Software Implementation of 10G-EPON Downstream Physical-Layer Processing Adopting CPU-GPU Cooperative Computing for Flexible Access Systems

Software Implementation of 10G-EPON Downstream Physical-Layer Processing Adopting CPU-GPU Cooperative Computing for Flexible Access Systems

The application of network function virtualization (NFV) and software-defined networks (SDNs) to optical access systems continues to attract a lot of attention. Their use on general-purpose hardware allows for cost-effective implementation and quick response to functional requirements. This paper de...

Full description

Bibliographic Details
Main Authors: Takahiro Suzuki, Sang-Yuep Kim, Jun-Ichi Kani, Jun Terada
Format: Article
Language:English
Published: IEEE 2019-01-01
Series:IEEE Access
Subjects:
SDN
NFV
access networks
GPU
PHY coding and scrambler
Online Access:https://ieeexplore.ieee.org/document/8664669/
id doaj-575e1cac92654dd2a505d97da60f7eb6
record_format Article
spelling doaj-575e1cac92654dd2a505d97da60f7eb62021-03-29T22:57:56ZengIEEEIEEE Access2169-35362019-01-017338883389710.1109/ACCESS.2019.29040838664669Software Implementation of 10G-EPON Downstream Physical-Layer Processing Adopting CPU-GPU Cooperative Computing for Flexible Access SystemsTakahiro Suzuki0https://orcid.org/0000-0001-5850-1172Sang-Yuep Kim1Jun-Ichi Kani2Jun Terada3NTT Access Network Service Systems Laboratories, NTT Corporation, Yokosuka, JapanNTT Access Network Service Systems Laboratories, NTT Corporation, Yokosuka, JapanNTT Access Network Service Systems Laboratories, NTT Corporation, Yokosuka, JapanNTT Access Network Service Systems Laboratories, NTT Corporation, Yokosuka, JapanThe application of network function virtualization (NFV) and software-defined networks (SDNs) to optical access systems continues to attract a lot of attention. Their use on general-purpose hardware allows for cost-effective implementation and quick response to functional requirements. This paper demonstrates the softwarization of the complete downstream physical (PHY)-layer functions of the optical line terminal (OLT), including a scrambler, which uses a serial processing algorithm that cannot be parallelized by a general-purpose graphics processing unit (GPU). We propose CPU-GPU cooperative implementation architecture that softwarizes the complete 10G-EPON OLT downstream PHY. We achieve 10.3125-Gbps real-time performance through experiments for the first time.https://ieeexplore.ieee.org/document/8664669/SDNNFVaccess networksGPUPHY coding and scrambler
collection DOAJ
language English
format Article
sources DOAJ
author Takahiro Suzuki
Sang-Yuep Kim
Jun-Ichi Kani
Jun Terada
spellingShingle Takahiro Suzuki
Sang-Yuep Kim
Jun-Ichi Kani
Jun Terada
Software Implementation of 10G-EPON Downstream Physical-Layer Processing Adopting CPU-GPU Cooperative Computing for Flexible Access Systems
IEEE Access
SDN
NFV
access networks
GPU
PHY coding and scrambler
author_facet Takahiro Suzuki
Sang-Yuep Kim
Jun-Ichi Kani
Jun Terada
author_sort Takahiro Suzuki
title Software Implementation of 10G-EPON Downstream Physical-Layer Processing Adopting CPU-GPU Cooperative Computing for Flexible Access Systems
title_short Software Implementation of 10G-EPON Downstream Physical-Layer Processing Adopting CPU-GPU Cooperative Computing for Flexible Access Systems
title_full Software Implementation of 10G-EPON Downstream Physical-Layer Processing Adopting CPU-GPU Cooperative Computing for Flexible Access Systems
title_fullStr Software Implementation of 10G-EPON Downstream Physical-Layer Processing Adopting CPU-GPU Cooperative Computing for Flexible Access Systems
title_full_unstemmed Software Implementation of 10G-EPON Downstream Physical-Layer Processing Adopting CPU-GPU Cooperative Computing for Flexible Access Systems
title_sort software implementation of 10g-epon downstream physical-layer processing adopting cpu-gpu cooperative computing for flexible access systems
publisher IEEE
series IEEE Access
issn 2169-3536
publishDate 2019-01-01
description The application of network function virtualization (NFV) and software-defined networks (SDNs) to optical access systems continues to attract a lot of attention. Their use on general-purpose hardware allows for cost-effective implementation and quick response to functional requirements. This paper demonstrates the softwarization of the complete downstream physical (PHY)-layer functions of the optical line terminal (OLT), including a scrambler, which uses a serial processing algorithm that cannot be parallelized by a general-purpose graphics processing unit (GPU). We propose CPU-GPU cooperative implementation architecture that softwarizes the complete 10G-EPON OLT downstream PHY. We achieve 10.3125-Gbps real-time performance through experiments for the first time.
topic SDN
NFV
access networks
GPU
PHY coding and scrambler
url https://ieeexplore.ieee.org/document/8664669/
work_keys_str_mv AT takahirosuzuki softwareimplementationof10gepondownstreamphysicallayerprocessingadoptingcpugpucooperativecomputingforflexibleaccesssystems
AT sangyuepkim softwareimplementationof10gepondownstreamphysicallayerprocessingadoptingcpugpucooperativecomputingforflexibleaccesssystems
AT junichikani softwareimplementationof10gepondownstreamphysicallayerprocessingadoptingcpugpucooperativecomputingforflexibleaccesssystems
AT junterada softwareimplementationof10gepondownstreamphysicallayerprocessingadoptingcpugpucooperativecomputingforflexibleaccesssystems
_version_ 1724190428010381312

Similar Items