A Survey on Machine Learning Accelerators and Evolutionary Hardware Platforms

Sathwika Bavikadi; Abhijitt Dhavlle; Amlan Ganguly; Anand Haridass; Hagar Hendy; Cory Merkel; Vijay Janapa Reddi; Purab Ranjan Sutradhar; Arun Joseph; Sai Manoj Pudukotai Dinakarrao

doi:10.1109/MDAT.2022.3161126

A Survey on Machine Learning Accelerators and Evolutionary Hardware Platforms

Sathwika Bavikadi, Abhijitt Dhavlle, Amlan Ganguly, Anand Haridass, Hagar Hendy, Cory Merkel, Vijay Janapa Reddi, Purab Ranjan Sutradhar, Arun Joseph, Sai Manoj Pudukotai Dinakarrao

Electrical Engineering Department

Research output: Contribution to journal › Article › peer-review

33 Scopus citations

Abstract

Advanced computing systems have long been enablers for breakthroughs in artificial intelligence (AI) and machine learning (ML) algorithms, either through sheer computational power or form-factor miniaturization. However, as AI/ML algorithms become more complex and the size of data sets increases, existing computing platforms are no longer sufficient to bridge the gap between algorithmic innovation and hardware design. This article presents a survey about various ML accelerators.

Original language	English
Pages (from-to)	91-116
Number of pages	26
Journal	IEEE Design and Test
Volume	39
Issue number	3
DOIs	https://doi.org/10.1109/MDAT.2022.3161126
State	Published - 1 Jun 2022

Keywords

ASICs
Accelerators
EDA
Energy Efficiency
FPGAs
Hardware Design
In-memory computing
ML as Service
Machine learning
Microservices
Neuromorphic computing
Processing-in-memory

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1109/MDAT.2022.3161126

Cite this

@article{c8c31066e5474e49b8f0ef7dcf0b3045,

title = "A Survey on Machine Learning Accelerators and Evolutionary Hardware Platforms",

abstract = "Advanced computing systems have long been enablers for breakthroughs in artificial intelligence (AI) and machine learning (ML) algorithms, either through sheer computational power or form-factor miniaturization. However, as AI/ML algorithms become more complex and the size of data sets increases, existing computing platforms are no longer sufficient to bridge the gap between algorithmic innovation and hardware design. This article presents a survey about various ML accelerators.",

keywords = "ASICs, Accelerators, EDA, Energy Efficiency, FPGAs, Hardware Design, In-memory computing, ML as Service, Machine learning, Microservices, Neuromorphic computing, Processing-in-memory",

author = "Sathwika Bavikadi and Abhijitt Dhavlle and Amlan Ganguly and Anand Haridass and Hagar Hendy and Cory Merkel and Reddi, {Vijay Janapa} and Sutradhar, {Purab Ranjan} and Arun Joseph and {Pudukotai Dinakarrao}, {Sai Manoj}",

note = "Publisher Copyright: {\textcopyright} 2013 IEEE.",

year = "2022",

month = jun,

day = "1",

doi = "10.1109/MDAT.2022.3161126",

language = "English",

volume = "39",

pages = "91--116",

number = "3",

}

TY - JOUR

T1 - A Survey on Machine Learning Accelerators and Evolutionary Hardware Platforms

AU - Bavikadi, Sathwika

AU - Dhavlle, Abhijitt

AU - Ganguly, Amlan

AU - Haridass, Anand

AU - Hendy, Hagar

AU - Merkel, Cory

AU - Reddi, Vijay Janapa

AU - Sutradhar, Purab Ranjan

AU - Joseph, Arun

AU - Pudukotai Dinakarrao, Sai Manoj

PY - 2022/6/1

Y1 - 2022/6/1

N2 - Advanced computing systems have long been enablers for breakthroughs in artificial intelligence (AI) and machine learning (ML) algorithms, either through sheer computational power or form-factor miniaturization. However, as AI/ML algorithms become more complex and the size of data sets increases, existing computing platforms are no longer sufficient to bridge the gap between algorithmic innovation and hardware design. This article presents a survey about various ML accelerators.

AB - Advanced computing systems have long been enablers for breakthroughs in artificial intelligence (AI) and machine learning (ML) algorithms, either through sheer computational power or form-factor miniaturization. However, as AI/ML algorithms become more complex and the size of data sets increases, existing computing platforms are no longer sufficient to bridge the gap between algorithmic innovation and hardware design. This article presents a survey about various ML accelerators.

KW - ASICs

KW - Accelerators

KW - EDA

KW - Energy Efficiency

KW - FPGAs

KW - Hardware Design

KW - In-memory computing

KW - ML as Service

KW - Machine learning

KW - Microservices

KW - Neuromorphic computing

KW - Processing-in-memory

UR - http://www.scopus.com/inward/record.url?scp=85127060383&partnerID=8YFLogxK

U2 - 10.1109/MDAT.2022.3161126

DO - 10.1109/MDAT.2022.3161126

M3 - Article

AN - SCOPUS:85127060383

SN - 2168-2356

VL - 39

SP - 91

EP - 116

JO - IEEE Design and Test

JF - IEEE Design and Test

IS - 3

ER -

A Survey on Machine Learning Accelerators and Evolutionary Hardware Platforms

Abstract

Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this