Economic Incentives in Content-Centric Networking: Implications for Protocol Design and Public Policy
Economic Incentives in Content-Centric Networking: Implications for Protocol Design and Public Policy
复制标题
DOI：
--
复制
发表时间：
2011-09
期刊：
影响因子：
--
通讯作者：
M. Sirbu;Patrick Agyapong
中科院分区：
文献类型：
--
作者：
M. Sirbu;Patrick Agyapong
许多关于未来互联网体系结构的建议设想在网络中广泛分布的缓存，以促进命名数据的传递。这些建议通常被称为以内容为中心的网络(CCN)，旨在解决当前互联网的一些挑战，即缺乏数据持久性、安全性和高效的网络内内容交付支持。然而，以前在这一领域的工作在很大程度上忽略了诸如不同网络参与者部署CCN体系结构的激励以及部署以内容为中心的网络的政策含义等问题。在本文中，我们开发了一个简单的经济模型来评估不同参与者参与以内容为中心的网络的动机以及影响这些动机的因素。我们发现，网络会根据域间路由策略在网络中的不同位置放置缓存。此外，我们模型的结果表明，通常情况下，网络可能没有动力来部署社会上最优的网内缓存数量，除非出版商为此提供一些明确的金钱补偿。然而，对于不同的网络参与者，使网内缓存配置在经济上可行所需的补偿水平有所不同。具体地说，我们表明，当出版商愿意为缓存服务付费时，如果所有最终用户都集中在一个网络中，那么Eyball网络在为首次购买者定价缓存服务方面享有显著的竞争优势。然而，当终端用户分布在多个眼球网络中时，眼球网络在缓存服务定价方面的竞争力就会降低。因此，眼球网络很可能会投资于缓存，直到透明缓存具有经济意义。超过这一门槛，眼球网络将依赖CDN和交通网络来提供缓存服务，并通过适当调整他们向CDN和交通网络收取的费用来共同定位和终止流量，从而赚取尽可能多的租金。因此，我们发现，向出版商提供缓存服务的竞争将在公交网络和CDN之间展开。此外，我们的模型还列出了CDN可以在CCN生态系统中与提供缓存服务的传输网络竞争的条件。这些情况有助于我们解释最近的观察结果，例如Level 3 Communications能够以低于Akamai Technologies的价格提供Netfix的视频内容。最后，我们确定并讨论CCN协议设计考虑因素和政策干预，以确保通过部署CCNS获得理想的社会经济结果。
Many proposals for a future Internet architecture envision widely distributed caches within the network that facilitate delivery of named-data. Such proposals, generally referred to as content-centric networking (CCN), aim to address some challenges of the current Internet namely lack of data persistence, security and efficient in-network content delivery support. However, previous work in this area has largely ignored issues such as the incentives of different network players to deploy CCN architectures and the policy implications of deploying content-centric networks. In this paper, we develop a simple economic model to assess the incentives of various players to partake in a content-centric network and the factors that affect these incentives. We find that networks will place caches at different points in the network depending on the inter-domain routing policy. Furthermore, results from our model suggest that, in general, networks may not have incentives to deploy the socially optimal number of in-network caches unless there is some explicit monetary compensation from publishers for doing so. However, the level of compensation required to make in-network cache provisioning economically viable differs for different network players. Specifically, we show that when a publisher’s willingness to pay for caching services does not depend on the type of network that provides the service, then eyeball networks enjoy a significant competitive advantage in pricing caching services for first-time purchasers, provided all the end - users are concentrated in a single network. However, when end-users are distributed in multiple eyeball networks, then eyeball networks become less competitive in pricing caching services. Thus, it is likely that eyeball networks will invest in caching up to the point where transparent caching makes economic sense. Above this threshold, eyeball networks will rely on CDNs and transit networks to provide caching services and extract as much rent as possible by properly adjusting the fees they charge the CDNs and transit networks to co-locate and terminate traffic. Consequently, we find that competition to offer caching services to publishers will play out between transit networks and CDNs. In addition, our model lays out conditions under which CDNs can compete with transit networks that provide caching services in a CCN ecosystem. These conditions help us to explain recent observations such as Level 3 Communication’s ability to under-price Akamai Technologies to deliver Netfix’s video content. Finally, we identify and discuss CCN protocol design considerations and policy interventions that may be necessary to ensure a desirable socio-economic outcome from deploying CCNs.