Automatic network configuration with dynamic churn prediction

Abstract

Peer-to-Peer (P2P) systems have been deployed on millions of nodes worldwide in environments that range from static to very dynamic and therefore exhibit different churn levels. Typically, P2P systems introduce redundancy to cope with loss of nodes. In distributed hash tables, redundancy often fixed during development or at initial deployment of the system. This can limit the applicability of the system to stable environments or make them inefficient in such environments. Automatic network configuration can make a system more adaptable to changing environments and reduce manual configuration tasks. Therefore, this paper proposes an automatic replication configuration based on churn prediction that automatically adapts its replication configuration to its environment. The mechanism termed dynamic replication mechanism (dynamic RM) developed and evaluated in this paper is based on exponential moving averages to predict churn that is used itself to determine a replication factor meeting a certain reliability threshold. Simulations with synthetic data and experiments with data from torrent trackers show that the behavior can be predicted accurately in any environment, from low churn rates to diurnal and high churn rates.

Augmenting home routers for socially-aware traffic management

“…B-Tracker uses a Distributed Hash Table (DHT) plus direct messages for tracking, which balances better than pure DHT tracking approaches. The RB-Tracker [19] approach is an extension to B-Tracker involving replication of content tracked between nodes and locality. These extensions are not possible in a pure DHT tracking approach, since direct messages are required.…”

“…However, node identifiers (ID) in a DHT are distributed randomly and subsequently the load on nodes makes it not feasible to add features like locality and prediction. For this reason, RB-HORST builds an overlay following the RB-Tracker approach [19], which uses a DHT for the first lookup and sends direct messages to nodes sharing a video to find closer ones. The main benefit of RB-Tracker compared to a pure DHT is the load balancing, because the more nodes share a specific video the more nodes will be responsible for managing the overlay.…”

“…In this context, this paper proposes the RB-HORST mechanism; a sophisticated combination of the "Replicating Balanced Tracker" (RB-Tracker) [19] and the "Home Router Sharing based on Trust" (HORST) [26] mechanisms. RB-HORST improves the QoE of end-users by exploiting available social networking information so as to prefetch interesting content likely to be consumed.…”