# iMARS: An In-Memory-Computing Architecture for Recommendation Systems

arXiv: https://arxiv.org/abs/2202.09433 · HTML: https://mlsanity.com/paper/2202.09433/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 18 Feb 2022.

## Built on by (most cited first, 2 of 2)

- [Robust Implementation of Retrieval-Augmented Generation on Edge-based Computing-in-Memory Architectures](https://mlsanity.com/paper/2405.04700/) (2024)
- [DeepCAM: A Fully CAM-based Inference Accelerator with Variable Hash Lengths for Energy-efficient Deep Neural Networks](https://mlsanity.com/paper/2302.04712/) (2023)

## Abstract (arXiv)

Recommendation systems (RecSys) suggest items to users by predicting their preferences based on historical data. Typical RecSys handle large embedding tables and many embedding table related operations. The memory size and bandwidth of the conventional computer architecture restrict the performance of RecSys. This work proposes an in-memory-computing (IMC) architecture (iMARS) for accelerating the filtering and ranking stages of deep neural network-based RecSys. iMARS leverages IMC-friendly embedding tables implemented inside a ferroelectric FET based IMC fabric. Circuit-level and system-level evaluation show that fw achieves 16.8x (713x) end-to-end latency (energy) improvement compared to the GPU counterpart for the MovieLens dataset.