Postdoctoral Researcher - Just-In-Time Compilation and Optimizations of HPC Applications

$106,200 - $106,200/Yr

Lawrence Livermore National Laboratory - Livermore, CA

Full-time - Entry Level

Livermore, CA

Professional, Scientific, and Technical Services

About the position

The Postdoctoral Researcher position at Lawrence Livermore National Laboratory focuses on Just-In-Time (JIT) compilation and optimization methods for high-performance computing (HPC) applications. The role involves collaborating with researchers to develop innovative JIT techniques aimed at enhancing the performance of HPC applications, particularly on advanced heterogeneous many-core and GPU-enabled systems. The researcher will engage in independent research, publish findings, and contribute to the scientific community.

Responsibilities

• Research, design and implement JIT compilation techniques using state-of-the-art compilation frameworks, i.e., LLVM/MLIR, targeting C/C++ HPC application codes.
• Research, design and develop dynamic JIT optimization methods that reduce the execution time of HPC applications, especially related to GPU execution.
• Research, design and develop techniques for scaling JIT compilation and optimization for large-scale, distributed, multi-node application deployments.
• Document research by publishing papers in peer-reviewed media and presenting papers within the DOE community and at academic conferences.
• Pursue independent (but complementary) research interests and interact with a broad spectrum of scientists internally and externally to define and carry out the research.
• Perform other related duties as assigned.

Requirements

• Ph.D. in Computer Science, or a related field.
• Ability to perform research and development in state-of-the-art compilation technologies, as demonstrated by published research articles and/or software.
• Ability to perform research and development in program analysis techniques, including static and dynamic analysis, as demonstrated by published research articles and/or software.
• Ability to conduct high-quality research and develop implementations to evaluate research ideas independently as demonstrated by published research articles.
• Experience in programming in C/C++, Python, and GPU programming models (CUDA/HIP/OpenMP offload) in a Unix/Linux environment.
• Proficient verbal and written communication skills necessary to interact in a clear and concise manner, author technical and scientific reports and papers, and deliver scientific presentations.
• Ability to take the initiative and have interpersonal communication skills necessary to work effectively in a dynamic team environment.

Nice-to-haves

• Experience with state-of-the-art compilation tools and technologies, including LLVM, MLIR.
• Experience with parallel programming models, including OpenMP, MPI, CUDA/HIP.

Benefits

• 401(k)
• Flexible schedule
• Relocation assistance
• Tuition reimbursement