Multi-bit Ferroelectric-NAND for High-throughput Massive Database Search
Abstract
The growing demand for large-scale database search in data-intensive applications, ranging from proteomics to autonomous systems, has exposed fundamental limitations in von Neumann architectures due to memory bandwidth and energy constraints. Hyperdimensional (HD) computing offers a robust and parallelizable framework for such tasks, but its practical implementation remains challenged by high memory demands. Ultra-high-density, energy-efficient ferroelectric NAND (FE-NAND) memory provides a potential solution by enabling in-situ computation. We fabricate quad-level FE-NAND strings with wide memory windows, disturb resilience, and robust retention. Using these planar FE-NAND strings as building blocks, we experimentally demonstrate in-situ multi-level cell (MLC) dot product operations at the single-cell level and, using experimentally calibrated physics-based simulations, demonstrate Hamming similarity calculations between reference and query hypervectors (HV). This platform leverages the inherent error tolerance of HD computing to achieve >90% search accuracy even at high logic levels (TLC, QLC). When benchmarked on Open Modification Search (OMS) tasks in proteomics with TB-scale datasets, our system shows nearly 1,000x speedup and over 10,000x energy efficiency improvement compared to incumbent solutions. These results establish FE-NAND as a viable in-storage compute architecture for large-scale, high-dimensional data processing.
Related
- Implementation of Grover’s Search Algorithm for Tectonic Activity Detection in an Unsorted Seismic Database
- Inpainting to search for premerger massive black hole binaries in realistic LISA datasets
- Compiling Together: High-Throughput Distributed Quantum Computing via Multi-Compilation
- VEQDB: A Compact and Reconstructible Multi-Device Tokamak Equilibrium Database
- Bond-informed search for ordered configurations in multi-principal element alloys
BibTeX
@misc{venkatesan2026multibit,
title = {{Multi-bit Ferroelectric-NAND for High-throughput Massive Database Search}},
author = {Prasanna Venkatesan and Tanvir H. Pantha and Po-Kai Hsu and Sumukh Pinge and Zheyu Li and Chinsung Park and Priyankka Ravikumar and Hari Jayasankar and Lance Fernandes and Weihong Xu and Zihan Xia and Flavio Ponzina and Keming Fan and Amrit Garlapati and Huy Tran and Taeyoung Song and Mengkun Tian and Hang Chen and Winston Chern and Kijoon Kim and Kwangyou Seo and Suhwan Lim and Kwangsoo Kim and Wanki Kim and Daewon Ha and Duygu Kuzum and Shimeng Yu and Mingu Kang and Tajana Rosing and Sourav Dutta and Asif Khan},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.05002},
eprint = {2610.05002},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.05002},
note = {CC BY-NC-ND 4.0},
}
RIS
TY - UNPB
TI - Multi-bit Ferroelectric-NAND for High-throughput Massive Database Search
AU - Prasanna Venkatesan
AU - Tanvir H. Pantha
AU - Po-Kai Hsu
AU - Sumukh Pinge
AU - Zheyu Li
AU - Chinsung Park
AU - Priyankka Ravikumar
AU - Hari Jayasankar
AU - Lance Fernandes
AU - Weihong Xu
AU - Zihan Xia
AU - Flavio Ponzina
AU - Keming Fan
AU - Amrit Garlapati
AU - Huy Tran
AU - Taeyoung Song
AU - Mengkun Tian
AU - Hang Chen
AU - Winston Chern
AU - Kijoon Kim
AU - Kwangyou Seo
AU - Suhwan Lim
AU - Kwangsoo Kim
AU - Wanki Kim
AU - Daewon Ha
AU - Duygu Kuzum
AU - Shimeng Yu
AU - Mingu Kang
AU - Tajana Rosing
AU - Sourav Dutta
AU - Asif Khan
PY - 2026
DA - 2026/10/04
PB - arXiv
DO - 10.48550/arxiv.2610.05002
UR - https://arxiv.org/abs/2610.05002
AB - The growing demand for large-scale database search in data-intensive applications, ranging from proteomics to autonomous systems, has exposed fundamental limitations in von Neumann architectures due to memory bandwidth and energy constraints. Hyperdimensional (HD) computing offers a robust and parallelizable framework for such tasks, but its practical implementation remains challenged by high memory demands. Ultra-high-density, energy-efficient ferroelectric NAND (FE-NAND) memory provides a potential solution by enabling in-situ computation. We fabricate quad-level FE-NAND strings with wide memory windows, disturb resilience, and robust retention. Using these planar FE-NAND strings as building blocks, we experimentally demonstrate in-situ multi-level cell (MLC) dot product operations at the single-cell level and, using experimentally calibrated physics-based simulations, demonstrate Hamming similarity calculations between reference and query hypervectors (HV). This platform leverages the inherent error tolerance of HD computing to achieve >90% search accuracy even at high logic levels (TLC, QLC). When benchmarked on Open Modification Search (OMS) tasks in proteomics with TB-scale datasets, our system shows nearly 1,000x speedup and over 10,000x energy efficiency improvement compared to incumbent solutions. These results establish FE-NAND as a viable in-storage compute architecture for large-scale, high-dimensional data processing.
ER -