论文标题
IMARS：用于推荐系统的内存中计算体系结构
iMARS: An In-Memory-Computing Architecture for Recommendation Systems
论文作者
论文摘要
推荐系统（RECSYS）通过根据历史数据预测其偏好向用户建议项目。典型的recsys处理大型嵌入表和许多相关表的嵌入式表。传统计算机体系结构的内存大小和带宽限制了RECSYS的性能。这项工作提出了一种内存计算（IMC）体系结构（IMAR），用于加速基于神经网络的Recsys的过滤和排名阶段。 IMARS利用IMC友好型嵌入桌子内实现的桌子内的IMC织物。电路级别和系统级别的评估表明，与Movielens数据集的GPU对应物相比，\ fw的端到端延迟（能量）改进达到16.8倍（713倍）。
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.
