WO2000036805A1

WO2000036805A1 - A communications network

Info

Publication number: WO2000036805A1
Application number: PCT/GB1999/004115
Authority: WO
Inventors: Andrew Stephen Chittenden
Original assignee: Madge Networks Limited
Priority date: 1998-12-11
Filing date: 1999-12-07
Publication date: 2000-06-22
Also published as: GB9827383D0

Abstract

The present invention relates to a communications network (1, 5, 6) which transfers data between communications devices (2, 3, 4, 7) in accordance with network addresses. The communications network has a group (3, 7) of at least two communications devices coupled to it with each device in the group having a common network address. The network is then adapted to transfer data destined for the common network address to a selected one of the communications devices in the group.

Description

A COMMUNICATIONS NETWORK

The present invention relates to a communications network which operates to transfer data between communications devices in accordance with network addresses. The present invention also relates to a group of communications devices for use in the communications network as well as a method of providing the two communications devices for use in a communications network. Typically, communications networks transfer data between respective communications devices based upon a network address. Thus, for example, in the case of the Internet, each end station connected to the Internet is assigned a respective Internet Protocol (IP) address. The unique network address of the intended destination end station is then included in any data packets, thereby allowing the network to transfer the data through the network, to the intended destination.

Important end stations, such as network servers or the like, usually include a back-up system which provides redundancy should a fault develop with a primary system. Thus, this may be provided in the form of a back-up end station which operates should the main end station fail, or alternatively it may be a back-up network interface card (NIC) within the end station which functions if the main NIC fails.

However, because networks assign a unique network address to each communications device, different network addresses are assigned to the main and back-up systems. Accordingly, end stations coupled to the network have to be configured to communicate with either network address as required. This not only requires that the other communications devices have to store a record of the network address of both the primary and the back-up communications devices, but also requires the network to be configured to update the end station should the primary system fail, such that subsequent communication can be carried out with the back-up device.

In accordance with a first aspect of the present invention we provide a communications network which operates to transfer data between communications devices in accordance with network addresses, the communications network having a group of at least two communications devices coupled thereto, the communications devices in the group having a common network address, wherein the network transfers data destined for the common network address to a selected one of the communications devices in the group.

Accordingly, we provide a communications network which includes two communications devices having the same network address. Data is transferred to one of the communications devices in accordance with this network address. If this communications device should fail, the data can be transferred to the second communications device without having to change the network address .

Typically the network further comprises an address designating device which assigns each communications device a network address based on a respective individual device address, wherein the communications devices in the group have a common device address, the common network address being assigned to each of the group of communications devices in accordance with the common device address. This allows a common network address to be easily assigned to the two communications devices. However, it will be appreciated that a common network address could be assigned to the communications devices based on different device addresses if suitable apparatus is provided. It will be appreciated that for certain network protocols an address database, which stores the individual device address of the various end stations may also be provided.

The address designating device is preferably a Dynamic Host Configuration Protocol (DHCP) device which is standard in many communications networks. However, any suitable method of assigning the network addresses could be used. Typically, each communications device in the group has a respective individual device address (such as a MAC address) , the data being transferred to the selected communications device in accordance with the respective individual device address and the common network address. Thus data is transferred to the group of devices in accordance with the network address and then to the selected device in accordance with a separate device address. However, any suitable alternative may be used, such as disabling a back-up communications devices whilst a primary communications device is working, then allowing the back-up to be enabled should a fault develop with the primary communications device.

Typically, the communications devices in the group cooperate so as to determine the selected one of the communications devices.

Furthermore, the selected communications device is preferably determined in accordance with information representative of the operational status of each communications device in the group. This allows a back-up communications device to be activated automatically should a fault occur with a normally selected device.

In accordance with a second aspect of the present invention, we provide a group of communications devices for use in a communications network, the communications network operating to transfer data to communications devices coupled thereto in accordance with corresponding network addresses, the group comprising: at least two communications devices having a common network address, the communications devices being arranged so that data destined for the common network address is transferred to a selected one of the communications devices .

Accordingly, the two communications devices of the group are provided with a common device address. The network address is then assigned in accordance with the common device address allowing the common network address to be assigned in accordance with normal operation protocols .

Typically the device address is a Media Access Control (MAC) address. Currently, every network connectable device is assigned a unique MAC address which is used by network protocols to assign a unique network address to the respective device. By also providing a MAC address common to the at least two communications devices, this allows a common network address to be assigned to the communications devices in accordance with normal operation protocols. Furthermore, data can then be transferred to the selected one of the communications devices in accordance with the individual MAC address .

