Research Aide– MCS – Shim, Junhyung – 2.2.26
About The Position
Compute nodes in exascale supercomputers and AI factories integrate multiple heterogeneous processing units. The portable and efficient programming of these architectures is a significant challenge. Application developers address this challenge by using multiple parallel programming models (CUDA, HIP, Level Zero, SYCL, OpenMP, MPI...) -- often mixing them to get the best from all worlds. Programming as such may introduce correctness issues throughout the entire software stack, including data races or misuse of programming model semantics, leading to undefined behavior. It is particularly true in deep software stacks -- such as machine-learning-oriented frameworks like oneCCL and pytorch, that relies on low-level parallel programming model at the bottom. The increasing complexity of hardware, programming models, and frameworks, requires new tools to strengthen the correctness of parallel programs. In this internship we will develop new semantical tools to analyze parallel programming models. The primary targets are CUDA and Level Zero, the two low-level programming models that drive workloads on Aurora, Polaris, and future NVIDIA AI machines. An initial goal of the internship is to extend (or re-develop if needed) a Level Zero semantical validator built on top of THAPI traces. This will allow applications to validate that their code is producing expected Level Zero calls.
Requirements
- The entirety of the appointment must be conducted within the United States.
- Applicants must be: ‒ Currently enrolled in undergraduate or graduate studies at an accredited institution. ‒ Graduated from an accredited institution within the past 3 months; or ‒ 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.
- If accepting an offer, must pass a screening drug test
- Must complete a satisfactory background check.
Responsibilities
- Develop new semantical tools to analyze parallel programming models.
- Extend (or re-develop if needed) a Level Zero semantical validator built on top of THAPI traces.
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
Full-time
Career Level
Intern
Education Level
No Education Listed
