Research Aide - LCF - Srinivasan, Agnes - 6.22.26.
About The Position
This internship focuses on chipStar, a community-supported implementation of CUDA/HIP programming models, which is currently the only way to run unmodified HIP/CUDA codes on Aurora. Prior work indicates that chipStar can match or outperform native SYCL on some test cases, while lagging on others. The intern will review test cases and applications where chipStar performs worse than SYCL on Intel GPUs, diagnose the root causes (such as kernel-launch overhead, memcpy patterns, synchronization primitives), potentially using AI tools for assistance, and contribute fixes upstream to chipStar or document remaining issues. Performance improvements are directly beneficial to current users of Aurora, including two active INCITE/ALCC applications, GENESIS and ZeroRK, which already rely on chipStar.
Requirements
- Currently enrolled in undergraduate or graduate studies at an accredited institution.
- Graduated from an accredited institution within the past 3 months.
- Actively enrolled in a graduate program at an accredited institution.
- Must be 18 years or older at the time the appointment begins.
- Must possess a cumulative GPA of 3.0 on a 4.0 scale.
- The entirety of the appointment must be conducted within the United States.
- Candidates may be required to complete pre-employment drug testing based on appointment length.
- Must complete a satisfactory background check.
Responsibilities
- Review test cases and applications where chipStar performs worse than SYCL on Intel GPUs.
- Diagnose the root causes of performance issues, such as kernel-launch overhead, memcpy patterns, and synchronization primitives.
- Contribute fixes upstream to chipStar or document remaining issues.
Benefits
- Comprehensive benefits are part of the total rewards package.
Stand Out From the Crowd
Upload your resume and get instant feedback on how well it matches this job.
What This Job Offers
Job Type
Part-time
Career Level
Entry Level
Education Level
No Education Listed
