General Purpose Graphics Processing Units potentially offer high memory bandwidth and performance across a range of applications, however programming them can be challenging. These hackathons help overcome this challenge by pairing new and existing GPU programmers with experienced mentors from universities and NVIDIA. By the end of the 5-day workshop teams will have applications running on GPUs or have a clear road map of how to get there.
The Open Call is already under way but there’s still plenty of time to submit a proposal. Ideally you will be a team of 3-5 developers, ideally postdocs, research software engineers (RSEs) or graduate students with a scalable application to port (or optimize on) GPU accelerators. Collectively the team should know the application intimately.
Programming experience with CUDA is not a requirement.
Full details of the event can be found here: http://gpuhack.shef.ac.uk/.
- 15 June – End of the Open Call
- 1 July 2019 – Notification of admittance and Open Registration
- 19 – 23 August 2019 – Sheffield GPU Hackathon
Travel and Accommodation
Registration for the GPUHackathon is free and includes all meals and refreshments during the meeting when accommodation is booked at the venue.
Participants are expected to cover their accommodation and travel. Staying at the venue is highly preferred, since it maximises networking with all participants and gets the most out of the hacking.
Please note this event is not organised by N8 CIR but is certain to be of interest to research software engineers across the partnership.