SC25 Home
Close

Presentation

Distributed Sparse Tensor Computations in MLIR
SessionLLVM-HPC2025: The 11th Workshop on the LLVM Compiler Infrastructure in HPC
DescriptionMLIR (Multi-Level Intermediate Representation) is a popular framework for implementing domain-specific compilers for optimizing matrix/tensor computations. However, no support currently exists in MLIR for distributed sparse tensor computations. In this paper, we describe the design and implementation of a new MLIR dialect for high-level specification of distributed sparse tensor computations. This specification is then lowered to MPI-based code for distributed execution. We illustrate the expressiveness of the new dialect by implementing a Graph Attention Network computation with multiple sparse tensor operators.
Author/Presenters
Miheer Vaidya
University of Utah
Shreya Singh
University of Utah
Devanshu Mantri
University of Utah
Michael Shannon Eydenberg
Sandia National Laboratories
Brian Michael Kelley
Sandia National Laboratories
Sivasankaran Rajamanickam
Sandia National Laboratories
Atanas Rountev
Ohio State University
P. Sadayappan
University of Utah
Event Type
Workshop
TimeMonday, 17 November 20252:30pm - 3:00pm CST
Location260
Next PresentationNext Presentation
Afternoon Break - LLVM-HPC2025
Similar Presentations
An MLIR pipeline for offloading Fortran to FPGAs via OpenMP
ASaP: Automatic Software Prefetching for Sparse Tensor Computations in MLIR
Accelerating Sparse Tensor Contractions