Fault-tolerant mechanism optimised for peer to peer network

Abstract

#CMT# # / CMT# The device has an admission unit i.e. admission component, for inserting a new node within a peer-to-peer network (Np2p) of a communication element (E1) e.g. application server in a communication architecture, where the network is composed of nodes (N1-N5). The admission unit determines the adjacent nodes deployed on a same processing equipment, and inserts the new node between the nodes. The nodes (N1-N5) are arranged in circular form such that each node possesses a single successor node and a single predecessor node. #CMT# : # / CMT# An independent claim is also included for a communication element comprising a peer-to-peer network. #CMT#USE : # / CMT# Device for managing a peer-to-peer network of a communication element (claimed) e.g. application server in a communication architecture i.e. Internet protocol multimedia subsystem architecture, or signaling element e.g. proxy conforming to a session initiation protocol defined by Request for Comments (RFC) 3261 of Internet Engineering Task Force (IETF), or call session control function within the architecture. #CMT#ADVANTAGE : # / CMT# The device is configured such that two of the nodes are not present in the same processing equipment, thus limiting the simultaneous failure of the nodes, while eliminating the need for changing fault-tolerant mechanisms. #CMT#DESCRIPTION OF DRAWINGS : # / CMT# The drawing shows a schematic view of a communication element. E1, E2 : Communication equipments LB : Load distribution module Np2p : Peer-to-peer network N-TEL : Communication network N1-N5 : Nodes.

Description

technical field [0001] The invention relates to the deployment of a peer-to-peer network on a set of processing devices. More specifically, the invention relates to the use of these peer-to-peer networks for communication applications. Background technique [0002] In particular it is known to distribute certain telecommunication applications over a group of processing nodes. In this way, each node handles only the portion of requests addressed to the application, and the number of nodes can be adjusted dynamically (if necessary) based on the resources required to handle those requests. This architecture also means that applications can easily be fault-tolerant through the inherent redundancy between nodes. [0003] However, for some applications, there is still the location of contextual information within the set of nodes. Certain applications, such as a signaling unit may require context to be stored between two messages it sends. This context can be used to process s...

AI Technical Summary

Problems solved by technology

[0016] However, in practice, the technical problem of fault tolerance has always existed

[0019] Therefore, existing techniques are not suitable for providing sufficient fault tolerance

Images