Network load-aware content distribution in overlay networks

Abstract

Abstract-Massive content distribution on overlay networks stresses both the server and the network resources because of the large volumes of data, relatively high bandwidth requirement, and many concurrent clients. While the server limitation can be circumvented by replicating the data at more nodes, the network limitation is far less easy to cope with, due to the difficulty in determining the cause and location of congestion and in provisioning extra resources. In this paper, we present novel schemes for massive content distribution, that assign the clients to appropriate servers, so that the network load is reduced and also well balanced, and the network resource consumption is low. Our schemes allow scaling to very large system because the algorithms are very efficient and do not require network measurement, or topology or routing information. The core problems are formulated as partitioning the clients into disjoint subsets according to the degree of interference criterion, which reflects network resource usage and the interference among the concurrent connections. We prove that these problems are NPcomplete, and present heuristic algorithms for them. Through simulation, we show that the algorithms are simple yet effective in achieving the design goals.

Network Load-aware User Grouping for Internet Media Streaming Systems

“…It gives the maximum affordable number of subscribers and service channels. The issue of reducing link stress has also been considered in several other works ( [2], [13]). We found that content distribution sessions following our user grouping algorithm make a balanced use of the network bandwidth, which tends to cause the least interference to other users (See Section IV).…”

“…The particular metrics we choose to capture the network load are known as the worst link stress (WLS) [2] and the degree of interference (DOI) [3].…”

“…Intuitively, it is easy to see that (c) is better than (b) from the point of view of network load balance. 2 For the measurement of the network load, we employ two metrics known as the worst link stress (WLS) [2] and the degree of interference (DOI) [3]. The WLS is the largest number of streams on any link and directly related to the worst congestion level in the network.…”

Design, optimization and performance evaluation of a content distribution overlay for streaming

“…As regards related works focused on architectures relying on overlays, reference [8] deals with selection strategies of clients by optimally partitioning them into disjoint subsets according to different criteria, such as network resource usage and the degree of congestion. Besides, optimal rate allocation in overlay-based CDN for efficient utilization of limited bandwidths has been investigated in [9].…”

Overview and Challenges of Overlay Networks: A Survey

“…For example, Chul Han et al [61] try to assign the clients to appropriate servers, so that the network load is reduced and also well balanced, and the network resource consumption is low. This scheme allows scaling to very large systems because the algorithms are efficient and do not require network measurements nor topology or routing information.…”