MATSA: An MRAM-Based Energy-Efficient Accelerator for Time Series Analysis

Abstract

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\times /6.15\times /6.31\times $ and energy efficiency by $11.29\times /4.21\times /2.65\times $ over server-class CPU, GPU, and Processing-Near-Memory platforms, respectively.

Keywords

• Time series analysis
• processing-using-memory
• memory-bound
• emerging technologies