4th Thematic CERN School of Computing

Europe/Zurich
MEDILS in Split, Croatia

MEDILS in Split, Croatia

Alberto Pace (CERN), Ivica Puljak (University of Split. Fac.of Elect. Eng., Mech. Eng. and Nav.Arc)
Description

t CSC2016
"Efficient and Parallel Processing of Scientific Data"

The 4th Thematic CERN School of Computing will be held in Split, Croatia from Monday 23 May  to Friday 27 May 2016 evening. Students should arrive on Sunday 22 May and leave on Saturday 28 May 2016.

The school is organized by CERN in collaboration with the University of Split (FESB - www.fesb.hr) and will be hosted at the Mediterranean Institute For Life Sciences (MEDILS) Conference Centre (www.medils.org). The Centre is a historical renovated building situated in a wooded and landscaped park located on the Adriatic Sea coast, a few kilometres from the centre of Split.

Reference letter
Virtual appliance
    • 3:00 PM
      Registration
    • 8:00 PM
      Dinner at MEDILS
    • Opening Session
      Conveners: Alberto Pace (CERN), Ivica Puljak (University of Split. Fac.of Elect. Eng., Mech. Eng. and Nav.Arc)
      • 1
        Introduction to FESB
        Speaker: Damir Lelas (University of Split. Fac.of Elect. Eng., Mech. Eng. and Nav.Arc)
      • 2
        Introduction to the thematic School of Computing
        Speaker: Alberto Pace (CERN)
    • Inaugural Scientific Lecture
      Convener: Ivica Puljak (Technical University of Split FESB)
      • 3
        Inaugural Scientific Lecture
    • 11:00 AM
      Updates
    • 11:15 AM
      Coffee
    • Technologies and Platforms
      Convener: Andrzej Nowak
    • 12:45 PM
      Lunch
    • Technologies and Platforms
      Convener: Andrzej Nowak
    • 3:00 PM
      Drinks & Fruits
    • 3:30 PM
      Sport afternoon introduction
    • Technologies and Platforms
      Convener: Andrzej Nowak
      • 6
        Exercise 1
        Speaker: Andrzej Nowak
      • 7
        Exercise 2
        Speaker: Andrzej Nowak
    • 5:45 PM
      Free time
    • 6:00 PM
      Transportation down town & Guided tour of Split
    • 8:15 PM
      Outside Welcome Dinner Chops Grill

      Chops Grill

      CHOPS GRILL Steak & Seafood KAMILA TONČIĆA 4, SPLIT, CROATIA

      http://www.chops-grill.com/en/

    • Effective I/O for Scientific Applications
      Convener: Sebastien Ponce (CERN)
    • Technologies and Platforms
      Convener: Andrzej Nowak
    • 11:00 AM
      Updates
    • 11:15 AM
      Coffee break
    • Effective I/O for Scientific Applications
      Convener: Sebastien Ponce (CERN)
    • 12:45 PM
      Lunch
    • 1:30 PM
      Study time and/or Daily Sports Program
    • 4:00 PM
      Drinks & Fruits
    • Programming for concurrency and correctness
      Convener: Danilo Piparo (CERN)
    • Technologies and Platforms
      Convener: Andrzej Nowak
      • 12
        Exercise 3
        Speaker: Andrzej Nowak
    • Effective I/O for Scientific Applications
      Convener: Sebastien Ponce (CERN)
    • 7:30 PM
      Free time
    • 8:00 PM
      Dinner at MEDILS
    • Programming for concurrency and correctness
      Convener: Danilo Piparo (CERN)
    • Programming for concurrency and correctness
      Convener: Danilo Piparo (CERN)
    • 11:00 AM
      Updates
    • 11:15 AM
      Coffee
    • Programming for concurrency and correctness
      Convener: Danilo Piparo (CERN)
    • 12:45 PM
      Lunch
    • 1:30 PM
      Sports afternoon

