基于软件电路的快速包转发引擎

M. Makkes, A. Varbanescu, C. D. Laat, R. Meijer
{"title":"基于软件电路的快速包转发引擎","authors":"M. Makkes, A. Varbanescu, C. D. Laat, R. Meijer","doi":"10.1145/2742854.2742862","DOIUrl":null,"url":null,"abstract":"Forwarding packets is part of the performance critical path of routing devices, and affects the network performance at any scale. This operation is typically performed by dedicated routing boxes, which are fast, but expensive and inflexible. Recent work has shown that in many cases commodity hardware is becoming an alternative to these specialized boxes. In this work, we present a new technique - based on bitslicing - to improve the performance of forward decision-making on modern commodity hardware. Specifically, we propose to replace memory lookups with logical operations, by evaluating the packet header information as a Boolean circuit. Being less memory-intensive, our algorithm has the potential to achieve high performance on both modern CPUs and GPUs. To measure and qulify the performance of our algorithm, we implemented it in OpenCL and performed a large set of experiments on 5 different platforms - two CPUs and three GPUs. Our results show that bitslicing has the ability to outperform the traditional, memory lookup approach in 70% of the cases, depending on the type of traffic and routing parameters.","PeriodicalId":417279,"journal":{"name":"Proceedings of the 12th ACM International Conference on Computing Frontiers","volume":"1 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2015-05-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Fast packet forwarding engine based on software circuits\",\"authors\":\"M. Makkes, A. Varbanescu, C. D. Laat, R. Meijer\",\"doi\":\"10.1145/2742854.2742862\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Forwarding packets is part of the performance critical path of routing devices, and affects the network performance at any scale. This operation is typically performed by dedicated routing boxes, which are fast, but expensive and inflexible. Recent work has shown that in many cases commodity hardware is becoming an alternative to these specialized boxes. In this work, we present a new technique - based on bitslicing - to improve the performance of forward decision-making on modern commodity hardware. Specifically, we propose to replace memory lookups with logical operations, by evaluating the packet header information as a Boolean circuit. Being less memory-intensive, our algorithm has the potential to achieve high performance on both modern CPUs and GPUs. To measure and qulify the performance of our algorithm, we implemented it in OpenCL and performed a large set of experiments on 5 different platforms - two CPUs and three GPUs. Our results show that bitslicing has the ability to outperform the traditional, memory lookup approach in 70% of the cases, depending on the type of traffic and routing parameters.\",\"PeriodicalId\":417279,\"journal\":{\"name\":\"Proceedings of the 12th ACM International Conference on Computing Frontiers\",\"volume\":\"1 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2015-05-06\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Proceedings of the 12th ACM International Conference on Computing Frontiers\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1145/2742854.2742862\",\"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 12th ACM International Conference on Computing Frontiers","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/2742854.2742862","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

摘要

报文转发是路由设备性能关键路径的一部分,对网络性能的影响是任何规模的。此操作通常由专用路由箱执行,速度快,但价格昂贵且不灵活。最近的研究表明,在许多情况下,商品硬件正在成为这些专用盒子的替代品。在这项工作中,我们提出了一种基于位切片的新技术来提高现代商用硬件上的前向决策性能。具体来说,我们建议用逻辑操作代替内存查找,通过将包头信息作为布尔电路进行评估。由于内存消耗较少,我们的算法有可能在现代cpu和gpu上实现高性能。为了测量和确定我们的算法的性能,我们在OpenCL中实现了它,并在5个不同的平台(两个cpu和三个gpu)上进行了大量的实验。我们的结果表明,根据流量类型和路由参数的不同,位切片在70%的情况下能够优于传统的内存查找方法。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Fast packet forwarding engine based on software circuits
Forwarding packets is part of the performance critical path of routing devices, and affects the network performance at any scale. This operation is typically performed by dedicated routing boxes, which are fast, but expensive and inflexible. Recent work has shown that in many cases commodity hardware is becoming an alternative to these specialized boxes. In this work, we present a new technique - based on bitslicing - to improve the performance of forward decision-making on modern commodity hardware. Specifically, we propose to replace memory lookups with logical operations, by evaluating the packet header information as a Boolean circuit. Being less memory-intensive, our algorithm has the potential to achieve high performance on both modern CPUs and GPUs. To measure and qulify the performance of our algorithm, we implemented it in OpenCL and performed a large set of experiments on 5 different platforms - two CPUs and three GPUs. Our results show that bitslicing has the ability to outperform the traditional, memory lookup approach in 70% of the cases, depending on the type of traffic and routing parameters.
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:604180095
Book学术官方微信