Application Closed
Hackathon Dates: 30 May (Online), 6 - 9 June 2023 (Hybrid)
Address: Birch Building 1.43 Interactive Theatre, ANU
Together with NVIDIA and OpenACC organization, NCI will host a hybrid Hackathon starting May 30, 2023 (virtually) and continuing June 6 - 9, 2023 (hybrid).
Open Hackathons are multi-day, intensive hands-on events designed to help computational and data scientists and researchers port, accelerate and optimize their applications on a variety of data centre parallel computing architectures. The event pairs participants with dedicated mentors experienced in programming and targeted application areas to realize performance gains and speedups using a variety of programming models, libraries, and tools.
The goal of the open hackathon is for computational and data scientists to port, accelerate and optimize their applications to modern computer architectures, including CPUs, GPUs, and other computing technologies. Participating teams should leave the event either with their applications accelerated and/or optimized on the latest supercomputing hardware or with a clear roadmap of the next steps needed to leverage these resources.
This hackathon is open to everyone looking to take their projects to the next level; however, priority acceptance will be given to NCI-affiliated scientists and their collaborators.
The NCI Open Hackathon will be hosted as a hybrid event. Day 0 (May 30, 2023) will be held virtually and attendees will have the option to attend Days 1 - 4 (June 06 - 09, 2023) either in-person or online only via a digital experience supported by online tools.
All communication for both remote and in-person participants will be done through Zoom, Slack, and email. This event will be conducted in the Australian Eastern Standard Time (AEST) zone. Attendees are responsible to cover their travel expenses.
Team | Sub-project | Project Details Presentation | Application Speedup (x times) | Node Replacement | GPU Compute Power Savings (W) | GPU Networking Power Savings (W) | Total Power (W) | Annual Compute Power Saving Per GPU Node (kWh/year) | Annual Networking Power Saving Per GPU Node (kWh/year) | Total Annual Energy Savings | $/kWh | Annual Cost Savings | 3-year Cost Savings | Metric Ton CO2 | Gas Car Driven for 1 year | Seedlings Trees growing for 10 years |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
MasterAI | Deepspeed | 3 | 24 | 19900 | 1022 | 20922 | 174324 | 8949 | 183273 | 0.34 | 62312.92 | 186938.75 | 130 | 28 | 2148 | |
Sophia | 1 | 1 | -5400 | -46 | -5446 | 47304 | 407 | 47711 | 0.34 | 16221.67 | 48665 | 34 | 7 | 559 | ||
ANSYS | Breathing | 10 | 15 | 10000 | 604 | 10604 | 87600 | 5288 | 92888 | 0.34 | 31581.99 | 94745 | 66 | 14 | 1089 | |
Submarine | 20.7 | 20.7 | 16313 | 870 | 17183 | 142901 | 7623 | 1505240 | 0.34 | 51178.08 | 153534.24 | 107 | 23 | 1764 | ||
Gridap | 100 | 6.3 | 375 | 197 | 572 | 3285 | 1729 | 5014 | 0.34 | 1704.71 | 5114.12 | 4 | 1 | 9 | ||
CompEcon | 1315 | 986.3 | 1078375 | 45705 | 1124080 | 9446565 | 400379 | 9846944 | 0.34 | 3347960.91 | 10043882.7 | 6981 | 1507 | 115406 | ||
ANUGA-CDAC | 60 | 0.34 | ||||||||||||||
HybridPIC | (Wrote source code) | 0.34 | ||||||||||||||
Gaia RVS | 25 | 0.8 | -5641 | -57 | -5697 | 49412 | 496 | 49908 | 0.34 | 16968.6 | 50905.8 | 35 | 8 | 585 | ||
Explorer | 21.2 | 15.9 | 10998 | 646 | 11644 | 96345 | 5657 | 102002 | 0.34 | 34680.69 | 104042.08 | 72 | 16 | 1195 | ||
Genome GPT | 22.4 | 16.8 | 11980 | 687 | 12667 | 104945 | 6020 | 110965 | 0.34 | 37728.18 | 113184.53 | 79 | 17 | 1301 | ||
Monofonia | 4.5 | 3.4 | -2788 | 64 | -2724 | 24419 | 559 | 23859 | 0.34 | 8112.12 | 24336.35 | 17 | 4 | 280 |
No advanced parallel computing or GPU skills are required, but it is helpful for teams to know the basics of GPU programming and profiling. We recommend familiarizing yourself with these concepts by accessing any of the no-charge lectures, tutorials, and labs available to all participants, or registering for our Bootcamps. Please contact the organizers for more information to help you prepare for the hackathon.
Teams attending the event will be given access to a GPU cluster for the duration of the hackathon.
If your team is accepted for the hackathon, registration information will be provided along with mentor introductions and computational resource access. Accepted teams should:
All times in the Australian Eastern Standard Time (AEST) zone
Day 0 - Team/Mentor Meeting: Tuesday, May 30, 2023: 09:00 AM - 11:00 AM (Online Only)
Day 1 - Tuesday, June 6, 2023: 09:00 AM - 05:00 PM (Hybrid*)
Days 2 and 3 - Wednesday and Thursday, June 7-8, 2023: 09:00 AM - 05:00 PM (Hybrid*)
Day 4 - Friday, June 9, 2023: 09:00 AM - 03:30 PM (Hybrid*)