High-Performance Centralized Content Delivery Infrastructure: Models and Asymptotics
Authors:
- 25 August 2015
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ACM Transactions on Networking
- Vol. 23 (5) , 1674-1687
Abstract
We consider a centralized content delivery infrastructure where a large number of storage-intensive files are replicated across several collocated servers. To achieve scalable mean delays in file downloads under stochastic loads, we allow multiple servers to work together as a pooled resource to meet individual download requests. In such systems, basic questions include: How and where to replicate files? What is the impact of dynamic service allocation across request types, and whether such allocations can provide substantial gains over simpler load balancing policies? What are tradeoffs among performance, reliability and recovery costs, and energy? This paper provides a simple performance model for large systems towards addressing these basic questions.Keywords
This publication has 29 references indexed in Scilit:
- Queueing system topologies with limited flexibility2013
- Power-aware speed scaling in processor sharing systems: Optimality and robustnessPerformance Evaluation, 2012
- A unifying model and analysis of P2P VoD replication and scheduling2012
- Counting YouTube videos via random prefix sampling2011
- Stochastic networks with multipath flow control2011
- Optimality analysis of energy-performance trade-off for server farm managementPerformance Evaluation, 2010
- The resource pooling principleACM SIGCOMM Computer Communication Review, 2008
- Multi-Path TCP: A Joint Congestion Control and Routing Scheme to Exploit Path Diversity in the InternetIEEE/ACM Transactions on Networking, 2006
- Performance of peer-to-peer networks: Service capacity and role of resource sharing policiesPerformance Evaluation, 2006
- Topics in propagation of chaos1991