An ISP Controlled P2P Overlay Network for Content Sharing

2022 17th Iberian Conference on Information Systems and Technologies (CISTI)(2022)

引用 0|浏览2
暂无评分
摘要
Nowadays, there is a strong tendency for the development of new Internet applications, many of which based on the overlay networks paradigm and oriented towards the goal of content sharing. However, as a consequence of this effective growth of overlay networks, Internet Service Providers (ISPs) infrastructures are overloaded with many Peer-to-Peer (P2P) related traffic, inducing more difficulties in the perspective of several ISP management and traffic engineering tasks. This work promotes the development of an overlay network based system allowing generic content sharing functionalities among the users, but with the ability to be controlled by the ISP. In the proposed system, the ISP is able to take advantage of specific P2P traffic control mechanisms, with the devised overlay architecture also supporting collaborative mechanisms with the ISP. Within this purpose, two basic contents transfer modes are supported by the system (client-server or P2P based ones). The devised overlay system also integrates two distinct application level routing strategies and several Traffic Engineering mechanisms that can be activated by the ISP. In particular, it is possible for the ISP to protect specific links and routers of the infrastructure from the overlay generated traffic. In a distinct perspective, it is also possible to prioritise specific links or routers of the network in order to be traversed by larger volumes of the overlay traffic. Such decisions can be permanently assumed, or activated by the ISP only during specific time periods. Additionally, the overall amount of P2P related traffic circulating in the overlay network can be also easily controlled by the ISP. A real prototype of the proposed collaborative system was implemented and tested in the Common Open Research Emulator (CORE).
更多
查看译文
关键词
Content Sharing,Internet,ISP,Overlay Networks,P2P,Traffic Engineering
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要