GB2433376A

GB2433376A - Communications system with peer-to-peer backup network

Info

Publication number: GB2433376A
Application number: GB0604079A
Authority: GB
Inventors: Andrew Mark Hutton
Original assignee: Siemens PLC
Current assignee: Siemens PLC
Priority date: 2005-12-15
Filing date: 2006-03-01
Publication date: 2007-06-20
Anticipated expiration: 2026-03-01
Also published as: GB0525485D0; GB2433376B; GB0604079D0

Abstract

A communications system 1 includes a number of nodes utilising a central server 2. To cater for potential failure of the server a peer to peer network 3,4 is provided as backup. The arrangement may be used to provide a back up communications network for a SIP-based communications system and the peer to peer network may make use of the Chord protocol.

Description

Communications system and method

Field of the invention

The present invention relates to a communications system and method, particularly a system based on the Session Initiation Protocol (SIP). The invention has particular application to any Internet Protocol (IP) based telephony system which employs a centralised server with which endpoints (i.e. phones, P.C. clients, gateways etc.) under normal operation are required to route session establishment requests to facilitate the location of the destination endpoint.

Background to the invention

The architecture of many deployments of SIP based User Agents (IP Phones, Soft Clients and Gateways) in Enterprise or Carrier based environments is such that the User Agents are totally reliant on centralised SIP Servers (B2BUA's or Proxy's) for locating other user agents. Before a call can be made, the SIP user agent has to register with the centralised SIP server and, to make a call, must send the SIP INVITE request to the SIP Server which then locates the destination user agent.

One problem with the centralised server approach is that when it is not available or cannot be accessed, for example due to WAN failure, the SIP user agents cannot communicate with each other even if there is potential for connectivity between user agents.

It is known to provide Peer to Peer (P2P) networks. These are networks in which every node has equal access to the network and can send and receive data at any time without having to wait for permission or assistance from a central node. One of the advantages of a P2P based telephony solution over the centralised server solution is that there is no single point of failure and the P2P system can maintain communication between endpoints even when parts of the network are not available. An example of a P2P protocol is Chord (See http://pdos.csail.mit.edu/chord/).

Summary of the invention

According to the invention in its broadest aspect there is provided an IP communications system comprising a set of nodes and a central server operating in accordance with a Internet Protocol (IP) and a peer to peer (P2P) network formed by the nodes.

Preferably, the protocol is a Session Initiation Protocol (SIP).

In event of failure of the central server, communication between the nodes will be possible using the P2P network.

IP Telephony endpoints will determine the availability of a centralised telephony server through a process such as registration and/or keep alive mechanism. In a SIP based system this typically requires the endpoint to register with the SIP server.

In addition to determining the availability of the centralised telephony server, the endpoint also joins a peer to peer based network. Many ways of providing a peer to peer network are applicable. The result will be that the endpoint will join a peer to peer network which enables the endpoint to discover the location of other peers without the use of a centralised server, typically this uses the concept of a distributed hash table to locate resources in other peers.

In normal operation, the endpoint will route all requests involved in the management of communication sessions (e.g. call setup, clear down etc.) to the centralised telephony server. However, when the endpoint determines that the central server is unavailable, possibly due to lack of response to a request, it will fall back to a P2P mode to allow some telephony type communication to continue without the central server.

Alternatively, a polling process may be initiated in which a periodic enquiry is made to the server to ensure the IP network is functioning.

When in P2P mode, endpoints will use P2P techniques (e.g. Distributed Hash tables) to discover the location of another endpoint with which they want to communicate. When they discover the address of the other endpoint they will establish communication by establishing a communication session directly with the other endpoint. Some endpoints may perform specific services like enabling access to the PSTN/ISDN and the knowledge of what endpoints can perform what services will be held with the P2P network.

It is possible that some endpoints may connect only to the P2P network and not to the centralised telephony server. One example of this, could be an access device which is used to route calls to another network (e.g. PSTN/ISDN) when the centralised service is not available.

It is possible that groups of endpoints connected to a single central server may form separate peer networks for survivability. For example a peer network could be restricted to a single location in a larger enterprise or could be restricted to a single enterprise if the centralised server services more than one enterprise.

When in the failback P2P mode, some centralised functions may be lost. For example, it may not be possible to log calls in a centralised call detail record or receive presence information from a centralised presence service. However basic telephony functions will be possible between devices in the P2P network and access to other networks (e.g. PSTN/ISDN) can be made available through a P2P enabled access device.

