NaaS: Network-as-a-Service in the Cloud

NaaS:云中的网络即服务

基本信息

  • 批准号:
    EP/K032968/1
  • 负责人:
  • 金额:
    $ 84.88万
  • 依托单位:
  • 依托单位国家:
    英国
  • 项目类别:
    Research Grant
  • 财政年份:
    2013
  • 资助国家:
    英国
  • 起止时间:
    2013 至 无数据
  • 项目状态:
    已结题

项目摘要

Cloud computing has significantly changed the IT landscape. Today it is possible for small companies or even single individuals to access virtually unlimited resources in large data centres (DCs) for running computationally demanding tasks. This has triggered the rise of "big data" applications, which operate on large amounts of data. These include traditional batch-oriented applications, such as data mining, data indexing, log collection and analysis, and scientific applications, as well as real-time stream processing, web search and advertising.To support big data applications, parallel processing systems, such as MapReduce, adopt a partition/aggregate model: a large input data set is distributed over many servers, and each server processes a share of the data. Locally generated intermediate results must then be aggregated to obtain the final result.An open challenge of the partition/aggregate model is that it results in high contention for network resources in DCs when a large amount of data traffic is exchanged between servers. Facebook reports that, for 26% of processing tasks, network transfers are responsible for more than 50% of the execution time. This is consistent with other studies, showing that the network is often the bottleneck in big data applications.Improving the performance of such network-bound applications in DCs has attracted much interest from the research community. A class of solutions focuses on reducing bandwidth usage by employing overlay networks to distribute data and to perform partial aggregation. However, this requires applications to reverse-engineer the physical network topology to optimise the layout of overlay networks. Even with perfect knowledge of the physical topology, there are still fundamental inefficiencies: e.g. any logical topology with a server fan-out higher than one cannot be mapped optimally to the physical network if servers have only a single network interface. Other proposals increase network bandwidth through more complex topologies or higher-capacity networks. New topologies and network over-provisioning, however, increase the DC operational and capital expenditures-up to 5 times according to some estimates-which directly impacts tenant costs. For example, Amazon AWS recently introduced Cluster Compute instances with full-bisection 10 Gbps bandwidth, with an hourly cost of 16 times the default. In contrast, we argue that the problem can be solved more effectively by providing DC tenants with efficient, easy and safe control of network operations. Instead of over-provisioning, we focus on optimising network traffic by exploiting application-specific knowledge. We term this approach "network-as-a-service" (NaaS) because it allows tenants to customise the service that they receive from the network. NaaS-enabled tenants can deploy custom routing protocols, including multicast services or anycast/incast protocols, as well as more sophisticated mechanisms, such as content-based routing and content-centric networking.By modifying the content of packets on-path, they can efficiently implement advanced, application-specific network services, such as in-network data aggregation and smart caching. Parallel processing systems such as MapReduce would greatly benefit because data can be aggregated on-path, thus reducing execution times. Key-value stores (e.g. memcached) can improve their performance by caching popular keys within the network, which decreases latency and bandwidth usage compared to end-host-only deployments.The NaaS model has the potential to revolutionise current cloud computing offerings by increasing the performance of tenants' applications -through efficient in-network processing- while reducing development complexity. It aims to combine distributed computation and network communication in a single, coherent abstraction, providing a significant step towards the vision of "the DC is the computer".
云计算显著改变了IT格局。如今,小公司甚至个人都可以访问大型数据中心(DC)中几乎无限的资源,以运行计算要求高的任务。这引发了“大数据”应用程序的兴起,这些应用程序对大量数据进行操作。这些应用包括传统的面向批处理的应用,如数据挖掘、数据索引、日志收集和分析、科学应用,以及实时流处理、网络搜索和广告。为了支持大数据应用,并行处理系统,如MapReduce,采用分区/聚合模型:一个大的输入数据集分布在许多服务器上,每个服务器处理一部分数据。然后必须聚合本地生成的中间结果才能获得最终结果。分区/聚合模型的一个开放挑战是,当服务器之间交换大量数据流量时,它会导致DC中网络资源的高度争用。Facebook报告称,在26%的处理任务中,网络传输占用了超过50%的执行时间。这与其他研究一致,表明网络往往是大数据应用的瓶颈。提高DC中此类网络绑定应用的性能引起了研究界的极大兴趣。一类解决方案集中于通过采用覆盖网络来分发数据并执行部分聚合来减少带宽使用。然而,这需要应用程序对物理网络拓扑进行逆向工程,以优化覆盖网络的布局。即使完全了解物理拓扑,仍然存在根本的低效率:例如,如果服务器只有一个网络接口,则任何服务器扇出大于1的逻辑拓扑都不能最佳地映射到物理网络。其他建议通过更复杂的拓扑或更高容量的网络来增加网络带宽。然而,新的拓扑和网络过度配置会增加DC的运营和资本支出(据估计高达5倍),这会直接影响租户成本。例如,Amazon AWS最近推出了具有全平分10 Gbps带宽的集群计算实例,每小时成本是默认值的16倍。相比之下,我们认为,这个问题可以更有效地解决提供DC租户的网络操作的高效,简便和安全的控制。而不是过度配置,我们专注于优化网络流量,利用特定于应用程序的知识。我们将这种方法称为“网络即服务”(NaaS),因为它允许租户定制他们从网络接收的服务。支持NaaS的租户可以部署自定义路由协议,包括多播服务或任播/incast协议,以及更复杂的机制,如基于内容的路由和以内容为中心的网络。通过修改路径上数据包的内容,他们可以有效地实现高级的特定于应用的网络服务,如网内数据聚合和智能缓存。MapReduce等并行处理系统将大大受益,因为数据可以在路径上聚合,从而减少执行时间。键值存储(例如memcached)可以通过在网络中缓存流行的键来提高其性能,与仅终端主机部署相比,这可以减少延迟和带宽使用。NaaS模型有可能通过提高租户应用程序的性能(通过高效的网络内处理)来彻底改变当前的云计算产品,同时降低开发复杂性。它旨在将联合收割机分布式计算和网络通信结合在一个单一的、一致的抽象中,为实现“DC就是计算机”的愿景迈出了重要的一步。

