On Retargeting the AI Programming Framework to New Hardwares - Network and Parallel Computing
Conference Papers Year : 2018

On Retargeting the AI Programming Framework to New Hardwares

Abstract

Nowadays, a large number of accelerators are proposed to increase the performance of AI applications, making it a big challenge to enhance existing AI programming frameworks to support these new accelerators. In this paper, we select TensorFlow to demonstrate how to port the AI programming framework to new hardwares, i.e., FPGA and Sunway TaihuLight here. FPGA and Sunway TaihuLight represent two distinct and significant hardware architectures for considering the retargeting process. We introduce our retargeting processes and experiences for these two platforms, from the source codes to the compilation processes. We compare the two retargeting approaches and demonstrate some preliminary experimental results.

Domains

Computer Science [cs]
Fichier principal
Vignette du fichier
477597_1_En_4_Chapter.pdf (336.25 Ko) Télécharger le fichier
Origin Files produced by the author(s)

Hal Ifip  : Connect in order to contact the contributor

https://inria.hal.science/hal-02279561

Submitted on : Thursday, September 5, 2019-1:31:41 PM

Last modification on : Tuesday, November 12, 2024-11:34:03 AM

Long-term archiving on : Thursday, February 6, 2020-4:15:58 AM

Download

Dates and versions

hal-02279561 , version 1 (05-09-2019)

Licence

Attribution

Identifiers

Cite

Jiacheng Zhao, Yisong Chang, Denghui Li, Chunwei Xia, Huimin Cui, et al.. On Retargeting the AI Programming Framework to New Hardwares. 15th IFIP International Conference on Network and Parallel Computing (NPC), Nov 2018, Muroran, Japan. pp.39-51, ⟨10.1007/978-3-030-05677-3_4⟩. ⟨hal-02279561⟩

Export

BibTeX XML-TEI Dublin Core DC Terms EndNote DataCite

Collections

IFIP-LNCS IFIP IFIP-TC IFIP-TC10 IFIP-NPC IFIP-WG10-3 IFIP-LNCS-11276
74 View
142 Download

Altmetric

Share

More