To enable the switching from the centralised communication to the P2P based mode of communication, the endpoint includes means to monitor the availability status of the centralised server. For example, the endpoint will periodically poll the server and expect an acknowledgment. If no acknowledgment is received, the endpoint may decide to switch into a P2P mode for communication. In addition the endpoint may determine that the server is unavailable at the time of initiating a communication session and may decide to switch to a P2P mode to create the communication session.

Typically, when the endpoint determines that the centralised server is unavailable, it will try to re-establish communication with the server by periodically trying to communicate with it. When it determines the server is available again it will revert to using it to establish new communication sessions.

The P2P network will normally be maintained even when the centralised server is available and may advantageously be used for other services, for example a distributed database.

The invention also provides a method.

Brief Description of the Drawings

A specific embodiment of the invention will now be described, by way of example only, with reference to the drawings in which: Figure 1 shows a communications network including two enterprise offices or locations hosted by a single centralised telephony server and each office having a peer to peer network; Figure 2 shows a communications network including two enterprise offices or locations hosted by a single centralised telephony server and each office also connected by a peer to peer network to the other; and Figure 3 shows an endpoint used in the communications network of figure 2.

Detailed Description

As is shown in figure 1, a first embodiment provides a communications network 1 including a centralised server 2 for connecting networks 3 and 4 at different locations.

A peer to peer network (P2P) is formed within each location which spans that location.

Thus, network 3 covers location A and network 4 covers location B. In a second embodiment shown in figure 2, a centralised server 2 is provided as before but in this case the P2P network 3 extends over both locations.

Figure 3 shows an endpoint 5 for use in either embodiment. In this case, it is a phone, although other endpoints may be used. It includes a communications application 6 which provides the core communications functionality to enable the endpoint to communicate via the centralised server supported network 2. It operates in accordance with the SIP protocol.

A user application 7 allows a user to initiate and terminate calls. In/out devices 8 are involved in this and may include a keyboard or mouse, microphones and speakers, and a display.

The communication application 6 connects the endpoint to the centralised server supported network 2, and also to the P2P network 3. Normally communication is, in this embodiment, made via the centralised server 2. In the event that the centralised server is not detected by the communications application 6 an output is made to a network monitor 9 indicating a network problem or failure.

The network monitor 9 provides an output to a selection manager 10. The selection manager 10 applies a set of rules to determine an appropriate course of action. For example, failure may require an immediate switch to the P2P network 3 or failure beyond a certain time threshold may trigger a switch.

The selection manager 10 instructs a network selector 11 to switch to the alternative network in this case the P2P network 3. The network selector 11 instructs the communications application 6 to utilise the P2P network 3.

When the centralised server 2 recovers, its presence is detected by the communications application 6. The network monitor 9, selection manager 10 and network selector 11 are used in much the same way as before to instruct the communications application 6 to switch to use of the centralised server.

Claims

Claims 1. A communications system comprising: an IP network operable in

a first network configuration formed by a set of nodes and a central server and operable in a second network configuration of a peer to peer network formed from at least some of the set of nodes and not the central server; means to detect a failure of the first network configuration; and means to activate the second network configuration upon detection of the failure.

2. A communications system as claimed in claim 1 wherein the IP network is a Session Initiation Protocol (SIP) network.

3. A communications system as claimed in claims 1 or 2 comprising; means to detect a re-enablement of the first network configuration; and means to re-activate the first network configuration upon detection of the re-enablement.

4. A communications system as claimed in claim 1, 2 or 3 wherein the means to detect a failure of the first network configuration detects a failure of the central server.

5. A communications system as claimed in claim 3 or 4 wherein the means to detect re-enablement of the first network configuration detects a re-enablement of the central server.

6. A system as claimed in any preceding claim wherein the first network configuration provides telephony and or multimedia communications between users.

7. A system as claimed in any preceding claim wherein the second network configuration utilises a Chord Protocol.

8. A communication method comprising: providing an IP network operable in: a first network configuration of a set of nodes and a central server; and a second network configuration of a peer to peer network formed by at least some of the set of nodes and not the central server, detecting a failure of the first network configuration; and in resonse to the detection of a failure utilising the second network configuration.

9. A method as claimed in claim 8 comprising; detecting a re-enablement of the first network configuration and in response, to the detection of the re-enablement switching to the first network configuration.