Practical course on the CUDA to HIP porting


With the upcoming arrival of the LUMI supercomputer, we start a series of training to enable its future users to exploit its potentiality. In particular, the LUMI-G partition is based on AMD GPUs and therefore we start with a course that will help the users to port their applications to this new architecture with a first course on the porting of CUDA applications to HIP. It will provide an extensive dive into the Hipify tools as well as showcase many examples, and it includes hands-on sessions. During the hands-on session, you will work with applications provided by us as well as with your own application.

This training course is open to all users of the LUMI consortium and we strongly encourage all users who are planning to use LUMI to submit an application. All the selected participants will be provided training accounts on the CSC’s Puhti system with Nvidia GPUs which will be used during the hands-on sessions.

Agenda (Times are in CET)
09:00 – 10:00 Introduction to AMD architecture and HIP
10:00 – 10:15 Break
10:15 – 11:00 Deep dive to Hipify tools and some examples
11:00 – 11:45 Lunch
11:45 – 16:00 Hands-on sessions

Fee: Free for users within the LUMI consortium

Registration procedure:

Please use the registration form you find at the link below.  To send you code please prepare a txt file with the following information:

  • How to retrieve the code and the possible input data, be sure that there is access
  • The name of the code, the person of contact for any questions, and a list with the required dependencies
  • How to compile and execute the code

and send it to georgios.markomanolis(at)

The number of applications selected for the hand-on sessions is limited and it will depend upon the total number of requests we will receive.

Register for the training here.
The registration deadline is February 15th.

Organizer: EuroHPC CC Finland

Contact: Georgios Markomanolis, georgios.markomanolis(at)