User Tools

Site Tools


appds:arch17

Совещания 2017г

2017.12.19 (Skype)

Participants

  • Kryukov
  • Haungs
  • Kostunin

Working plan for 2018 (RU part)

Architecture of APPDS (A.Kryukov)
Events sample of TUNKA for ML (E.Korosteleva)
Comparision TensorFlow and PyTourch for APP tasks
Data Format (on example of KASCADE and TAIGA)
Meta data provenance system
On-fly Aggregation service

Protocol

Arrange the mailing lists

  • APPDS (Andreas: German members)
  • APPDS-manager (Kryukov, Haungs, Kostunin + ?)

IT-infrastructure

  • Git
  • Redmine

Upcoming events

Next steps

  • KCDC deployment at SINP
  • PhD students at KIT (beginning of work: April)

2017.11.9 (Skype)

Participants

  • Kryukov
  • Haungs
  • Kostunin

Protocol

Topics

  1. Organisational/Structural Matters
    • Agreement on monthly (skype) management meetings. Next: 19/12/2017 (after lunch)
    • A first KRAD meeting in Moscow envisaged: week of 5-9 March 2018?? (kind of kick-off meeting)
    • Establish a mailing list for the members (responsible: Kryukov; will appds@theory.sinp.msu.ru ) action: sample names for the list!
    • Create a wiki page for internal matters (action: Dima!)
    • Think about a GIT-Hub as infrastructure
    • Sample a list of possible IT/physics conferences for KRAD (action: all)
  2. Discussion on first actions/work plan for 2018:
    • In Russia: At the first stage (2018) the following works are planned:
      • to prepare an analytical review of modern scientific and technical literature;
      • to formulate general criteria for the functioning of the big data analysis system for astrophysical experiments and develop a list of the basic quantitative parameters of the system;
      • to develop a data description format for the KASCADE and TAIGA experiments and search and develop a method of aggregation of these data;
      • to develop a preliminary version of the particle type identification algorithm in the TAIGA experiment based on the machine learning method;
      • to develop the architecture of a system for storing, processing and analyzing big data of astrophysical experiments;
      • to develop a research stand on the basis of the KCDC system at SINP MSU to develop the proposed methods and algorithms;
      • to deploy the HubZero platform for filling it with educational materials for the preparation of students and graduate students in the field of particle astrophysics.
      • to prepare 3 articles for publication;
      • to prepare an interim report.
    • At KIT:
      • wp1 (KCDC extension): software extension of KCDC to include Tunka-related data
      • wp2 (Big Data Science Software): Run first jobs for KCDC or Tunka on a Tier/Grid environment (i.e. concept for Docker, Container, etc…)
      • wp3 (Multi-messenger Data Analysis): find/select 2 PhDs and define appropriate physics topics.
      • wp4 (go for public): define/create concept and design for KRAD webpages

Action: create a coherent working plan as preparation for the December meeting

appds/arch17.txt · Last modified: 08/11/2018 01:29 by admin