ICE: An Intelligent Cognition Engine with 3D NAND-based In-Memory Computing for Vector Similarity Search Acceleration
Journal
Proceedings of the Annual International Symposium on Microarchitecture, MICRO
Journal Volume
2022-October
ISBN
9781665462723
Date Issued
2022-01-01
Author(s)
Hu, Han Wen
Wang, Wei Chen
Chang, Yuan Hao
Lee, Yung Chun
Lin, Bo Rong
Wang, Huai Mu
Lin, Yen Po
Huang, Yu Ming
Lee, Chong Ying
Su, Tzu Hsiang
Hsieh, Chih Chang
Hu, Chia Ming
Lai, Yi Ting
Chen, Chung Kuang
Chen, Han Sung
Li, Hsiang Pang
Chang, Meng Fan
Wang, Keh Chung
Hung, Chun Hsiung
Lu, Chih Yuan
Abstract
Vector similarity search (VSS) for unstructured vectors generated via machine learning methods is a promising solution for many applications, such as face search. With increasing awareness and concern about data security requirements, there is a compelling need to store data and process VSS applications locally on edge devices rather than send data to servers for computation. However, the explosive amount of data movement from NAND storage to DRAM across memory hierarchy and data processing of the entire dataset consume enormous energy and require long latency for VSS applications. Specifically, edge devices with insufficient DRAM capacity will trigger data swap and deteriorate the execution performance. To overcome this crucial hurdle, we propose an intelligent cognition engine (ICE) with cognitive 3D NAND, featuring non-volatile in-memory computing (nvIMC) to accelerate the processing, suppress the data movement, and reduce data swap between the processor and storage. This cognitive 3D NAND features digital nvIMC techniques (i. e., ADClDAC-free approach), high-density 3D NAND, and compatibility with standard 3D NAND products with minor modifications. To facilitate parallel INT8/INT4 vector-vector multiplication (VVM) and mitigate the reliability issue of 3D NAND, we develop a bit-error-tolerance data encoding and a two's complement-based digital accumulator. VVM can support similarity computations (e.g., cosine similarity and Euclidean distance), which are required to search 'the most similar data' right where they are stored. In addition, the proposed solution can be realized on edge storage products, e.g., embedded Multi-Media Card (eMMC). The measured and simulated results on real 3D NAND chips show that ICE enhances the system execution time by 17× to 95× and energy efficiency by 11× to 140×, compared to traditional von Neumann approaches using state-of-the-art edge systems with MobileFaceNet on CASIA-WebFace dataset. To the best of our knowledge, this work demonstrates the first 3D NAND-based digital nvIMC technique with measured silicon data.
Subjects
3D NAND | In-Memory Computing | Unstructured Data Search | Vector Similarity Search
SDGs
Type
conference paper