Performance analysis and optimization workshop

11.-12.6.2024 Oslo, Norway

This two-day workshop in Oslo targets advanced LUMI users and developers who seek to enhance their code’s efficiency. Participants will engage in detailed lessons on code profiling and optimization, followed by practical demos and exercises. The course will give attendees the knowledge required to understand their application performance and what opportunities there are for optimization. Attendees are encouraged to bring their own applications for expert analysis and advice on achieving optimal performance on the LUMI platform.

Please note that this is an advanced workshop and not an HPC intro course. Participants are expected to have an understanding of high-performance computing and the general working of LUMI.  Note that there is considerable overlap with the content of this workshop and parts of the 4-day General LUMI course. This workshop contains more hands-on and live demo sessions.

Workshop Prerequisites

Participants are expected to have:

• Solid knowledge of HPC cluster computing.
• Familiarity with Linux command line and programming (C, Fortran, or Python).
• Good understanding of LUMI.
• Project on LUMI or already applied for one.

If you do need a refresher on the specifics of LUMI, join our 2-day LUMI intro course or have a look into our training material repository with slides, exercises and recordings of all our trainings including our advanced 4-day course.

Agenda

The workshop consists of a mix of short lectures, live demos and hands-on exercises, that cover the following key topics:

• Running CPU and GPU profilers and analyzing the output
• Identifying bottlenecks in applications
• Mitigation of the most common performance bottlenecks (for example serial code performance, application placement, MPI, data movement and I/O, GPU occupancy, registers occupancy, strided data access pattern)
• Expert advice on achieving optimal performance on the LUMI platform for your application

Each day will run from 9:00 to 16:30 CEST, with breaks scheduled throughout.

Event info

The worksop will be jointly conducted by the LUMI HPE Centre of Excellence (HPE CoE), AMD and the LUMI User Support Team (LUST). This event is sponsored by Norwegian AI Cloud

The course is intended for users with ongoing projects on LUMI, users with project proposals in one of the national or EuroHPC channels, and support staff of local organizations of the LUMI consortium members.

This workshop is an in-person event in Oslo only. All sessions will be recorded and published afterwards but we feel that we can only provide a good experience to an onsite audience. Several staff members from HPE, AMD and LUST will be on-site, so this is the perfect moment for interaction with them.

The venue for in-person attendance is at the University of Oslo in Norway. See the bottom of this page for more information on the venue. Participants are responsible themselves for all travel bookings.

If registrations exceed the number of planned participants, we will maintain a waiting list. The registration link can be found at the bottom of this page.

Attendance are free of charge. In-person participants will be provided with a sandwich lunch.

To participate in the hands-on exercises and access the slides and training materials on the course day, participants will need to join a course-specific training project. This project is managed by Puhuri, which means that CSC users will need to set up a new LUMI user account. Step-by-step instructions will be sent out after registration on how to join the project and set up an account if needed.

Registration

Please register for the course by 30. May 2024. Participants will receive a quick confirmation on a first-come, first-served basis so that they can arrange travel. If your plans change, we kindly ask you to cancel your registration as soon as possible, ideally before the registration deadline. The email acknowledging your registration will contain a link to manage it.

Users who do not have an account on LUMI yet will receive temporary access for the purpose of the course. The allocated compute time should only be used for the course exercises. Any abuse will result in removal from the allocation for this and future courses.

Venue

The course will take place at the University of Oslo in Norway. Participants are responsible for arranging their own travel and accommodations.

Ole-Johan Dahls hus
Seminarrom Python
Gaustadalléen 23B
0373 Oslo
Norway

This workshop is sponsored by Norwegian AI Cloud.