      Rafting in Omis

    • 8:00 PM
      Outside Dinner Radmanove mlinice

      Radmanove mlinice

      Kanjon rijeke Cetine,21310, Omiš

      http://www.radmanove-mlinice.hr/index.php/en/radman-s-mills

    • Programming for concurrency and correctness
      Convener: Danilo Piparo (CERN)
    • Programming for concurrency and correctness
      Convener: Danilo Piparo (CERN)
      • 18
        Exercise 2
        Speaker: Danilo Piparo (CERN)
    • 11:00 AM
      Updates
    • 11:15 AM
      Coffee
    • Programming for concurrency and correctness
      Convener: Danilo Piparo (CERN)
      • 19
        Exercise 3
        Speaker: Danilo Piparo (CERN)
    • 12:45 PM
      Lunch
    • 1:30 PM
      Study time and/or Daily Sports Program
    • 4:00 PM
      Drinks & Fruits
    • 4:20 PM
      Official School Photo
    • Effective I/O for Scientific Applications
      Convener: Sebastien Ponce (CERN)
      • 20
        Key ingredients to achieve effective I/O
        Lecture 4
        Speaker: Sebastien Ponce (CERN)
    • Effective I/O for Scientific Applications
      Convener: Sebastien Ponce (CERN)
    • Effective I/O for Scientific Applications
      Convener: Sebastien Ponce (CERN)
    • 7:30 PM
      Free time
    • 8:00 PM
      Dinner at MEDILS
    • Effective I/O for Scientific Applications
      Convener: Sebastien Ponce (CERN)
    • Technologies and Platforms
      Convener: Andrzej Nowak
    • 11:00 AM
      Updates
    • 11:15 AM
      Coffee
    • 11:45 AM
      Self-Evaluation by students
    • 12:45 PM
      Lunch
    • Closing Scientific Lecture: Hydrogen powered automobiles: status and future
      Convener: Prof. Frano Barbir (FESB)
      • 25
        Closing Scientific Lecture (tbc)
    • Students presentations
      • 26
        Student presentations
        • a) A quick trip from code profiling to file formats

          Several points should be taken into account when it comes to data import into Big Data solutions. Today we cover two, the first one is a data transfer engine (Apache Sqoop) and how we have profiled it with a profiler for distributed applications. Secondly, we provide some numbers to file formats that come from different use cases.

          Speaker: Daniel Lanza Garcia (Ministere des affaires etrangeres et europeennes (FR))
        • b) New versus malloc

          "new[] isn't always better", aka "Don't trust your compiler". During a code optimization exercise we very slowly discovered that a seemingly innocent C++ 'new double[n]' construct was up to a factor of ten slower than an archaic 'malloc'.

          Speaker: Jan Justinus Keijser
        • c) Development of distributed and concurrent applications with the actor model and akka

          An brief overview of the conceptual model to deal with concurrent computations called "Actor model" including the outline of one of its most famous implementations: Akka, a framework for Scala and Java

          Speaker: Cesar Aguilera Padilla (Universidad de Granada (ES))
        • d) Where are the low hanging fruits?

          When facing a big physics reconstruction codebase, there is always room for improvement. However, choosing what to explore is crucial in the search for success!

          Speaker: Daniel Hugo Campora Perez (Universidad de Sevilla (ES))
        • e) Using Jupyterhub in the Belle2 experiment

          Jupyter allows to develop (Python-based) applications, so-called notebooks, which bundle code, documentation and results, e.g. plots. The advantages over classical script-based approaches is the feature to recompute only parts of the analysis code, which allows for fast and iterative development, and a web-based user frontend, which can be hosted centrally and only requires a browser on the user side. This talk will give a short introduction into the jupyter ecosystem.

          Speaker: Nils Braun (KIT)
        • f) Performance comparison for NVIDIA CUDA and Intel Xeon Phi
          Speaker: Petra Loncar (Faculty of Electrical Engineering, Mechanical Engineering and N)
        • g) Challenges for the Level-1 Trigger Control Software of a High-Energy Physics Experiment

          The trigger subsystem is an important component of a high-energy physics experiment with particular software requirements. Concurrent control and monitoring of hundreds of different boards is not uncommon. This presentation provides a brief overview of the challenges and the solutions that LHC experiments have devised to deal with them.

          Speaker: Christos Lazaridis (University of Wisconsin-Madison (US))
        • h) Network performance optimization of database replications at WLCG

          To achieve a better network performance over database replications at WLCG, different studies have been done during several months in collaboration with specific Tier 1s. These studies will be explained during this presentation to show how was the optimization workflow in order to improve the replication performance.

          Speaker: Lorena Lobato Pardavila (Universidad de Oviedo (ES))
        • i) Embarrassingly parallel made embarrassingly difficult
          Speaker: Baldur van Lew (Leiden University Medical Centre)
    • 4:30 PM
      Drinks & Fruits
    • 5:00 PM
      Closing session
    • 5:30 PM
      Free time
    • 8:00 PM
      Outside Closing dinner Restoran – kavana Bajamonti

      Restoran – kavana Bajamonti

      Prokurativa d.o.o. Trg Republike 1, Split

      http://www.restoran-bajamonti.hr/en/

    • 9:00 AM
      Departure