CMS: A Computational Memory Solution for High-Performance and Power-Efficient Recommendation System
Minho Ha, Joonseop Sim, Donguk Moon, Myunghyun Rhee, Jungmin Choi, Byungil Koh, Euicheol Lim, Kyoung Park · 2022 IEEE 4th International Conference on Artificial Intelligence Circuits and Systems (AICAS) · 2022
This paper proposes a cost-effective and scalable memory solution for high-performance and power-efficient recommendation systems, called computational memory solution (CMS). To address the memory bandwidth challenges of the deep learning-based recommendation system, CMS offloads memory-intensive embedding operations to near-data processors equipped with large capacity and high bandwidth memory. In contrast to the other state-of-the-art near-memory processing accelerators that only support inference and have scalability restrictions, CMS supports training and scales as much as the high-speed serial interface allows. Our evaluation results show that CMS achieves up to$7.5\times$higher throughput and$12.2\times$higher power efficiency for training than state-of-the-art accelerators.