IXPUG Russia Workshop and Tutorials, hosted by Saint-Petersburg Polytechnic University named after Peter the Great on June 9-10, 2016

Below is the agenda of the IXPUG Russia. Workshop and tutorials will be held at the Saint-Petersburg Polytechnic University.

IXPUG Keynotes Video

Schedule Overview

  • Thursday, June 9, Applications Day with users’ application stories, roundtable, training and youth session
  • Friday, June 10, Technologies Day with Intel keynotes, master-classes and training

IXPUG/RU Workshop and Tutorials Agenda

Applications Day (June 9)

 

Session

Presentation

Presenter

9:30-10:00

Registration, coffee

 

 

10:00-10:15

Opening

Welcome – IXPUG and SPbPU

Rector SPbPU

10:15-10:30

Keynote

KNC in RSC Petastream @ SPbPU

Alexander Moskovsky, RSC

10:30-10:50

Applications

GROMACS @ Xeon Phi

Alexey Shvetsov, SPbPU

10:50-11:10

Applications

Numerical weather prediction

Mikhail Tolstykh, INM RAS

11:10-11:30

Applications

GAMESS (US) @ Xeon Phi

Vladimir Mironov, MSU

11:30-11:50

Applications

Extra-large net DB indexation

Leonid Sokolinskiy, SUSU

11:50-12:10

Coffee break

 

 

12:10-12:30

Applications

Aeroacoustics

Vladimir Titarev, MIPT

12:30-12:50

Applications

SeisSol @ Xeon Phi

Carsten Uphoff, TUM

12:50-13:10

Applications

PICADOR, plasma physics

Iosif Meerov, NNSU

13:10-13:30

Applications

Molecular dynamics

Dmitry Nolde, IBCh RAS

13:30-13:50

Applications

Computational astrophysics

Igor Chernykh, ICMMG SBRAS

13:50-14:00

Call to action

Preparation to the next steps – KNL/Omni-Path

Nikolay Mester, Intel

14:00-15:00

Lunch

 

 

15:00-18:00

Training 1
(Computer class)

Programming on the Intel® Xeon Phi™: from simple to complex

  1. Programming coprocessor Intel Xeon Phi: from simple to complex
  2. Improving performance of the N-Body problem
  3. Intel®Xeon Phi™ Workshop Optimizations for 3-Dimensional Finite Difference (3DFD)
  4. Intel®Xeon Phi™ Workshop 3-Dimensional Finite Difference (3DFD) MPI code
  5. Intel®Xeon Phi™ Workshop Convolutional Neural Network

Singularis Lab LLC

15:00-18:00

Roundtable
(VIP-hall)

Preparing workloads for KNL

Nikolay Mester, Intel

15:00-18:00

Youth section
(Main hall)

Technology of parallel computing in modeling

Igor Odintsov

15:00-18:00

Working discussions
(Winter garden)

 

 

15:00-18:00

Visit to SPbPU’s Supercomputer

Visit supercomputer area

Director of SCC

16:25-16:35

Coffee-break

 

 

18:00-TBD

Move to the boat, buffet on the boat

 

 

 

Technologies Day (June 10)

 

Session

Presentation

Presenter

9:30-10:00

Registration, coffee

 

 

10:00-10:30

Keynote

Road to Many Core & the View Ahead

Joseph Curley, Intel

10:30-11:20

Keynote

New opportunities for developers on KNL/OmniPath. Detailed information for SW developers.

Christopher Dahnken, Intel

11:20-11:40

Coffee break

 

 

11:40-12:30

Keynote

Leveraging Intel MPI for best performance on Intel® Xeon Phi™ and Intel® OmniPath hardware

Dmitry Durnov, Intel

12:30-13:00

Keynote

Intel SW tools support for KNL/OmniPath

Dmitry Sivkov, Intel

13:00-14:00

Lunch

 

 

14:00-17:00

Training 2
(Computer class)

Advanced programming on the Intel® Xeon Phi™: from simple to complex

  1. Programming coprocessor Intel Xeon Phi: from simple to complex
  2. Improving performance of the N-Body problem
  3. Intel®Xeon Phi™ Workshop Optimizations for 3-Dimensional Finite Difference (3DFD)
  4. Intel®Xeon Phi™ Workshop 3-Dimensional Finite Difference (3DFD) MPI code
  5. Intel®Xeon Phi™ Workshop Convolutional Neural Network

Singularis Lab LLC

14:00-17:00

Master-class 1
(VIP-hall)

New Intel® Xeon Phi™ and Intel® Omni-Path Architecture provide developers an access to a new set of technologies.

