Mojo: MLIR-based Performance-Portable HPC Science Kernels on GPUs for the Python Ecosystem
By:
Godoy, William F; Melnichenko, Tatiana A; Valero Lara, Pedro ; Elwasif, Wael R; Fackler, Philip W; Ferreira Da Silva, Rafael ; Teranishi, Keita ; Vetter, Jeffrey S
Page Number:
2114-2128
Book Title:
Proceedings of the SC '25 Workshops of the International Conference for High Performance Computing, Networking, Storage and Analysis
Publication Date:
November 2025
Publisher Location:
Association for Computing Machinery, New York, New York, United States of America
Conference Name:
The International Conference for High Performance Computing, Networking, Storage, and Analysis 2025 (SC '25)
We explore the performance and portability of the novel Mojo language for scientific computing workloads on GPUs. As the first language based on the LLVM’s Multi-Level Intermediate Representation (MLIR) compiler infrastructure, Mojo aims to close performance and productivity gaps by combining Python’s interoperability and CUDA-like syntax for compile-time portable GPU programming. We target four scientific workloads: a seven-point stencil (memory-bound), BabelStream (memory-bound), miniBUDE (compute-bound), and Hartree–Fock (compute-bound with atomic operations); and compare their performance against vendor baselines on NVIDIA H100 and AMD MI300A GPUs. We show that Mojo’s performance is competitive with CUDA and HIP for memory-bound kernels, whereas gaps exist on AMD GPUs for atomic operations and for fast-math compute-bound kernels on both AMD and NVIDIA GPUs. Although the learning curve and programming requirements are still fairly low-level, Mojo can close significant gaps in the fragmented Python ecosystem in the convergence of scientific computing and AI.
