Interest rates stochastic models Ioane Muni Toke Ecole Centrale Paris Option Math ematiques

Interest rates stochastic models Ioane Muni Toke Ecole Centrale Paris Option Math ematiques Appliqu ees Majeure Math ematiques Financi` eres December 2010 - January 2011 Ioane Muni Toke (ECP - OMA Finance) Interest rates stochastic

722 views • 61 slides
Ecien t relational learning from sparse data Lub o P op elnsk Kno wledge

Ecien t relational learning from sparse data Lub o P op elnsk Kno wledge

Ecien t relational learning from sparse data Lub o P op elnsk Kno wledge Disco v ery Group F acult y of Informatics, Masaryk Univ ersit y in Brno, Czec hia popel@fi.muni.cz http://www.fi.muni.cz/ kd

374 views • 11 slides
Multiresolution Algorithms for Sparse Matrix Representation By: Mario Barela Mentor: Professor

Multiresolution Algorithms for Sparse Matrix Representation By: Mario Barela Mentor: Professor

Multiresolution Algorithms for Sparse Matrix Representation By: Mario Barela Mentor: Professor Garcia-Cervera Big Picture What is it all about ? 1 4 3 7 8 6 Dense Matrix 2 5 9 0 4 0 0 0 0 Sparse Matrix 0 0 9 Importance?

473 views • 12 slides
Considerations for GPU SEE Testing Edward J. Wyrwas edward.j.wyrwas@nasa.gov 301-286-5213

Considerations for GPU SEE Testing Edward J. Wyrwas edward.j.wyrwas@nasa.gov 301-286-5213

Considerations for GPU SEE Testing Edward J. Wyrwas edward.j.wyrwas@nasa.gov 301-286-5213 Lentech, Inc. work performed in support of NEPP Acknowledgment: This work was sponsored by: NASA Electronic Parts and Packaging (NEPP) Program To be

403 views • 21 slides
Dr. Avidan Akerib, VP Associative Computing BU June 2nd 2019 2 CORPORATE SUMMARY FOUNDED IN

Dr. Avidan Akerib, VP Associative Computing BU June 2nd 2019 2 CORPORATE SUMMARY FOUNDED IN

Dr. Avidan Akerib, VP Associative Computing BU June 2nd 2019 2 CORPORATE SUMMARY FOUNDED IN 1995 1 PUBLIC COMPANY 2 Consistent profitability &amp; zero debt ~150 EMPLOYEES WORLDWIDE. 3 Design / R&amp;D in Sunnyvale, CA &amp; Israel;

513 views • 49 slides
In-memory processing of big data via succinct data structures Rajeev Raman University of

In-memory processing of big data via succinct data structures Rajeev Raman University of

In-memory processing of big data via succinct data structures Rajeev Raman University of Leicester SDP Workshop, University of Cambridge Introduction Succinct Data Structuring Succinct Tries Applications &amp; Libraries End Overview

321 views • 18 slides
Technology Update and Finances January 10, 2017 1 Presentation Outline Technology

Technology Update and Finances January 10, 2017 1 Presentation Outline Technology

Technology Update and Finances January 10, 2017 1 Presentation Outline Technology Strategic Direction Current Use in the Classrooms Key Technology Focus: 2014/15 2015/16 2016/17 Current District Standards

708 views • 33 slides
DES S-box 4 is not like the others Stian Fauskanger University of Bergen Department of

DES S-box 4 is not like the others Stian Fauskanger University of Bergen Department of

DES S-box 4 is not like the others DES S-box 4 is not like the others Stian Fauskanger University of Bergen Department of Informatics The Selmer Center October 14, 2014 DES S-box 4 is not like the others DES Data Encryption Standard Image

622 views • 14 slides
Sec ecur urity ity Fea eatures ures for or SSD Why Storage Security is Important ? Dilemma

Sec ecur urity ity Fea eatures ures for or SSD Why Storage Security is Important ? Dilemma

Sec ecur urity ity Fea eatures ures for or SSD Why Storage Security is Important ? Dilemma emmas ! A secret known by two is no longer a secret : https://www.symantec.com/about/newsroom/press-kits#

621 views • 15 slides
ITCC November 18,2015 ITD Room 438 Agenda 1:00 Update on EA Activity Jeff Quast 1:20 Update

ITCC November 18,2015 ITD Room 438 Agenda 1:00 Update on EA Activity Jeff Quast 1:20 Update

ITCC November 18,2015 ITD Room 438 Agenda 1:00 Update on EA Activity Jeff Quast 1:20 Update on ITD Activity Gary Vetter 1:40 NASCIO 2015 State CIO Survey Dan Sipes 2:00 LPO Directional Statement Justin Data 2:20 Security Positon Update

638 views • 14 slides

More recommend