Intel SW tools use HW awareness in order to make life of developers easier and get maximum performance from the platform.

During this master-class you will learn about the latest Intel® Xeon Phi™ and Intel® Omni-Path Architecture specific optimizations available in Intel MPI

Dmitry Durnov, Intel

14:00-17:00

Master-class 2
(Main hall)

Intel provides a wide spectrum of hardware and software tools for HPC and for developers. During this master-class Intel® Software tools will be overviewed for effective usage of modern Intel HW solutions for HPC

Dmitriy Sivkov, Intel

14:00-17:00

Working discussions
(Winter garden)

 

 

15:30-15:45

Coffee break

 

 

17:00-TBD

Buffet (Winter garden)

 

 

Disclaimer

Organizers reserve the right to:

  • amend the program;
  • refuse registration without stating reasons;
  • prohibit / restrict participation of a registered visitor under effect of alcohol and/or drugs;
  • prohibit /restrict participation of a registered visitor violating common behavior practices in public places and/or demonstrating disrespect to other visitors, organizers and/or sponsors of the conference.

This is a general information resource that does not constitute a public offer accordance with provisions of Article 437 (2) of the Civil Code of the Russian Federation.

Contacts

Please contact events@ixpug.ru for more information.

Location

St. Petersburg Polytechnic University named after Peter the Great (SPbPU), Scientific Research Building (NIK).
29, Polytekhnicheskaya street, St. Petersburg, (Polytekhnicheskaya subway station)

  • Carsten Uphoff
    TUM
    SeisSol @ Xeon Phi

    Carsten Uphoff received the M.Sc. degree in electrical engineering and information technology from the Technical University of Munich, Germany, in 2015. Since then, he is a research fellow in the group for algorithms and software for high performance computing at the department of informatics and he is currently pursuing the Ph.D. in computer science. He is one of the main developers of the earthquake simulator SeisSol.

  • Christopher Dahnken
    Intel Corp.
    New opportunities for developers on KNL/OmniPath. Detailed information for SW developers

    Position:

    Head of High Performance Computing Software Enabling in EMEA region, Intel Corporation

    Experience:

    Chris manages the software enabling team for High-Performance Computing in Europe, focusing on important scientific and industrial applications.

    Before his current role he held several HPC related engineering positions in Intel.

    Chris holds a Diploma and a PhD in theoretical physics from the University of Wuerzburg, Germany.

  • Joseph Curley
    Senior Director, Intel Corp.
    Road to Many Core & the View Ahead

    Joe Curley, Senior Director - HPC Platform and Ecosystem Enabling at Intel Corporation. Joe's primary responsibilities focus on establishing many core products and supporting development of the parallel computing ecosystem. Joe joined Intel in 2007 to manage planning and business activities that led up to the announcement of the Intel MIC Architecture in May of 2010, and launch of the first Intel Xeon Phi products in late 2012. Prior to joining Intel, Joe worked at Dell, Inc. leading the Dell Precision Workstation and the Dimension and XPS consumer / small business desktop product lines, as well as a series of engineering management roles. He began his career at computer graphics pioneer Tseng Labs.

