Second INFN International School on: "Architectures, tools and methodologies for developing efficient large scale scientific computing applications" ESC10 - Bertinoro (Forlì-Cesena) I 22-27 November 2010

Europe/Rome
Bertinoro

Bertinoro

Description
FINAL PROGRAM
group photo
photo in class
    • 16:00 20:30
      Registration and Welcome
    • 20:30 20:45
      Welcome Dinner 15m
    • 08:30 20:00
      Session 1
      • 08:30
        Welcome and Opening of the School 30m
        Speaker: Federico Ruggieri (RM3)
      • 09:00
        Concepts of performance and efficiency 45m
        Speaker: Dr Peter Elmer (Princeton University)
        Slides
      • 09:50
        Modern processors and related optimisation topics - Part 1 45m
        Speaker: Mr Sverre Jarp (CERN)
        Slides
      • 10:40
        Coffee break 20m
      • 11:00
        Modern processors and related optimisation topics - Part 2 45m
        Speaker: Mr Sverre Jarp (CERN)
      • 11:50
        Exercises - Introduction to Performance tuning tools 45m
        Speakers: Mr Andrzej Nowak (CERN), Mr Sverre Jarp (CERN)
        Exercises
      • 12:40
        Lunch break 1h 30m
      • 14:15
        Introduction to GPUs/CPU accelerators 45m
        Speaker: Mr Andrzej Nowak (CERN)
        Slides
      • 15:00
        Exercises - Continued focus on Performance tuning tools 45m
        Speakers: Mr Andrzej Nowak (CERN), Mr Sverre Jarp (CERN)
      • 15:45
        Coffee break 15m
      • 16:00
        Exercises - Continued focus on Performance tuning tools 45m
        Speakers: Mr Andrzej Nowak (CERN), Mr Sverre Jarp (CERN)
      • 16:45
        Exercises - Continued focus on Performance tuning tools 45m
        Speakers: Mr Andrzej Nowak (CERN), Mr Sverre Jarp (CERN)
    • 20:30 20:30
      Dinner
    • 08:30 20:00
      Session 2
      • 08:30
        Efficient C++ coding 45m
        Speaker: Dr Sebastien Binet (LAL/IN2P3)
        Slides
      • 09:20
        Efficient C++ coding 45m
        Speaker: Dr Sebastien Binet (LAL/IN2P3)
      • 10:10
        Coffee break 20m
      • 10:30
        Exercises - Basic C++ optimisations 45m
        Speakers: Dr Sebastien Binet (LAL/IN2P3), Vincenzo Innocente (CERN)
        Exercises
      • 11:15
        Exercises - Basic C++ optimisations 45m
        Speakers: Dr Sebastien Binet (LAL/IN2P3), Vincenzo Innocente (CERN)
      • 12:40
        Lunch break 1h 30m
      • 14:15
        The Memory Crisis 45m
        Speaker: Lassi Tuura (FNAL)
        Slides
      • 15:00
        How memory allocation works 45m
        Speaker: Lassi Tuura (FNAL)
      • 15:45
        Coffee break 15m
      • 16:00
        Exercises - Memory Allocations 45m
        Speakers: Lassi Tuura (FNAL), Dr Peter Elmer (Princeton University)
        Exercises
      • 16:45
        Exercises - Memory Allocations 45m
        Speakers: Lassi Tuura (FNAL), Dr Peter Elmer (Princeton University)
      • 18:30
        Evening Lecture: "The new C++ runtime library: more and better" (Paolo Carlini, Oracle Italia) 1h
        Slides
    • 20:30 20:30
      Dinner
    • 08:30 16:30
      Session 3
      • 08:30
        How memory allocation works 45m
        Speaker: Lassi Tuura (FNAL)
      • 09:20
        Monitoring/Debugging Memory Usage 45m
        Speaker: Dr Peter Elmer (Princeton University)
      • 10:10
        Coffee break 20m
      • 10:30
        Exercises - Monitoring Memory Usage 45m
        Speakers: Lassi Tuura (FNAL), Dr Peter Elmer (Princeton University)
      • 11:20
        Exercises - Monitoring Memory Usage 45m
        Speakers: Lassi Tuura (FNAL), Dr Peter Elmer (Princeton University)
      • 12:20
        Lunch break 1h 30m
      • 14:00
        Reducing Memory Usage 45m
        Speaker: Dr Peter Elmer (Princeton University)
      • 14:50
        Exercises - Reducing Memory Usage 45m
        Speakers: Lassi Tuura (FNAL), Dr Peter Elmer (Princeton University)
      • 15:40
        Exercises - Reducing Memory Usage 45m
        Speakers: Lassi Tuura (FNAL), Dr Peter Elmer (Princeton University)
    • 17:30 19:45
      Social tour 2h 15m

