Help us, help you! -- Help us improve the POP Service so that we can help you improve your HPC Applications
The primary aim of the POP User Forum is to provide the POP Users and non-Users alike with the opportunity to influence the direction of the service with their concrete questions and feedback. A secondary goal is to inform developers of HPC applications about the opportunity to obtain help in performance optimization by applying for services offered in the POP Project.
Session I. Introduction (14:30 - 16:30)
- A general presentation by the Project Coordinator about the objectives, provided services and procedures of the POP CoE (Speaker: Jesus Labarta, BSC) [Slides]
- Overview of success stories and reports from performance assessments performed in the first 18 months of operation (Speakers: Sally Bridgwater, NAG) [Slides]
- Several presentations by selected POP Users focused on their experiences from performance assessments and by POP Collaborators
- Harald Klimach, Universität Siegen: Results of Ateles und Musubi code analyses [Slides]
- Riccardo Rossi, UPC: Results of Kratos code analysis [Slides]
- Paul Gibbon, JSC: EoCoE/POP performance analysis workshops [Slides]
- Alan O'Cais, JSC: Collaboration with ECAM CoE [Slides]
- Ondrej Jakl, VSB: POP training at IT4I and results from PERMON analysis [Slides]
Session II. User Feedback and Wrap-up (17:00 - 19:00)
- Service Q&A session to get feedback (Moderator: Maike Gilliot, Ter@tec) [Slides]
- All participants will break into small groups (consisting of Users and non-Users) to provide feedback to the service as well as to convince non-Users to try the service or better understand what is required for non-Users to try the service (needs assessment). The Session will consist of a series of open questions as well as a quantitative assessment of the service.
- Group summaries of key feedback – 1 slide per small group, max 30 minutes.
This workshop is mainly targeted to Users of the PoP Service as well as anyone interested in learning what the POP Service can do for them, including:
- Organisations and individuals interested in receiving a detailed assessment of their code behaviour so that they may refactor the code in the most efficient manner.
- HPC code users interested in the assessment of achieved performance in specific production conditions or to learn about possible improvements by modifying their environment setup.
- Infrastructure operators interested in the assessment of achieved performance of codes running in production conditions, training of their support staff, or getting feedback for time computer time allocation processes.
- Vendors interested in benchmarking, customer support and system dimensioning / design.