Content delivery simulations supported by social network-awareness

详细信息    查看全文

• 作者：Irene Kilanioti ; ^{ekoila01@cs.ucy.ac.cy} ; George A. Papadopoulos ^{george@cs.ucy.ac.cy}
• 关键词：Social video sharing ; Social cascading ; Simulations ; Content distribution networks ; Internet measurements ; Social prediction ; Prefetching
• 刊名：Simulation Modelling Practice and Theory
• 出版年：2017
• 出版时间：February 2017
• 年：2017
• 卷：71
• 期：Complete
• 页码：114-133
• 全文大小：1197 K
• 卷排序：71

文摘

In this study we conduct experiments on a modified content delivery simulation framework, as we aspire to compare miscellaneous policies for dynamic OSN-aware content delivery. The incorporation of an OSN-aware dynamic mechanism becomes indispensable for CDN services, since (i)significantly large proportion of Internet traffic results from -easily produced via online media services and transmitted over OSNs- bandwidth-intensive multimedia content and (ii)multimedia content providers, such as YouTube, often rely on ubiquitous Content Distribution Networks (CDNs) infrastructures. Our policies take patterns of user activity over OSNs and exploit geo-social properties of users participating in retransmissions of items over OSNs (social cascades), proceed to incorporate various caching schemes of the underlying infrastructure, different policies for the handling of OSN data and various approaches that take into account the efficient timing of prefetching. The simulation framework we introduce can serve as the basis of further parameterized content delivery experimentation that exploits information transmission over OSNs and decreases replication costs by selectively copying items to locations where items are likely to be consumed.