Score: 3

Ember: A Compiler for Efficient Embedding Operations on Decoupled Access-Execute Architectures

Published: April 14, 2025 | arXiv ID: 2504.09870v1

By: Marco Siracusa , Olivia Hsu , Victor Soria-Pardos and more

BigTech Affiliations: Stanford University ARM

Potential Business Impact:

Makes computer recommendations much faster.

Business Areas:
Embedded Software Software

Irregular embedding lookups are a critical bottleneck in recommender models, sparse large language models, and graph learning models. In this paper, we first demonstrate that, by offloading these lookups to specialized access units, Decoupled Access-Execute (DAE) processors achieve 2.6$\times$ higher performance and 6.4$\times$ higher performance/watt than GPUs on end-to-end models. Then, we propose the Ember compiler for automatically generating optimized DAE code from PyTorch and TensorFlow. Conversely from other DAE compilers, Ember features multiple intermediate representations specifically designed for different optimization levels. In this way, Ember can implement all optimizations to match the performance of hand-written code, unlocking the full potential of DAE architectures at scale.

Country of Origin
🇺🇸 🇬🇧 United Kingdom, United States

Page Count
14 pages

Category
Computer Science:
Hardware Architecture