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

Abstract

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 systemlevel evaluation show that iMARS achieves 16.8× (713×) endto-end latency (energy) improvement compared to the GPU counterpart for the MovieLens dataset.