First INFN International School on: "Architectures, tools and methodologies for developing efficient large scale scientific computing applications" ESC09 - Bertinoro (Forlì-Cesena) I 12-16 October 2009

Europe/Rome
Bertinoro

Bertinoro

Friday announcement
Participants
    • Registration and Welcome
    • 20:30
      Welcome Dinner
    • Session 1
      • 1
        Welcome and Opening of the School
        Speaker: Vincenzo Maria Vagnoni (BO)
        Slides
      • 2
        Introduction/Setting the scene for why efficient use of modern architectures is important
        Speaker: Sverre Jarp (CERN)
        Slides
      • 3
        Review of modern CPU architectures (The 7 dimensions of performance)
        Speaker: Sverre Jarp (CERN)
      • 10:40
        Coffee break
      • 4
        Designing Architectures and Frameworks for HEP
        Speaker: Pere Mato (CERN)
        Slides
      • 5
        Designing Architectures and Frameworks for HEP
        Speaker: Pere Mato (CERN)
      • 12:40
        Lunch break
      • 6
        Understanding performance tuning
        Speaker: Mr Andrzej Nowak (CERN)
        Slides
      • 7
        Exercises
        Speakers: Mr Andrzej Nowak (CERN), Davide Salomoni (CNAF), Sverre Jarp (CERN)
      • 8
        Exercises
        Speakers: Davide Salomoni (CNAF), Vincenzo Innocente (CERN)
      • 16:30
        Coffee break
      • 9
        Exercises
        Speakers: Davide Salomoni (CNAF), Dr Pere Mato (CERN)
    • 10
      Evening Lecture: "High throughput data trasmission through network links"
      Speaker: Domenico Galli (BO)
      Slides
    • 20:30
      Dinner
    • Session 2
      • 11
        Memory management and use
        Speaker: Lassi Tuura (Northeastern University)
        Slides
      • 12
        Memory management and use
        Speaker: Lassi Tuura (Northeastern University)
      • 10:10
        Coffee break
      • 13
        Data structures
        Speaker: P. Calafiura (LBNL)
        Slides
      • 14
        Data structures
        Speaker: P. Calafiura (LBNL)
      • 12:00
        Lunch break
      • 15
        Software physical design
        Speaker: Pere Mato (CERN)
        Slides
      • 16
        Exercises
        Speakers: Davide Salomoni (CNAF), Lassi Tuura (Northeastern University), Peter Elmer (Princeton University)
        Minutes
      • 16:00
        Coffee break
      • 17
        Exercises
        Speakers: Davide Salomoni (CNAF), Paolo Calafiura (Lawrence Berkeley National Lab), Dr Pere Mato (CERN)
      • 18
        Exercises
        Speakers: Davide Salomoni (CNAF), Paolo Calafiura (Lawrence Berkeley National Lab), Dr Pere Mato (CERN)
    • 19
      Evening Lecture: "Getting it all with C++: Abstraction, Reusability, Performance and Future-Safety"
      Speaker: Ulrich Drepper (Red Hat)
      Slides
    • 20:30
      Dinner
    • Session 3
      • 20
        Basic C++ performance issues
        Speaker: Sebastien Binet (LAL - IN2P3)
        Slides
      • 21
        Basic C++ performance issues
        Speaker: Sebastien Binet (LAL - IN2P3)
      • 22
        Compiler Optimization including efficient programming with SIMD (SSE) instructions
        Speakers: Mr Andrzej Nowak (CERN), Sverre Jarp (CERN)
        Slides
      • 10:55
        Coffee break
      • 23
        Building the software
        Speaker: Peter Elmer (Princeton University)
        Slides
      • 24
        Building the software
        Speaker: Peter Elmer (Princeton University)
      • 13:00
        Lunch break
      • 25
        Exercises on Memory Management
        Speakers: Davide Salomoni (CNAF), Lassi Tuura (Northeastern University), Peter Elmer (Princeton University)
      • 26
        Exercises
        Speakers: Dr Alfio Lazzaro (CERN), Davide Salomoni (CNAF), Dr Sebastien Binet (LAL/IN2P3)
    • 16:30
      Tour to Ravenna
    • 20:30
      Social dinner (Ravenna)
    • Session 4
      • 27
        Introduction to multithreading/multiprocessing methodology and vocabulary
        Speaker: Vincenzo Innocente (CERN)
        Slides
      • 28
        Introduction to OpenMP
        Speaker: Alfio Lazzaro (CERN)
        Slides
      • 29
        Parallel Programming using std::thread
        Speaker: Vincenzo Innocente (CERN)
      • 10:55
        Coffee break
      • 30
        Support tools for multithreading
        Speaker: Mr Andrzej Nowak (CERN)
        Paper
        Slides
      • 31
        Introduction to MPI
        Speaker: Dr Alfio Lazzaro (CERN)
        Slides
      • 13:00
        Lunch break
      • 32
        Parallel Programming exercises using OpenMP and MPI
        Speakers: Dr Alfio Lazzaro (CERN), Sverre Jarp (CERN)
        exercise solutions
        Slides
      • 33
        Parallel Programming exercises using OpenMP and MPI
        Speakers: Alfio Lazzaro (MI), Vincenzo Innocente (CERN)
        code example using std algo
      • 16:00
        Coffee break
      • 34
        Exercises using std::thread
        Speakers: Davide Salomoni (CNAF), Vincenzo Innocente (CERN)
        code example 1
        code example 2
        Paper
        web tutorial
      • 35
        Exercises
        Speakers: Mr Andrzej Nowak (CERN), Davide Salomoni (CNAF), Mr Sverre Jarp (CERN)
    • 36
      Evening Lecture: Multicores, GPUs, FPGAs and custom processors for scientific computing: a delicate tradeoff"
      Speaker: Raffaele Tripiccione (FE)
      Slides
    • 20:30
      Dinner
    • Session 5
    • 46
      Evening lecture: Virtualization, Grid, Cloud: Integration Paths for Scientific Computing
      Speaker: Davide Salomoni (CNAF)
      Slides
    • 20:30
      Dinner
    • Session 6
      • 47
        Students feedback
      • 48
        Final examination
        Slides
      • 11:00
        Coffee break
      • 49
        Delivery of certificates of attendance
      • 12:00
        Lunch
      • 50
        Shuttle departure