Heterogeneous Programming on GPUs with MPI + OmpSs @ BSC (2012)
Date: 23 - 24 May 2012
Objectives:
The course will motivate the audience on the need for portable, efficient programming models that put less pressure on program developers while still getting good performance for clusters and clusters with GPUs.
More specifically, the tutorial will:
Introduce the hybrid MPI/OmpSs parallel programming model for future exascale systems
Demonstrate how to use MPI/OmpSs to incrementally parallelize/optimize:
MPI applications on clusters of SMPs, and
Leverage CUDA kernels with OmpSs on clusters of GPUs
Level:
For trainees with some theoretical and practical knowledge, some programming experience.Learning Outcomes:
The students who finish this course will be able to develop benchmarks and simple applications with the MPI/OmpSs programming model to be executed in clusters and clusters of GPUs.Prerequisites:
Good knowledge of C/C++
Basic knowledge of CUDA/OpenCL
Basic knowledge of Paraver/Extrae
Material for practical sessions will be provided during the course, and it is also interesting that students can provide their own application(s) for session 4, free hands-on.
You are expected to come with your own laptop with either linux, windows or MacOS operating system.
Dates:
from 23 May 2012 09:30 to 24 May 2012 18:00
Timezone:
CET
Location:
BSC, Barcelona
Additional info:
Course Program Outline:Day 1
Session 1
9am – 11am: Introduction to OmpSs11:30am – 1pm: OmpSs single node programming hands-on
Lunch Break 1pm to 2pm
Session 2
2 pm- 3 pm: More on OmpSs: GPU/CUDA programming3 pm- 6 pm: OmpSs single node programming hands-on with GPUs
Day 2
Session 3
9am- 10 am: Introduction to MPI/OmpSs10am- 1 pm: MPI/OmpSs hands-on
Lunch Break 1pm to 2pm
Session 4/ 2pm- 6 pm: Free hands-on: Students use OmpSs environment with prepared examples, except in the free hands-on session were they can bring their own application.
END of COURSE
Event types:
- Workshops and courses
Activity log