Post-Doctoral Researcher F/M Dynamic Parallelization of Sparse Codes for High-Performance Computing and Machine Learning

HiPEAC; Lyon; Sur place; EUR 42 000 - 54 000; Plein temps

Résumé du poste

Inria Lyon seeks a Post-Doctoral Researcher for 14 months to optimize dynamic parallelization of sparse codes in high-performance computing and machine learning, focusing on compiler algorithms and scheduling.

The fellow will propose code optimizations and data structures for scaling sparse propagation, address runtime scheduling by relying on existing parallel runtimes, and validate the approach on benchmarks from standard collections.

This position is funded by the French PEPR NumPEx program.

Qualifications

• PhD in CS/Applied Math required.
• Experience with compiler/runtime algorithm development.
• Background in HPC and machine learning is desirable.

Responsabilités

• Propose code optimizations and data structures for scaling sparse propagation.
• Address runtime scheduling by relying on existing parallel runtimes.
• Validate the complete approach on benchmarks using sparse tensors from standard collections.

Connaissances

Compiler optimization

Runtime scheduling

Parallel computing

Machine learning

Formation

PhD in Computer Science / Applied Math

Outils

BLAS/LAPACK

HPC runtimes

Description du poste

Most kernels of interest in machine learning and high-performance computing manipulate sparse tensors. Sparse codes are highly irregular and make use of array indirections and dynamic control which jeopardize static automatic parallelization algorithms.

The overall objective of this postdoctoral fellowship is to investigate compiler and runtime algorithms to delay the specialization of the dense code at runtime when the sparse structure is known.

From a dense specification, we seek to compile a code able to specialize itself on the sparse input data. The specialization will involve a set of sub-computations, which are expected to be achievable by standard linear algebra routines (e.g. gemm), using state-of-the art linear algebra libraries.Several issues must be investigated:

• How to specialize the code? In particular, how propagate efficiently the sparsity along the computation flow?
• How to detect library kernels on the specialized code?
• How to enforce a proper scheduling for the parallel runtime?

Points 1 and 2 have been partially addressed by a PhD student.

The postdoctoral fellow will:

• Propose code optimizations and data structures for scaling sparse propagation (point 1)
• Address runtime scheduling (point 3) by relying on existing parallel runtimes
• Validate the complete approach (points 1, 2 and 3) on scientific benchmarks by using sparse tensors from the Florida sparse matrix collection as well as machine learning applications.

This position is funded by the French prioritary research program for exascale computing in France (PEPR NumPEx).

Inria Lyon is a leading research center in computer science and applied mathematics, dedicated to advancing knowledge and technology. We foster innovation, collaboration, and excellence in research and education.

Metadata

Topics: Compilation, High-performance computing, Optimization, Parallel computing, Performance engineering, Runtime performance

Summary

Inria Lyon seeks a Post-Doctoral Researcher for 14 months to optimize dynamic parallelization of sparse codes in high-performance computing and machine learning, focusing on compiler algorithms and scheduling.

Post-Doctoral Researcher F/M Dynamic Parallelization of Sparse Codes for High-Performance Computing and Machine Learning

Full-time

Inria Lyon

Lyon, FR

Inria Lyon seeks a Post-Doctoral Researcher for 14 months to optimize dynamic parallelization of sparse codes in high-performance computing and machine learning, focusing on compiler algorithms and scheduling.

Context & Goals

Most kernels of interest in machine learning and high-performance computing manipulate sparse tensors. Sparse codes are highly irregular and make use of array indirections and dynamic control which jeopardize static automatic parallelization algorithms.

The overall objective of this postdoctoral fellowship is to investigate compiler and runtime algorithms to delay the specialization of the dense code at runtime when the sparse structure is known.

From a dense specification, we seek to compile a code able to specialize itself on the sparse input data. The specialization will involve a set of sub-computations, which are expected to be achievable by standard linear algebra routines (e.g. gemm), using state-of-the art linear algebra libraries.Several issues must be investigated:

• How to specialize the code? In particular, how propagate efficiently the sparsity along the computation flow?
• How to detect library kernels on the specialized code?
• How to enforce a proper scheduling for the parallel runtime?

Points 1 and 2 have been partially addressed by a PhD student.

The postdoctoral fellow will:

• Propose code optimizations and data structures for scaling sparse propagation (point 1)
• Address runtime scheduling (point 3) by relying on existing parallel runtimes
• Validate the complete approach (points 1, 2 and 3) on scientific benchmarks by using sparse tensors from the Florida sparse matrix collection as well as machine learning applications.

This position is funded by the French prioritary research program for exascale computing in France (PEPR NumPEx).

The HiPEAC project has received funding from the European Union's Horizon Europe research and innovation funding programme under grant agreement number 101296676. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union. Neither the European Union nor the granting authority can be held responsible for them.