项目成果

期刊论文数量(10)
专著数量(0)
科研奖励数量(0)
会议论文数量(0)
专利数量(0)
Optimizing Network Performance in Distributed Machine Learning
  • DOI:
  • 发表时间:
    2015-07
  • 期刊:
  • 影响因子:
    0
  • 作者:
    Luo Mai;C. Hong;Paolo Costa
  • 通讯作者:
    Luo Mai;C. Hong;Paolo Costa
SquirrelJoin: Network-Aware Distributed Join Processing with Lazy Partitioning
  • DOI:
    10.14778/3137628.3137636
  • 发表时间:
    2017-08
  • 期刊:
  • 影响因子:
    0
  • 作者:
    Lukas Rupprecht;W. Culhane;P. Pietzuch
  • 通讯作者:
    Lukas Rupprecht;W. Culhane;P. Pietzuch
CloudScope: Diagnosing and Managing Performance Interference in Multi-tenant Clouds
NetAgg: Using Middleboxes for Application-specific On-path Aggregation in Data Centres
Faces in the Clouds: Long-Duration, Multi-User, Cloud-Assisted Video Conferencing
  • DOI:
    10.1109/tcc.2017.2680440
  • 发表时间:
    2019-07
  • 期刊:
  • 影响因子:
    6.5
  • 作者:
    R. Clegg;R. Landa;D. Griffin;M. Rio;Peter Hughes;Ian Kegel;T. Stevens;P. Pietzuch;Doug Williams
  • 通讯作者:
    R. Clegg;R. Landa;D. Griffin;M. Rio;Peter Hughes;Ian Kegel;T. Stevens;P. Pietzuch;Doug Williams
{{ item.title }}
{{ item.translation_title }}
  • DOI:
    {{ item.doi }}
  • 发表时间:
    {{ item.publish_year }}
  • 期刊:
  • 影响因子:
    {{ item.factor }}
  • 作者:
    {{ item.authors }}
  • 通讯作者:
    {{ item.author }}

