Third INFN International School on: "Architectures, tools and methodologies for developing efficient large scale scientific computing applications" ESC11 - Bertinoro (Forlì-Cesena) Italy 23-29 October 2011

Contribution List

0. Welcome and Opening of the School

24/10/2011, 08:30

Session 1

1. Concepts of performance and efficiency

Dr Peter Elmer (Princeton University)

24/10/2011, 09:00

Session 1

2. Modern processors and related optimisation topics - Part 1

Mr Sverre Jarp (CERN)

24/10/2011, 09:50

Session 1

3. Modern processors and related optimisation topics - Part 2

Mr Sverre Jarp (CERN)

24/10/2011, 11:00

Session 1

4. Introduction to Performance tuning tools

Lassi Tuura (Fermilab), Dr Peter Elmer (Princeton University)

24/10/2011, 11:50

Session 1

5. Floating point computation: accuracy, optimization, vectorization (with exercises)

Vincenzo Innocente (CERN)

24/10/2011, 14:15

Session 1

6. Floating point computation: accuracy, optimization, vectorization (with exercises)

Vincenzo Innocente (CERN)

24/10/2011, 15:00

Session 1

7. Floating point computation: accuracy, optimization, vectorization (with exercises)

Vincenzo Innocente (CERN)

24/10/2011, 16:00

Session 1

8. Floating point computation: accuracy, optimization, vectorization (with exercises)

Vincenzo Innocente (CERN)

24/10/2011, 16:45

Session 1

57. Evening Lecture: "GPU for scientific computing" - Alessandro Lonardo - INFN Roma 1

24/10/2011, 18:30

Session 1

16. Efficient C++ coding

Dr Sebastien Binet (LAL/IN2P3), Vincenzo Innocente (CERN)

25/10/2011, 08:30

Session 2

17. Efficient C++ coding

Dr Sebastien Binet (LAL/IN2P3), Vincenzo Innocente (CERN)

25/10/2011, 09:20

Session 2

18. Exercises - Basic C++ optimisations

Dr Sebastien Binet (LAL/IN2P3), Vincenzo Innocente (CERN)

25/10/2011, 10:30

Session 2

19. Exercises - Basic C++ optimisations

Dr Sebastien Binet (LAL/IN2P3), Vincenzo Innocente (CERN)

25/10/2011, 11:15

Session 2

20. The Memory Crisis

Lassi Tuura (Fermilab)

25/10/2011, 14:15

Session 2

21. How memory allocation works

Lassi Tuura (Fermilab)

25/10/2011, 15:00

Session 2

22. Exercises - Memory Allocations

Lassi Tuura (Fermilab)

25/10/2011, 16:00

Session 2

23. Exercises - Memory Allocations

Lassi Tuura (Fermilab)

25/10/2011, 16:45

Session 2

24. Evening Lecture: "Architectures of High Throughput DAQ systems" - Niko Neufeld (CERN)

25/10/2011, 18:30

Session 2

9. Exercises (Floating Point, Memory use, C++)

26/10/2011, 08:30

Session 3

10. Exercises (Floating Point, Memory use, C++)

26/10/2011, 09:20

Session 3

11. Exercises (Floating Point, Memory use, C++)

26/10/2011, 10:30

Session 3

12. Exercises (Floating Point, Memory use, C++)

26/10/2011, 11:20

Session 3

13. Exercises (Floating Point, Memory use, C++)

26/10/2011, 14:00

Session 3

14. Exercises (Floating Point, Memory use, C++)

26/10/2011, 14:50

Session 3

15. Exercises (Floating Point, Memory use, C++)

26/10/2011, 15:40

Session 3

34. Parallel programming theory

Dr Tim Mattson (Intel)

27/10/2011, 08:30

Session 4

35. An introduction to OpenMP

Dr Tim Mattson (Intel)

27/10/2011, 09:20

Session 4

36. Synchronization in OpenMP

Dr Tim Mattson (Intel)

27/10/2011, 10:30

Session 4

37. Work sharing constructs

Dr Tim Mattson (Intel)

27/10/2011, 11:15

Session 4

38. The OpenMP data environment

Dr Tim Mattson (Intel)

27/10/2011, 12:05

Session 4

39. I/O Efficiency

Vincenzo Maria Vagnoni (BO)

27/10/2011, 14:30

Session 4

41. I/O Efficiency

Vincenzo Maria Vagnoni (BO)

27/10/2011, 16:15

Session 4

42. Evening Lecture: "OpenCL and the quest for Performance Portability" - Timothy G. Mattson (Intel)

27/10/2011, 18:30

Session 4

25. OpenMP tasks

Dr Tim Mattson (Intel)

28/10/2011, 08:30

Session 5

26. OpenMP Memory model

Dr Tim Mattson (Intel)

28/10/2011, 09:20

Session 5

27. A survey of programming models

Dr Tim Mattson (Intel)

28/10/2011, 10:30

Session 5

52. A survey of programming models

Dr Tim Mattson (Intel)

28/10/2011, 11:30

Session 5

53. Lecture on low-level CPU profiling

Mr Sverre Jarp (CERN)

28/10/2011, 14:30

Session 5

54. Physical design of SW

Benedikt Hegner (CERN)

28/10/2011, 15:15

Session 5

55. Physical Design Examples/HEP Frameworks

Benedikt Hegner (CERN)

28/10/2011, 16:15

Session 5

56. Software Development Tools

Benedikt Hegner (CERN)

28/10/2011, 17:00

Session 5

58. Announcements

Federico Ruggieri (ROMA3)

28/10/2011, 17:45

Session 5

33. Evening Lecture: "Linux on Multicore: challanges and perspective" - Andrea Arcangeli (RedHat)

28/10/2011, 18:30

Session 5

43. Students feedback

29/10/2011, 08:30

Session 6

44. Final examination

29/10/2011, 09:00

Session 6

45. Delivery of certificates of attendance

29/10/2011, 11:30

Session 6

46. Shuttle departure

29/10/2011, 14:30

Session 6

32. A survey of programming models

Dr Tim Mattson (Intel)

Lecture

49. Evening Lecture: "Getting it all with C++: Abstraction, Reusability, Performance and Future-Safety"

Ulrich Drepper (Red Hat)

47. Evening Lecture: "High throughput data trasmission through network links"

Domenico Galli (BO)

50. Evening Lecture: "How to program a 1000 core processor" (Timothy G. Mattson, Intel Corp.)

51. Evening Lecture: Multicores, GPUs, FPGAs and custom processors for scientific computing: a delicate tradeoff"

Raffaele Tripiccione (FE)

48. Evening lecture: Virtualization, Grid, Cloud: Integration Paths for Scientific Computing

Davide Salomoni (CNAF)

40. Exercises - I/O Efficiency

Vincenzo Maria Vagnoni (BO)

31. OpenMP Memory model

Lecture and exercises

30. OpenMP tasks

Lecture and exercises

29. The OpenMP data environment

Dr Tim Mattson (Intel)

Lecture and exercises

28. Work sharing constructs

Dr Tim Mattson (Intel)

Lecture and exercises