Typically each individual MAC address is stored in a respective store in the respective communications device. This is usually burnt into a respective flash ROM during the production of each communications device. However, any suitable method of designating a communications device with a MAC address may be used. Preferably the common device address is stored in a common store accessible by each communications device. However, it will be realised that a store may be provided at an alternative location in the network, for example at a network controller, which may then be accessed by each communications device, as required.

Typically the communications devices comprise respective network interface cards of a single communications network end station. However, each communications device may comprise a separate end station or alternatively a separate communications network, depending on the circumstance in which the present invention is implemented.

In accordance with a third aspect of the present invention, we provide a method of operating a communications network interconnecting a number of communications devices and including a group of at least two communications devices, the communications network operating to transfer data between communications devices in accordance with network addresses assigned to the communications devices, the method comprising: assigning a common network address to the communications devices in the group; and, transferring data destined for the common network address to a selected one of the communications devices in the group .

Accordingly, we also provide a method of assigning a common network address to each communications device in the group so that the group of at least two communications devices can be used on a standard network, without requiring any network modification.

Typically the method of transferring data to the selected one of the communications devices comprises transferring initialisation data to each of the communications devices in the group in accordance with a network address, causing the selected communications device to generate response data including its respective individual device address and, transferring subsequent data to the selected communications device in accordance with the common network address and the respective individual device address .

This allows an end station coupled to the network to initialise communication with the group of communications devices using the common network address. The end station then receives an indication of the respective individual device address of the selected communications device so that any further data can be transferred directly thereto. However, should a fault occur, the respective individual device address of an alternative communications device can be transferred to the end station such that operation of the system can continue substantially uninterrupted.

An example of the present invention will now be described with reference to the accompanying drawings, in which: - Figure 1 is a schematic view of a communications network including groups of communications devices according to the present invention;

Figure 2 is an example of a data packet suitable for transfer over the network of Figure 1;

Figure 3 is a schematic diagram the modified end station of Figure 1;

Figure 4 is a schematic diagram of the group of end stations of Figure 1; and, Figure 5 is a schematic diagram of a second example of the modified end station of Figure 1.

Figure 1 shows two communications networks 1,5, coupled via a router 6, which operate in accordance with the Internet Protocol (IP) . Coupled to the networks 1,5 are a number of end stations 2a, 2b, 2c, 2d, a modified end station 3 and a dynamic host configuration protocol (DHCP) device 4 and an end station group 7, as shown.

Communication between two standard end stations 2 will now be described with reference to Figure 1. In use, data packets are transferred between the end station 2,3,7 using respective individual MAC addresses. The MAC addresses are defined within the computer industry such that every device which couples to a computer network has a unique MAC address. Accordingly, if a data packet to be transferred includes a MAC address, the data packet is simply transferred directly to the device having the given MAC address in accordance with predetermined routing information.

If however an end station 2a is to communicate with an end station 2b for the first time then the end station 2a will not have a record of the MAC addresses of the end station 2b. Accordingly, the end station 2a initiates data transfer to the end station 2b by generating an initialisation data packet including the IP address of the end station 2b.

A data packet 30, suitable for transfer over the communications network 1, is shown in Figure 2. This comprises a header 31, including a destination IP field 32, and a destination Media Access Control (MAC) field 33, which indicate the IP and Media Access Control (MAC) addresses of the destination end station 2b and a source IP field 34 and a source MAC field 35, which indicate the IP and MAC addresses of the source end station 2a. The data packet 30 also includes a payload 36 including the data to be transferred. The remainder of the header, indicated generally at 37 will include any other header fields required by the IP protocol.

The initialisation data packet is transferred to the network 1.

However, in the present case, the initialisation data packet does not include an address in the destination MAC field 33 because this is unknown to the source end station 2a. Accordingly, the network operates to transfer the data to each end station 2b, 3 coupled to the network 1. The data packet is also transferred to the router 6, which passes the data packet on to each of the other end stations 2c, 2d, 7 coupled to the network 5, as will be described in more detail below.

Each end station then compares the IP address contained in the destination IP field 32 and determines whether it is the intended destination end station. In this case, the end stations 2c, 2d, 3, 7 determine that they are not the intended recipient of the data packet and the data packet is ignored.

