Evaluation of heterogeneous AIoT Accelerators within VEDLIoT
Paper i proceeding, 2023

Within VEDLIoT, a project targeting the development of energy-efficient Deep Learning for distributed AIoT applications, several accelerator platforms based on technologies like CPUs, embedded GPUs, FPGAs, or specialized ASICs are evaluated. The VEDLIoT approach is based on modular and scalable cognitive IoT hardware platforms. Modular microserver technology enables the integration of different, heterogeneous accelerators into one platform. Benchmarking of the different accelerators takes into account performance, energy efficiency and accuracy. The results in this paper provide a solid overview regarding available accelerator solutions and provide guidance for hardware selection for AIoT applications from far edge to cloud. VEDLIoT is an H2020 EU project which started in November 2020. It is currently in an intermediate stage. The focus is on the considerations of the performance and energy efficiency of hardware accelerators. Apart from the hardware and accelerator focus presented in this paper, the project also covers toolchain, security and safety aspects. The resulting technology is tested on a wide range of AIoT applications.

Författare

R. Griessl

Universität Bielefeld

F. Porrmann

Universität Bielefeld

Nils Kucza

Universität Bielefeld

K. Mika

Universität Bielefeld

Jens Hagemeyer

Universität Bielefeld

Martin Kaiser

Universität Bielefeld

Mario Porrmann

Universität Bielefeld

M. Tassemeier

Universität Osnabrück

M. Flottmann

Universität Osnabrück

Fareed Mohammad Qararyah

Chalmers, Data- och informationsteknik, Datorteknik

Muhammad Waqar Azhar

Chalmers, Data- och informationsteknik, Datorteknik

Pedro Petersen Moura Trancoso

Chalmers, Data- och informationsteknik, Datorteknik

D. Odman

EmbeDL AB

K. Gugala

Antmicro

G. Latosinski

Antmicro

Proceedings -Design, Automation and Test in Europe, DATE

15301591 (ISSN)

Vol. 2023-April
9783981926378 (ISBN)

2023 Design, Automation and Test in Europe Conference and Exhibition, DATE 2023
Antwerp, Belgium,

Very Efficient Deep Learning in IOT (VEDLIoT)

Europeiska kommissionen (EU) (EC/H2020/957197), 2020-11-01 -- 2023-10-31.

Ämneskategorier

Programvaruteknik

Datorsystem

Mer information

Senast uppdaterat

2024-07-09