Description

The course includes topics on code optimization for x86 platforms and
efficient code parallelization using OpenMP threading. Advanced aspects
of threading and optimization, such as new features of OpenMP 4.0, will
be covered during the course. Some performance aspects of MPI will also be discussed.

Learning outcome

Awareness of modern features of x86 CPUs
Ability to vectorize computations
Ability to use advanced features of OpenMP
Ability to increase code performance using threading and x86 optimization
Prerequisites

Good knowledge of C/C++ or Fortran
Good knowledge of threading using OpenMP
Basic knowledge of MPI
Basic knowledge of modern CPU architectures
Program
 

Day 1: Monday, April 18

  9:00-  9:45    Introduction
10:00-10:15   Coffee break
10:15-12:00   Performance Analysis methods & tools. Interactive demo.
12:00-13:00   Lunch break
13:00-13:45   Microarchitectural optimization
13:45-14:00   Coffee break
14:00-14:45   Optimization for SIMD
14:45-16:00   Exercises

Day 2: Tuesday, April 19

   9:00- 9:45   Optimization for SIMD cont'd
  9:45-10:00   Coffee break
10:00-11:15    Exercises
11:15-12:00    Memory access optimization
12:00-13:00   Lunch break
13:00-13:45    Memory access optimization cont'd
13:45-14:45    Exercises
14:45-15:00   Coffee break
15:00-16:00   Exercises

Day 3: Wednesday, Aprl 20

  9:00-  9:45   Advanced OpenMP
  9:45-10:00   Coffee break
10:00-11:15    Exercises
11:15-12:00    OpenMP performance considerations
12:00-13:00   Lunch break
13:00-14:00   Exercises
14:00-14:45   MPI performance considerations
14:45-15:00   Coffee break
15:00-16:00   ExercisesLecturers:  Sebastian von Alfthan (CSC), Mikko Byckling (Intel), Pekka Manninen (CSC)

Language:  EnglishPrice:           Free of charge

https://events.prace-ri.eu/event/492/