This course will be delivered as an ONLINE COURSE for remote participation due to the COVID-19 measures enforced by most European governments.
Description
The course gives an introduction to GPU programming using HIP. Topics covered in the course include concepts specific to GPUs and GPU programming, writing of GPU kernels, performance considerations, profiling tools, and how to use multiple GPUs together with MPI.
This course is part of CSC Autumn of HPC 2021, a set of five two day modules teaching essential skills in parallel programming for modern GPU-accelerated supercomputers and organised by CSC – IT Center for Science Ltd (Finland) under PRACE Training Centres activity.
Learning outcome
After the course, the participants will be able to write simple GPU programs, know how to analyse the performance of their GPU programs, and understand how to combine GPUs with MPI.
Prerequisites and content level
The participants are assumed to have a working knowledge of the C++ (or C) programming language and to be familiar with MPI.
The content level of the course is broken down as: beginner's - 0%, intermediate - 60%, advanced - 40%, community-targeted content - 0%.
Preliminary agenda
Day 1, Thursday December 9
-
Introduction to GPUs and GPU programming
-
Kernels, synchronisation, streams
-
Memory allocation/access, unified memory
Day 2, Friday December 10
-
Fortran and HIP
-
Kernel optimisation, profiling, and libraries
-
Multi-GPU programming
Lecturers
Martti Louhivuori (CSC), Georgios Markomanolis (CSC), Jaro Hokkanen (CSC)
Language: English
Price: Free of charge (2 training days)
REGISTRATION is OBLIGATORY since the details to access the online course will be provided to the registered and accepted attendees only. If you have registered to this course and you are not able to attend, please CANCEL your registration in advance by sending an email to patc@csc.fi