数据更新时间:{{ journalArticles.updateTime }}

{{ item.title }}
  • 作者:
    {{ item.author }}

数据更新时间:{{ monograph.updateTime }}

{{ item.title }}
  • 作者:
    {{ item.author }}

数据更新时间:{{ sciAawards.updateTime }}

{{ item.title }}
  • 作者:
    {{ item.author }}

数据更新时间:{{ conferencePapers.updateTime }}

{{ item.title }}
  • 作者:
    {{ item.author }}

数据更新时间:{{ patent.updateTime }}

Peter Pietzuch其他文献

CAP-VMs: Capability-Based Isolation and Sharing in Clouds
CAP-VM:云中基于能力的隔离和共享
CubicleOS: A Library OS with Software Componentisation for Practical Isolation Extended Abstract
CubicleOS:具有软件组件化功能的库操作系统,可实现实际隔离扩展摘要
  • DOI:
  • 发表时间:
    2020
  • 期刊:
  • 影响因子:
    0
  • 作者:
    V. Sartakov;Llu'is Vilanova;Peter Pietzuch
  • 通讯作者:
    Peter Pietzuch

Peter Pietzuch的其他文献

{{ item.title }}
{{ item.translation_title }}
  • DOI:
    {{ item.doi }}
  • 发表时间:
    {{ item.publish_year }}
  • 期刊:
  • 影响因子:
    {{ item.factor }}
  • 作者:
    {{ item.authors }}
  • 通讯作者:
    {{ item.author }}

{{ truncateString('Peter Pietzuch', 18)}}的其他基金

Cloud Open Source Research Mobility Network
云开源研究移动网络
  • 批准号:
    EP/Y030346/1
  • 财政年份:
    2023
  • 资助金额:
    $ 84.88万
  • 项目类别:
    Research Grant
CloudCAP: Capability-based Isolation for Cloud Native Applications
CloudCAP:云原生应用程序基于能力的隔离
  • 批准号:
    EP/V000365/1
  • 财政年份:
    2020
  • 资助金额:
    $ 84.88万
  • 项目类别:
    Research Grant
CloudSafetyNet: End-to-End Application Security in the Cloud
CloudSafetyNet:云中的端到端应用程序安全
  • 批准号:
    EP/K008129/1
  • 财政年份:
    2013
  • 资助金额:
    $ 84.88万
  • 项目类别:
    Research Grant
CloudFilter: Practical Confinement of Sensitive Data Across Clouds
CloudFilter:跨云敏感数据的实际限制
  • 批准号:
    EP/J020370/1
  • 财政年份:
    2012
  • 资助金额:
    $ 84.88万
  • 项目类别:
    Research Grant
Smart Flow - Extendable Event-Based Middleware
智能流 - 可扩展的基于事件的中间件
  • 批准号:
    EP/F042469/1
  • 财政年份:
    2008
  • 资助金额:
    $ 84.88万
  • 项目类别:
    Research Grant
DISSP: Dependable Internet-Scale Stream Processing
DISSP:可靠的互联网规模流处理
  • 批准号:
    EP/F035217/1
  • 财政年份:
    2008
  • 资助金额:
    $ 84.88万
  • 项目类别:
    Research Grant

相似国自然基金

多维在线跨语言Calling Network建模及其在可信国家电子税务软件中的实证应用
  • 批准号:
    91418205
  • 批准年份:
    2014
  • 资助金额:
    170.0 万元
  • 项目类别:
    重大研究计划
基于Wireless Mesh Network的分布式操作系统研究
  • 批准号:
    60673142
  • 批准年份:
    2006
  • 资助金额:
    27.0 万元
  • 项目类别:
    面上项目

