Uniform handling and abstraction of NFV hardware accelerators

Author

Bronstein, Zvika ; Roch, Evelyne ; Jinwei Xia ; Molkho, Adi

Author_Institution

Huawei Technol., Shenzhen, China

Volume

29

Issue

3

fYear

2015

fDate

May-June 2015

Firstpage

22

Lastpage

29

Abstract

This article outlines some NFV topics related to implementing VNF over NFVI where HWA may be needed. The document highlights the commonalities and differences among various HWA approaches and suggests uniform handling, common architecture, and abstraction layer for the hardware acceleration components. This uniform handling will allow deployment of various hardware accelerators within NFVI and facilitate interoperability between various VNFs and HWAs.

Keywords

graphics processing units; open systems; virtual private networks; virtualisation; HWA approach; NFV hardware accelerator abstraction; NFV hardware accelerator handling; NFVI; VNF; abstraction layer; common architecture; hardware acceleration components; interoperability; network function virtualization infrastructure; uniform handling; Accelerators; Dynamic scheduling; Hardware; Integrated circuits; Network architecture; Resource management; Telecommunication network management; Virtualization;

fLanguage

English

Journal_Title

Network, IEEE

Publisher

ieee

ISSN

0890-8044

Type

jour

DOI

10.1109/MNET.2015.7113221

Filename

7113221