  • Vladimir Titarev
    MIPT
    Aeroacoustics

    Владимир Александрович Титарев, к.ф.-м.н.
    Основное место работы: ведущий научный сотрудник ФИЦ “Информатика и управление” РАН.  Так же заместитель заведующего в лаборатории "Математическое моделирование нелинейных процессов в газовых средах" МФТИ (совместитель) и ведущий научный сотрудник в ФГУП “ЦАГИ” (совместитель в НИО-9)

    В октябре 2003 года защитил кандидатскую диссертацию по физ.-мат. наукам (руководитель – профессор, д.ф.-м.н. Е. М. Шахов).   Параллельно аспирантуре в России с 2001 по 2005 г. учился в аспирантуре Университета Тренто, Италия, и в октябре 2005 года получил степень Ph.D. по прикладной математике (руководитель – профессор Э. Ф. Торо).  C октября 2005 г. по январь 2007 года работал в должности Postdoctoral Research fellow в Университете Тренто, Италия. В марте 2007 перешел на работу в Cranfield University, Великобритания, в должности Research fellow, а с ноября 2008 г. по март 2011 г. – Research Fellow/Lecturer.   В марте 2011 перешел на работу в Вычислительный Центр в качестве ведущего научного сотрудника в сектор кинетической теории газов. Научные интересы включают в себя вычислительную механику жидкости и газа, высокопроизводительные вычисления, разработку методов повышенного порядка точности для решения системы гиперболических уравнений. 

    Рабочие странички:
    http://www.ccas.ru/personal/titarev/titarevr.htm
    http://flowmodellium.ru/

  • Dmitry Durnov
    Intel Corp.
    Leveraging Intel MPI for best performance on Intel® Xeon Phi™ and Intel® OmniPath hardware

    Области профессиональных интересов включают в себя высокопроизводительные вычисления, распределённые/телекоммуникационные системы, встроенные системы.
    Получил степень магистра по направлению электроника и микроэлектроника в Нижегородском Государственном Техническом Университете им. Р.Е. Алексеева.

  • Dmitry Sivkov
    Intel Corp.
    Intel SW tools support for KNL/OmniPath

    Инженер-консультант по кластерным инструментам

    Карьера: с 2000 года работал в Удмуртском Государственном университете  по направлению «Вычислительная математика», развивал использование высокопроизводительных вычислений и кластерных систем в УдГУ. 
    С 2011 года инженер-консультант Intel по кластерным инструментам.

    Образование: УдГУ, кандидат физико-математических наук по специальности «Дифференциальные уравнения».

  • Leonid Sokolinsky
    SUSU
    Extra-large net DB indexation

    Leonid Sokolinsky is the Vice-Rector for Informatization of South Ural State University (National Research University), doctor of science (phys.&math.), professor, honorary teacher of higher professional education of Russian Federation. Leonid Sokolinsky published more than 100 scientific papers. He is a member of the ACM since 1997. His research interests: parallel computing, database management systems, software for supercomputers, cloud computing, e-learning, mathematical programming.

    Homepage: http://sok.susu.ru/e_index.html.

  • Mikhail Tolstykh
    INM RAS
    Numerical weather prediction

    Tolstykh Mikhail, born in 1963. Graduated from Moscow Institute of Physics and Technology (MIPT), Dr. habil. in physics and mathematics. Lead researcher at the Institute of Numerical Mathematics Russian Academy of Sciences since 2003, head of the laboratory (part-time) at Hydrometcentre of Russia (since 2005), teacher at MIPT.

    Main developer of the SLAV atmosphere model which is used for operational medium-range weather prediction at Hydrometcentre of Russia. Currently, he works on model development targeting its application for climate change modeling.

  • Rostislav Fadeev
    INM RAS
    Numerical weather prediction

    Fadeev Rostislav, born in 1982. PhD in physics and mathematics, graduated from Moscow Institute of Physics and Technology (MIPT). Researcher at the Institute of Numerical Mathematics Russian Academy of Sciences (since 2009), part-time work at Hydrometcentre of Russia (since 2003), teacher at MIPT.

    He is one of the main developers of the SLAV atmosphere model which is used for operational medium-range weather prediction at Hydrometcentre of Russia. Currently, he works on the development of the non-hydrostatic  version for SLAV model, and also on the development of the SLAV version intended for seasonal forecast and climate change modeling.