The Relaxed-Ring: A Fault-Tolerant Topology for Structured Overlay Networks

Abstract

Fault-tolerance and lookup consistency are considered crucial properties for building applications on top of structured overlay networks. Many of these networks use the ring topology for the organization or their peers. The network must handle multiple joins, leaves and failures of peers while keeping the connection between every pair of successor-predecessor correct. This property makes the maintenance of the ring very costly and temporarily impossible to achieve, requiring periodic stabilization for fixing the ring. We introduce the relaxed-ring topology that does not rely on a perfect successorpredecessor relationship and it does not need any periodic maintenance. Leaves and failures are considered as the same type of event providing a fault-tolerant and selforganizing maintenance of the ring. Relaxed-ring's limitations with respect to failure handling are formally identified, providing strong guarantees to develop applications on top of the architecture. Besides permanent failures, the paper analyses temporary failures and false suspicions caused by broken links, which are often ignored.

OSIRIS-SR: A Safety Ring for self-healing distributed composite service execution

“…However, those problems are outside of the scope of this paper, and can be overcome by applying solutions as presented in [8].…”

Modeling the Performance of Ring Based DHTs in the Presence of Network Address Translators

“…From a local perspective, a peer knows that it is the root of a branch if the size of its predecessor list is greater that one. Such predecessor list (PL) routing is used for instance in the relaxed-ring system [11] to handle non-transitive underlay networks.…”

From Mini-clouds to Cloud Computing

“…We have extensively studied structured overlay networks in the Selfman project [16], where peer-to-peer networks can scale well and quickly. Some of the networks developed in Selfman, Beernet [13,8] and Scalaris [15,12], provide not only self-organization of peers to deal with churn, but they also provide self-managing replicated storage. These networks are prepared to deal with unanticipated churn, because it cannot be known in advance when peers are going fail, join or leave the network.…”