相似海外基金

In-service Catastrophic Failure: or Controlling the Grain Boundary Network Evolution
运行中的灾难性故障:或控制晶界网络的演化
  • 批准号:
    2888456
  • 财政年份:
    2023
  • 资助金额:
    $ 84.88万
  • 项目类别:
    Studentship
A radiation oncology clinician scientist devoted to building an inclusive clinical research program in an integrated academic satellite network in service to the National Cancer Institute
放射肿瘤学临床科学家致力于在为国家癌症研究所服务的综合学术卫星网络中建立包容性临床研究项目
  • 批准号:
    10561290
  • 财政年份:
    2023
  • 资助金额:
    $ 84.88万
  • 项目类别:
Implementation of telemedicine and social network driven HIV service uptake for comprehensive HIV service integration in rural syringe service programs
实施远程医疗和社交网络驱动的艾滋病毒服务,将艾滋病毒服务全面纳入农村注射器服务计划
  • 批准号:
    10682889
  • 财政年份:
    2023
  • 资助金额:
    $ 84.88万
  • 项目类别:
Traffic Control Schemes in Decentralized In-Network Computing toward Democratization of Network Service Platform
面向网络服务平台民主化的分散式网内计算中的流量控制方案
  • 批准号:
    22K17888
  • 财政年份:
    2022
  • 资助金额:
    $ 84.88万
  • 项目类别:
    Grant-in-Aid for Early-Career Scientists
Next Generation Software-defined Intelligent Radio Access Network (SIRAN) - Leveraging Deep Learning for Autonomous and Intelligent Service Provisioning
下一代软件定义智能无线接入网络 (SIRAN) - 利用深度学习实现自主和智能服务提供
  • 批准号:
    RGPIN-2019-06348
  • 财政年份:
    2022
  • 资助金额:
    $ 84.88万
  • 项目类别:
    Discovery Grants Program - Individual
CISE-MSI: RCBP-RF: S&CC: Building a Smart Mobility Network for the San Antonio Transit to Improve Transit Service and Social Impact (SmartSAT)
CISE-MSI:RCBP-RF:S
  • 批准号:
    2131193
  • 财政年份:
    2021
  • 资助金额:
    $ 84.88万
  • 项目类别:
    Standard Grant
Construction of a distributed control all-optical network utilized flexibly according to service demands
构建根据业务需求灵活运用的分布式控制全光网络
  • 批准号:
    21H03433
  • 财政年份:
    2021
  • 资助金额:
    $ 84.88万
  • 项目类别:
    Grant-in-Aid for Scientific Research (B)
Next Generation Software-defined Intelligent Radio Access Network (SIRAN) - Leveraging Deep Learning for Autonomous and Intelligent Service Provisioning
下一代软件定义智能无线接入网络 (SIRAN) - 利用深度学习实现自主和智能服务提供
  • 批准号:
    RGPIN-2019-06348
  • 财政年份:
    2021
  • 资助金额:
    $ 84.88万
  • 项目类别:
    Discovery Grants Program - Individual
FMitF: Track I: ComScaN: Composition and Scaling of Network Service Functions
FMITF:第一轨:ComScaN:网络服务功能的组成和扩展
  • 批准号:
    2123987
  • 财政年份:
    2021
  • 资助金额:
    $ 84.88万
  • 项目类别:
    Standard Grant
Network Topology-Aware Service Chaining in NFV Network
NFV 网络中的网络拓扑感知服务链
  • 批准号:
    21K21288
  • 财政年份:
    2021
  • 资助金额:
    $ 84.88万
  • 项目类别:
    Grant-in-Aid for Research Activity Start-up
{{ showInfoDetail.title }}

作者:{{ showInfoDetail.author }}

知道了