FlutPIM: A Look-Up Table-Based Processing in Memory Architecture with Floating-Point Computation Support for Deep Learning Applications

Purab Ranjan Sutradhar, Sathwika Bavikadi, Mark A. Indovina, Sai Manoj Pudukotai Dinakarrao, Amlan Ganguly

Research output: Chapter in Book/Report/Conference proceedingChapter

Abstract

Processing-in-Memory (PIM) has shown great potential for a wide range of data-driven applications, especially Deep Learning and AI. However, it is a challenge to facilitate the computational sophistication of a standard processor (i.e. CPU or GPU) within the limited scope of a memory chip without contributing significant circuit overheads. To address the challenge, we propose a programmable LUT-based area-efficient PIM architecture capable of performing various low-precision floating point (FP) computations using a novel LUT-oriented operand-decomposition technique. We incorporate such compact computational units within the memory banks in a large count to achieve impressive parallel processing capabilities, up to 4x higher than state-of-the-art FP-capable PIM. Additionally, we adopt a highly-optimized low-precision FP format that maximizes computational performance at a minimal compromise of computational precision, especially for Deep Learning Applications. The overall result is a 17% higher throughput and an impressive 8-20x higher compute Bandwidth/bank compared to the state-of-the-art of in-memory acceleration.
Original languageAmerican English
Title of host publicationGLSVLSI '23: Proceedings of the Great Lakes Symposium on VLSI 2023
DOIs
StatePublished - 2023
Externally publishedYes

Keywords

  • DRAM
  • Deep Learning
  • floating point
  • processing in memory

EGS Disciplines

  • Electrical and Computer Engineering

Fingerprint

Dive into the research topics of 'FlutPIM: A Look-Up Table-Based Processing in Memory Architecture with Floating-Point Computation Support for Deep Learning Applications'. Together they form a unique fingerprint.

Cite this