RT Journal Article
T1 MATSA: An MRAM-Based Energy-Efficient Accelerator for Time Series Analysis.
A1 Fernández Vega, Iván
A1 Giannoula, Christina
A1 Manglik, Aditya
A1 Quislant-del-Barrio, Ricardo
A1 Mansouri-Ghiasi, Nika
A1 Gómez-Luna, Juna
A1 Gutiérrez-Carrasco, Eladio Damián
A1 Plata-González, Óscar Guillermo
A1 Mutlu, Onur
K1 Proceso de datos
K1 Dispositivos de almacenamiento de datos
K1 Gestión de memoria
K1 Análisis de series temporales
AB Time Series Analysis (TSA) is a critical workload to extract valuable information from collections of sequential data, e.g., detecting anomalies in electrocardiograms. Subsequence Dynamic Time Warping (sDTW) is the state-of-the-art algorithm for high-accuracy TSA. We find that the performance and energy efficiency of sDTW on conventional CPU and GPU platforms are heavily burdened by the latency and energy overheads of data movement between the compute and the memory units. sDTW exhibits low arithmetic intensity and low data reuse on conventional platforms, stemming from poor amortization of the data movement overheads. To improve the performance and energy efficiency of the sDTW algorithm, we propose MATSA, the first Magnetoresistive RAM (MRAM)-based Accelerator for TSA. MATSA leverages Processing-Using-Memory (PUM) based on MRAM crossbars to minimize data movement overheads and exploit parallelism in sDTW. MATSA improves performance by 7.35×/6.15×/6.31× and energy efficiency by 11.29×/4.21×/2.65× over server-class CPU, GPU, and Processing-Near-Memory platforms, respectively.
PB IEEE Computer Society
YR 2024
FD 2024
LK https://hdl.handle.net/10630/38997
UL https://hdl.handle.net/10630/38997
LA eng
NO Ivan Fernandez, Christina Giannoula, Aditya Manglik, Ricardo Quislant, Nika Mansouri-Ghiasi, Juan Gómez-Luna, Eladio Gutiérrez, Oscar G. Plata, Onur Mutlu. MATSA: An MRAM-Based Energy-Efficient Accelerator for Time Series Analysis. IEEE Access 12: 36727-36742 (2024)
NO Gobierno de España
NO High Performance, Edge And Cloud computing (HiPEAC)
NO SAFARI Group’s Industrial Partners
NO Semiconductor Research Corporation
DS RIUMA. Repositorio Institucional de la Universidad de Málaga
RD 20 ene 2026