Overview of the Hackathon
This Hackathon was in continuation to a successful execution of a pilot test run at IIT Madras.
The key objective of the hackathon has been to optimize, scale and tune the user codes to achieve better performance and/or solve larger problem sizes. The key highlight of this activity is cross disciplinary collaboration and teamwork between Domain Experts (end-users) and Computer Scientists (mentors). This provided a platform for participants to learn new skills and technologies related to parallel programming which will be useful to take their simulation codes to the next level of computational performance. Applications across various departments were received.Events Dates and Place
The event was jointly organized by C-DAC and IIT Guwahati from 5th Feb 2025 to 7th Feb 2025.
Participating Teams
In total, 20 teams had applied for this event while 6 of them were selected for participating in the event.
Pre-Hackathon Activities (2 weeks prior to the event in online mode)
The teams were introduced to their respective mentors. As many of the codes were focussed on OpenACC based GPU implementation, a short online training programme on OpenACC was conducted for the participants.
Following activities were carried out before the actual mini-hackathon:
- Get the code compiled and run on the target platform
- Select appropriate input test cases and setup code correctness validation mechanism
- Get the code profiled with Intel VTune profiler and identify the hotspots
Conclusion
The hackathon was successfully executed, with actively engaging the participants and extending their efforts beyond the event itself. Performance improvements were impressive, with the highest speed-up reaching 5832 times, and the lowest at 2 times.
Future Work and Suggestions
Due to limited time during the mini-hackathon, we could not get to a fully optimized version for many codes. Hence, we have requested teams and their respective mentors to continue interacting in online mode till the code gets to reasonable performance.
Suggestions for future hackathons:
- Target more user/ legacy codes on GPU clusters with emerging tools (OpenACC)
- Identify user codes which can scale on bigger clusters (20 PF) and extend necessary support (including additional system time under NSM)
- Target codes catering ‘Grand Challenge Problems’
- Conduct User’s Meets at regular intervals (say monthly) for effective engagement among both domain experts and computer scientists
- Identify codes which can be catered to quantum computing using hybrid environment