Aerial: 5G GPU超融合平台

Anupa Kelkar, C. Dick
{"title":"Aerial: 5G GPU超融合平台","authors":"Anupa Kelkar, C. Dick","doi":"10.1145/3472716.3472864","DOIUrl":null,"url":null,"abstract":"In addition to high-throughput, low-latency and high-reliability, softwarization, virtualization and open interfaces are key tenets of 5G wireless. In this paper we present Aerial, a pure software implementation of a MIMO multi-cell gNB. All of the baseband signal processing is implemented as CUDA running on an NVIDA GPU, layer-2 is hosted on a CPU and an O-RAN 7.2 split fronthaul interface is supported with a Mellanox ConnectX-6 DX NIC. The accompanying video demonstrates the three use-cases that are described in this paper.","PeriodicalId":178725,"journal":{"name":"Proceedings of the SIGCOMM '21 Poster and Demo Sessions","volume":"22 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2021-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":"{\"title\":\"Aerial: a GPU hyper-converged platform for 5G\",\"authors\":\"Anupa Kelkar, C. Dick\",\"doi\":\"10.1145/3472716.3472864\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"In addition to high-throughput, low-latency and high-reliability, softwarization, virtualization and open interfaces are key tenets of 5G wireless. In this paper we present Aerial, a pure software implementation of a MIMO multi-cell gNB. All of the baseband signal processing is implemented as CUDA running on an NVIDA GPU, layer-2 is hosted on a CPU and an O-RAN 7.2 split fronthaul interface is supported with a Mellanox ConnectX-6 DX NIC. The accompanying video demonstrates the three use-cases that are described in this paper.\",\"PeriodicalId\":178725,\"journal\":{\"name\":\"Proceedings of the SIGCOMM '21 Poster and Demo Sessions\",\"volume\":\"22 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2021-08-23\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"2\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Proceedings of the SIGCOMM '21 Poster and Demo Sessions\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1145/3472716.3472864\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of the SIGCOMM '21 Poster and Demo Sessions","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/3472716.3472864","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 2

摘要

除了高吞吐量、低延迟和高可靠性之外,软件化、虚拟化和开放接口是5G无线的关键原则。在本文中,我们提出了天线,一个纯软件实现的MIMO多小区gNB。所有基带信号处理都是在nvidia GPU上运行的CUDA实现的,第二层是在CPU上托管的,O-RAN 7.2分裂前传接口由Mellanox ConnectX-6 DX网卡支持。附带的视频演示了本文中描述的三个用例。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Aerial: a GPU hyper-converged platform for 5G
In addition to high-throughput, low-latency and high-reliability, softwarization, virtualization and open interfaces are key tenets of 5G wireless. In this paper we present Aerial, a pure software implementation of a MIMO multi-cell gNB. All of the baseband signal processing is implemented as CUDA running on an NVIDA GPU, layer-2 is hosted on a CPU and an O-RAN 7.2 split fronthaul interface is supported with a Mellanox ConnectX-6 DX NIC. The accompanying video demonstrates the three use-cases that are described in this paper.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信