# Automatic BLAS Offloading on Unified Memory Architecture: A Study on NVIDIA Grace-Hopper

arXiv: https://arxiv.org/abs/2404.13195 · HTML: https://mlsanity.com/paper/2404.13195/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 19 Apr 2024.

## Built on by (most cited first, 1 of 1)

- [Characterizing and Optimizing LLM Inference Workloads on CPU-GPU Coupled Architectures](https://mlsanity.com/paper/2504.11750/) (2025)

## Abstract (arXiv)

Porting codes to GPU often requires major efforts. While several tools exist for automatically offload numerical libraries such as BLAS and LAPACK, they often prove impractical due to the high cost of mandatory data transfer. The new unified memory architecture in NVIDIA Grace-Hopper allows high bandwidth cache-coherent memory access of all memory from both CPU and GPU, potentially eliminating bottleneck faced in conventional architecture. This breakthrough opens up new avenues for application development and porting strategies. In this study, we introduce a new tool for automatic BLAS offload, the tool leverages the high speed cache coherent NVLink C2C interconnect in Grace-Hopper, and enables performant GPU offload for BLAS heavy applications with no code changes or recompilation. The tool was tested on two quantum chemistry or physics codes, great performance benefits were observed.