Scientific computing, Big data analytics and machine learning technology for megascience projects

Europe/Moscow
LIT JINR

LIT JINR

Description
International school  “Scientific computing, Big data analytics and machine learning technology for megascience projects” in frames of the Conference "Distributed Computing and Grid-technologies in Science and Education" will be held on September, 10 – 14, 2018 at the Laboratory of Information Technologies (LIT) of the Joint Institute for Nuclear Research (JINR), Dubna.

The registration for the school will take place on September 9 from 17.00 to 20.00 in the hotel "Dubna" (Vekslera str. 8).
During the registration, you will receive your laboratory permit to enter the JINR territory where the school will take place at the conference hall of the laboratory of information technologis.
In case you are not able to register on September 9, please pick up your laboratory permit at JINR Visit Centre (Molodezhnaya str. 5) from 09.00 t0 18:00 except lunch break from 13:00 to 14:00. 
For more information, please follow the link http://www.jinr.ru/about-en/visit-centre/
Please be aware that it is utterly important to return your laboratory permit back to the JINR Security Department. Don’t forget to drop it in the mailbox of JINR Visit Centre before your departure.


The goal of the school is to attract young scientists, students and postgraduate students to solve IT-tasks and challenges related to various aspects of megaprojects in the field of high-energy physics and to familiarize participants with modern methods of Big data analytics, machine learning and high-performance computing systems and to use this knowledge to solve IT tasks in the field of high-energy physics.
 

The main topics of the school are:
  • Big data by the example of the NICA megaproject and experiments at the Large Hadron Collider as the main source of big data in high-energy physics;
  • Distributed systems for collecting, processing, managing and storing information;
  • Use of high-performance systems (supercomputers, computing clusters) for data processing and modeling of physical experiments;
  • Machine learning. 

Working languages: Russian and English.

Selection of participants takes place in universities.
Please DO NOT register at the GRID 2018 conference website (GRID 2018 conference website is for those participants who make a report at the conference).

Organizers provide accommodation and per diem (at a rate of 500 rub/day.)



Контакты
    • 08:00 12:00
      Plenary Reports Conference Hall

      Conference Hall

      LIT JINR

    • 12:00 13:30
      Lunch 1h 30m
    • 13:30 14:15
      HTC and HPC as a parts of distributed computing system 45m 407

      407

      LIT JINR

      -How to combine two different computing paradigm into one distributed computing system -How to join GRID to Supercomputers
      Speaker: Mr Danila Oleynik (JINR LIT)
      Slides
    • 14:15 15:00
      COMPASS Production System 45m 407

      407

      LIT JINR

      Speaker: Mr Artem Petrosyan (JINR)
      Slides
    • 15:00 15:20
      Coffee 20m
    • 15:20 16:10
      DIRAC Interware for building distributed computing systems for scientific applications 50m 407

      407

      LIT JINR

      Modern scientific applications require very large computing and storage capacity to treat steadily increasing data volumes with more and more sophisticated algorithms. In many cases, in particular, for large international scientific collaborations, such capacity cannot be provided by just one even very large computing center. Instead, all the computing and storage resources available for a scientific community must be integrated in a coherent distributed system that can be seen and used by the community members as a single easy to access computer. Therefore, multiple projects were carried out recently to provide distributed computing infrastructures to multiple scientific communities. Among such infrastructures one can mention WLCG Computing Grid for the LHC program or European Open Science Cloud for European research projects. The DIRAC platform provides all the necessary software components for building such distributed computing infrastructures. In the lecture, the example of the DIRAC interware will be used to describe architectures and technologies applied for building complex distributed computing infrastructures involving various types of resources such as computing grids, clouds, supercomputers or volunteer computing platforms based on the BOINC technology. Integration of data stores with different storage technologies will be discussed. Examples of usage of DIRAC based infrastructures for High Energy Physics and other Big Data scientific applications will be presented.
      Speaker: Dr Andrei Tsaregorodtsev (CPPM-IN2P3-CNRS)
    • 16:10 17:00
      Poster Session
    • 17:00 18:30
      Welcome Party 1h 30m
    • 08:00 12:30
      Plenary Reports
    • 12:30 13:30
      Lunch 1h
    • 13:30 15:00
      Data and Workload Management Systems in the High Energy Physics Experiments 1h 30m 407

      407

      LIT JINR

      Speaker: Dr Alexei Klimentov (Brookhaven National Lab)
      Slides
    • 15:00 15:30
      Coffee 30m
    • 15:30 17:00
      Introduction to Big Data analytics 1h 30m 407

      407

      LIT JINR

      Speaker: Sergey Belov (Joint Institute for Nuclear Research)
    • 08:00 11:00
      Plenary Reports
    • 11:00 17:00
      Boad and Picnic party 6h
    • 08:00 12:30
      Plenary Reports Conference Hall

      Conference Hall

      LIT JINR

    • 12:30 13:30
      Lunch 1h
    • 13:30 15:00
      Brief introduction to deep learning 1h 30m 407

      407

      LIT JINR

      Tutorial - Deep learning for beginners with exciting examples of computing poetry
      Speakers: Prof. Gennady Ososkov (Joint Institute for Nuclear Research), Mr Pavel Goncharov (Sukhoi State Technical University of Gomel, Gomel, Belarus)
    • 15:00 15:30
      Coffee 30m
    • 15:30 17:00
      Нейросети в задачах распознавания образов с Intel® Optimized TensorFlow* with Keras. 1h 30m 407

      407

      LIT JINR

      Speaker: Ms Надежда Плотникова (Intel)
      notes
      Slides
    • 09:00 10:00
      Plenary Reports
    • 10:00 10:30
      Coffee 30m
    • 10:30 12:00
      ROOT Tutorial 1h 30m 407

      407

      LIT JINR

      Speaker: Martin Vala (JINR)
    • 12:00 13:00
      Conference Closing 1h Conference Hall

      Conference Hall

      LIT JINR

    • 13:00 14:00
      Lunch 1h
    • 14:00 15:30
      Mastering React Tutorial 1h 30m 407

      407

      LIT JINR

      Speaker: Michal Dančo (JINR Dubna, The Institute of Experimental Physics SAS in Košice)