What is this Hackathon?
This hackathon is focused on migrating your code or project from CUDA to C++ with SYCL enabling the ability to use CPUs and GPUs using Intel(tm) Developer Cloud and 4th Gen Intel® Xeon® Scalable processors with Intel® Data Center GPU Max Series.
What is oneAPI?
oneAPI is an open, cross-industry, standards-based, unified, multiarchitecture, multi-vendor programming model that delivers a common developer experience across accelerator architectures – for faster application performance, more productivity, and greater innovation.
If your registration is accepted by Intel, you will be invited to participate in the hackathon where you you can port your CUDA code to SYCL.
|Registration begins||July 13, 2023|
|Optional Onboarding Workshop Day / Hackathon Begins 9am CT - 5pm CT||July 26, 2023|
|Check Point #1||August 17, 2023|
|Check Point #2||August 24, 2023|
|Project Submission Deadline||September 6, 2023|
|Participants continue to hack on their projects||August 24 -September 6|
|Hackathon Ends / Project Showcase||September 6, 2023|
Over the three weeks of the hackathon, participants will port their codes from CUDA to C++ with SYCL using a variety of tools, technologies, and techniques enabling them to run on multiple architectures.
1. Apply to hackathon on DevPost
2. Attend the hackathon workshop day on July 26 sign up here
- Mentors will be available on Discord to provide assistance during that time period.
4. At the end of the hackathon you will have functional code that not only runs on your existing NVIDIA hardware but on a number of GPUs and CPUs like 4th Gen Intel® Xeon® Scalable processors (code-named Sapphire Rapids), the Intel® Xeon® CPU Max Series (code-named Sapphire Rapids HBM) and the Intel® Data Center GPU Max Series (code-named Ponte Vecchio).
5. You will have the opportunity to share your newly ported project with the oneAPI community for greater visibility.
Your code will now be able to run on some of the fastest accelerators on the market with Intel® Xeon® CPU Max Series (code-named Sapphire Rapids HBM) and the Intel® Data Center GPU Max Series (code-named Ponte Vecchio).
Come join us, port your code from CUDA* to SYCL and get a demonstration of the speed and functionality of the Intel's latest processors and GPUs through the Intel Developer Cloud!
What to Bring
Bring your project or software with CUDA codes that you'd like to port over.
What to Submit
- Provide a video of your working project using something like asciinema (https://asciinema.org/) or a presentation.
- [optional and will not affect standing] Submit your project with SYCL code with a link to your GitHub/GitLab or other code repository site under a git branch that clearly signifies where your porting activities reside like for example 'cuda-to-sycl-hackathon' or a code fork.
- [optional and will not affect standing] : A blog post of your experience
$6,900 in prizes
Technical Phase Special Resources
All teams who are accepted into the technical phase will have special access to Intel mentors, free access to paid tools, and more great exposure opportunities!
Intel® Arc A770 Limited Edition Graphics Card
Fifteen Intel® Arc A770 Limited Edition 16GB PCI Express 4.0 Graphics Cards
Graphics Cards – ARV $459.99 Odds of winning depend on the total number of eligible entries
received. The total potential ARV (Approximate Retail Value) of all prizes is $5,519.88.
There will be no judging. Instead, prizes will be distributed based on eligibility rules based on location and random assignment.
In order to be considered for the prize, you will need to submit a slides detailing your porting journey, feedback, and status. While you do not have to provide code - you must provide proof of a project by website, a published paper or some other method that shows that a project exists.
All participants will be considered for further exposure through articles on the Intel® Developer Zone.
Please refer to the terms and conditions at https://images.plan.intel.com/Web/IntelCorporation/%7Bf08eeb4d-d95f-4af6-a82c-d89ae1f96ea1%7D_Online_Sweepstakes_Rules_CUDA_to_SYCL_Migration.pdf.
Participants who are legal residents of the following countries are eligible for prizing – legal residents of the 50 U.S. States, the District of Columbia, the United Kingdom, France, Netherlands, Switzerland, Germany, Japan, and Canada (excluding Quebec), South Korea and Taiwan. Participants will need to have reached the age of majority in their state or country of residence at the time of entry.
By opting in for the prizing you are confirming you are age 18 years or older. Intel may contact me for marketing-related communications. To learn about Intel's practices, including how to manage your preferences and settings, you can visit Intel's Privacy and Cookies notices.
Submitting to this hackathon could earn you:
Intel will vet registration information to select projects and participants who are most likely ready to migrate during the technical phase.