Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

SpecPCM: A Low-power PCM-based In-Memory Computing Accelerator for Full-stack Mass Spectrometry Analysis

Keming Fan, Ashkan Moradifirouzabadi, Xiangjin Wu, Zheyu Li, Flavio Ponzina, Anton Persson, Eric Pop, Tajana Rosing, Mingu Kang

TL;DR

SpecPCM, an in-memory computing (IMC) accelerator designed to achieve substantial improvements in energy and delay efficiency for both MS spectral clustering and database search, leverages a robust hyperdimensional computing algorithm with a novel dimension-packing method and develops specialized hardware for the end-to-end MS pipeline to overcome the nonideal behavior of PCM devices.

Abstract

Mass spectrometry (MS) is essential for proteomics and metabolomics but faces impending challenges in efficiently processing the vast volumes of data. This paper introduces SpecPCM, an in-memory computing (IMC) accelerator designed to achieve substantial improvements in energy and delay efficiency for both MS spectral clustering and database (DB) search. SpecPCM employs analog processing with low-voltage swing and utilizes recently introduced phase change memory (PCM) devices based on superlattice materials, optimized for low-voltage and low-power programming. Our approach integrates contributions across multiple levels: application, algorithm, circuit, device, and instruction sets. We leverage a robust hyperdimensional computing (HD) algorithm with a novel dimension-packing method and develop specialized hardware for the end-to-end MS pipeline to overcome the non-ideal behavior of PCM devices. We further optimize multi-level PCM devices for different tasks by using different materials. We also perform a comprehensive design exploration to improve energy and delay efficiency while maintaining accuracy, exploring various combinations of hardware and software parameters controlled by the instruction set architecture (ISA). SpecPCM, with up to three bits per cell, achieves speedups of up to 82x and 143x for MS clustering and DB search tasks, respectively, along with a four-orders-of-magnitude improvement in energy efficiency compared with state-of-the-art CPU/GPU tools.

SpecPCM: A Low-power PCM-based In-Memory Computing Accelerator for Full-stack Mass Spectrometry Analysis

TL;DR

SpecPCM, an in-memory computing (IMC) accelerator designed to achieve substantial improvements in energy and delay efficiency for both MS spectral clustering and database search, leverages a robust hyperdimensional computing algorithm with a novel dimension-packing method and develops specialized hardware for the end-to-end MS pipeline to overcome the nonideal behavior of PCM devices.

Abstract

Mass spectrometry (MS) is essential for proteomics and metabolomics but faces impending challenges in efficiently processing the vast volumes of data. This paper introduces SpecPCM, an in-memory computing (IMC) accelerator designed to achieve substantial improvements in energy and delay efficiency for both MS spectral clustering and database (DB) search. SpecPCM employs analog processing with low-voltage swing and utilizes recently introduced phase change memory (PCM) devices based on superlattice materials, optimized for low-voltage and low-power programming. Our approach integrates contributions across multiple levels: application, algorithm, circuit, device, and instruction sets. We leverage a robust hyperdimensional computing (HD) algorithm with a novel dimension-packing method and develop specialized hardware for the end-to-end MS pipeline to overcome the non-ideal behavior of PCM devices. We further optimize multi-level PCM devices for different tasks by using different materials. We also perform a comprehensive design exploration to improve energy and delay efficiency while maintaining accuracy, exploring various combinations of hardware and software parameters controlled by the instruction set architecture (ISA). SpecPCM, with up to three bits per cell, achieves speedups of up to 82x and 143x for MS clustering and DB search tasks, respectively, along with a four-orders-of-magnitude improvement in energy efficiency compared with state-of-the-art CPU/GPU tools.

Paper Structure

This paper contains 19 sections, 1 equation, 15 figures, 6 tables.

Table of Contents

  1. Introduction
  2. Background and Motivation
  3. Hyperdimensional computing
  4. Mass Spectrometry and its accelerations
  5. IMC for MS analysis and application of PCM
  6. Proposed SpecPCM using Analog IMC based on PCM
  7. SpecPCM overview
  8. Dimension packing in HD algorithm
  9. Hardware Operations
  10. Efficiency vs. Accuracy Trade-offs
  11. PCM Device Optimization
  12. Instruction Set
  13. Evaluation
  14. Experimental Setup
  15. Experiment Results
  16. ...and 4 more sections

Figures (15)

  • Figure 1: Overview of spectral clustering for MS analysis.
  • Figure 2: Overview of MS database (DB) search.
  • Figure 3: Latency breakdown for GPU tools, (a) Clustering on dataset PXD000561 using HyperSpec, (b) Database search on dataset HEK293 using HyperOMS.
  • Figure 4: Overview of the SpecPCM accelerator.
  • Figure 5: Dimension packing for MLC with 3 bits per cell.
  • ...and 10 more figures