      Guided tour to Casa Artusi (centre of gastronomic culture dedicated to Italian home cooking).
      Cooking demonstration and tasting of Piadina Romagnola

    • 19:45 19:45
      Social dinner

      Casa Artusi Restaurant

    • 08:30 20:00
      Session 4
      Announcements
      • 08:30
        Floating point computation 45m
        Speaker: Vincenzo Innocente (CERN)
        Slides
      • 09:20
        Exercises - Optimization floating point computation in C++ 45m
        Speaker: Vincenzo Innocente (CERN)
        tar file with exercises and doc
      • 10:10
        Coffee break 20m
      • 10:30
        Vectorization using SIMD (part I) 45m
        Speaker: Vincenzo Innocente (CERN)
      • 11:15
        vectorization using SIMD (part II) 45m
        Speaker: Vincenzo Innocente (CERN)
      • 12:05
        Exercises - Vectorization using gcc 45m
        Speakers: Alfio Lazzaro (CERN), Vincenzo Innocente (CERN)
        exercises
      • 12:50
        Lunch break 1h 30m
      • 14:30
        I/O Efficiency 45m
        Speaker: Dr Gerhard Brandt (DESY)
        Slides
      • 15:15
        Exercises - I/O Efficiency 45m
        Speaker: Dr Gerhard Brandt (DESY)
        Minutes
      • 16:00
        Coffee break 15m
      • 16:15
        Exercises - I/O Efficiency 45m
        Speaker: Dr Gerhard Brandt (DESY)
      • 18:30
        Evening Lecture: "Exploiting GPU'sfor scientific computing" (Leone Bosi, INFN Perugia) 1h
    • 20:30 20:30
      Dinner
    • 08:30 19:30
      Session 5
      • 08:30
        Parallel programming theory 45m
        Lecture
        Speaker: Dr Tim Mattson (Intel)
        Slides
      • 09:20
        An introduction to OpenMP 45m
        Lecture and exercises
        Speaker: Dr Tim Mattson (Intel)
        Exercises
      • 10:10
        Coffee break 20m
      • 10:30
        Synchronization in OpenMP 30m
        Lecture and exercises
        Speaker: Dr Tim Mattson (Intel)
      • 11:00
        Work sharing constructs 45m
        Lecture and exercises
        Speaker: Dr Tim Mattson (Intel)
      • 11:50
        The OpenMP data environment 45m
        Lecture and exercises
        Speaker: Dr Tim Mattson (Intel)
      • 12:50
        Lunch break 1h 30m
      • 14:30
        OpenMP tasks 45m
        Lecture and exercises
        Speaker: Dr Tim Mattson (Intel)
      • 15:15
        OpenMP Memory model 45m
        Lecture and exercises
        Speaker: Dr Tim Mattson (Intel)
      • 16:00
        Coffee break 15m
      • 16:15
        A survey of programming models 45m
        Lecture
        Speaker: Dr Tim Mattson (Intel)
      • 18:30
        Evening Lecture: "How to program a 1000 core processor" (Timothy G. Mattson, Intel Co.) 1h
    • 20:30 20:30
      Dinner
    • 08:30 14:50
      Session 6
      • 08:30
        Students feedback 30m
      • 09:00
        Final examination 2h
      • 11:00
        Coffee break 30m
      • 11:30
        Delivery of certificates of attendance 30m
      • 12:00
        Lunch 1h 15m
      • 14:30
        Shuttle departure 1m