Interconnect-Centric Benchmarking of In-Memory Acceleration for DNNS

Gakul Krishnan, Sumit K. Mandai, Chaitali Chakrabarti, Jae Sun Seo, Umit Ogras, Yu Cao

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Scopus citations

Abstract

In-memory computing (IMC) provides a dense and parallel structure for high performance and energy-efficient acceleration of deep neural networks (DNNs). The increased computational density of IMC architectures results in increased on -chip communication costs, stressing the interconnect fabric. In this work, we develop a novel performance benchmark tool for IMC architectures that incorporates device, circuits, architecture, and interconnect under a single roof. The tool assesses the area, energy, and latency of the IMC accelerator. We analyze three interconnect cases to illustrate the versatility of the tool: (1) Point-to-point (P2P) and network-on-chip (NoC) based IMC architectures to demonstrate the criticality of the interconnect choice; (2) Area and energy optimization to improve IMC utilization and reduce on-chip interconnect cost; (3) Evaluation of a reconfigurable NoC to achieve minimum on-chip communication latency. Through these studies, we motivate the need for future work in the design of optimal on-chip and off-chip interconnect fabrics for IMC architectures.

Original languageEnglish (US)
Title of host publicationChina Semiconductor Technology International Conference 2021, CSTIC 2021
EditorsCor Claeys, Steve X. Liang, Qinghuang Lin, Ru Huang, Hanming Wu, Peilin Song, Linyong Pang, Ying Zhang, Beichao Zhang, Xinping Xinping Qu, Cheng Zhuo, Hsiang-Lan Lung
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781665449458
DOIs
StatePublished - Mar 14 2021
Event2021 China Semiconductor Technology International Conference, CSTIC 2021 - Shanghai, China
Duration: Mar 14 2021Mar 15 2021

Publication series

NameChina Semiconductor Technology International Conference 2021, CSTIC 2021

Conference

Conference2021 China Semiconductor Technology International Conference, CSTIC 2021
Country/TerritoryChina
CityShanghai
Period3/14/213/15/21

ASJC Scopus subject areas

  • Hardware and Architecture
  • Electrical and Electronic Engineering
  • Electronic, Optical and Magnetic Materials

Fingerprint

Dive into the research topics of 'Interconnect-Centric Benchmarking of In-Memory Acceleration for DNNS'. Together they form a unique fingerprint.

Cite this