GPU Hackathons are five day intensive hands-on events designed to help computational scientists port their applications to GPUs using libraries, OpenACC, CUDA and other tools by pairing participants with dedicated mentors experienced in GPU programming and development. Representing distinguished scholars and preeminent institutions around the world, these teams of mentors and attendees work together to realize performance gains and speedups by taking advantage of parallel programming on GPUs. This event is jointly organized by Helmholtz-Zentrum Dresden-Rossendorf (HZDR) and Jülich Supercomputing Centre (JSC) in association with the Helmholtz Federated IT Services Software Cluster (HIFIS).
The goal of the 5-day hands-on workshop is for current or prospective user groups of large-scale, GPU-accelerated systems to port their applications to GPU accelerators or to further optimize already ported applications. The teams should leave with applications running on GPUs, or at least with a clear roadmap of how to get there. Access will be provided to the following GPU-accelerated supercomputers:
- The TAURUS system at TU Dresden with NVIDIA K80 GPUs and Intel Haswell processors as well as NVIDIA V100 GPUs and IBM POWER9 processors
- The JUWELS cluster at JSC with NVIDIA V100 GPUs and Intel Skylake processors
The selected projects will receive a mentor that will prepare them and their application for the hackathon week and will work with them in person during that week.
Please note that teams need to have at least three team members to ensure a sustainable hackathon experience for the team. Join us for this one week of extreme immersion in GPU computing. Prior GPU programming knowledge is not required, but expect homework given before the event in this case. If you are currently pondering if or how to move your application to GPUs, this event is for you and you should apply!
Space is limited, so only six to seven teams can be accepted. In case of over-subscription preference will be given to teams with more widely used, open-source applications and/or teams for which a suitable mentor is available.