However, the end station 2b will determine that it is the intended destination end station and will generate a response data packet accordingly. In this case, the response data packet includes the IP address and the MAC address which were contained in the source IP and MAC fields 34,35 of the received initialisation data packet in the respective destination IP and MAC address fields 32,33. Furthermore, the end station 2b includes its own respective IP and MAC addresses in the source IP and MAC fields 34,55 of the response data packet . The response data packet is then transferred onto the network 1, which in turn transfers it to the end station 2a, in accordance with the MAC address contained in the destination MAC field 33. The end station 2a can then extract the MAC address of the end station 2b, from the source MAC field 35, allowing any subsequent data to be transferred between the end stations 2a, 2b to be transferred directly using MAC addresses.

In the case in which communication is to be established between the end stations 2a, 2d, the initialisation data packet is transferred to the end station 2d via the router 6. In this case, the router modifies the data packet by inserting its own MAC address in the source MAC field 35 before passing the data packet on to each of the other end stations 2c, 2d, 7 coupled to the network 5. The router keeps a record of the IP and MAC addresses of the end station 2a.

Upon receipt of the initialization data packet, the end station 2d will generate the required response data packet. This includes the MAC address of the router 6 in the destination MAC field 33, as this is the MAC address included in the source MAC field 35 of the initialisation data packet while the IP address of the end station 2a is included in the destination address field 32 of the response data packet.

Accordingly, the response data packet is transferred directly to the router 6. This then uses the IP address of the end station 2a in the field 32 to determine, from the stored record, the MAC address of the end station 2a. This is then placed in the destination MAC field 33. The router 6 also places its own MAC address in the source MAC field 35 and keeps a record of the IP and MAC addresses of the end station 2d.

The response data packet is then transferred to the end station 2a, which determines the MAC address of the router therefrom. This allows all subsequent data to be transferred to the end station 2d directly, via the router 6, using MAC addresses.

Once the network management mechanisms have used the above mentioned method to establish a suitable route for communication between any two end stations, details of the route are stored as routing information. This routing information is determined for given IP addresses and is stored at a number of points in the network (depending upon its type and features) , typically including within each end station, allowing data packets to be transferred as required.

However, if there is no communication between the two such end stations for a set interval of time, the routing information is erased from the store, in an event known in the art as a "time-out" . Any subsequent attempt by one of those end stations to communicate with the other will then require it to invoke again the initialisation procedure described above .

IP addresses are assigned to end stations by the DHCP device 4 in response to an IP address request from one of the end stations 2. The request is generated by an end station when it requires an IP address and this generally occurs when the end station 2 is initially connected to the network 1, although it can also happen under other circumstances known in the art.

The method so far described is already known. Operation of the network in conjunction with the modified end station 3, will now be described. As shown in Figure 3 , the end station 3 comprises two computer network interface cards (NICs) 10,11 having respective memories 13,14 and respective card processors 15,16. Each NIC is coupled to the network 1 via a bus 12, and is coupled to a host processor 18 of the modified end station 3 via a second bus 17. There is also provided a further memory 19 coupled to the bus 17.

The card processors 15,16 are designed to transfer NIC status information, via the bus 17, such that each NIC 10,11 periodically receives details of the current operating status of the other NIC 11,10. Furthermore, the NICs 10,11 are adapted to operate as a primary and a backup respectively. Accordingly, under normal circumstances, all communication with the modified end station 3 will be carried out via the primary NIC 10. However, should the NIC 10 develop a fault, this will be detected by the backup NIC 11 allowing subsequent communication to be carried out via the back-up NIC 11. Each NIC 10,11 is identified by an individual MAC address . The manner in which the MAC addresses are assigned will depend on the circumstances in which this occurs. Thus, for example, if the MAC address is defined on production of the NIC 10,11, the MAC address is usually burnt into a flash ROM, or the like. Alternatively, the MAC address may be assigned before coupling the NIC to the network, in which case it may be programmed into a standard ROM using software. A further alternative is for a controller to be provided on the network, with the controller storing MAC addresses for each device coupled to the network. It will be realised that any suitable method may be used.

An additional common MAC address is also provided in the memory 19 such that it can be accessed by the card processors 15,16, via the bus 17. Again, the manner in which the common MAC address is assigned will depend on the circumstances. Thus, the memory 19 may comprise a flash ROM into which the common MAC address is burnt, upon production of the end station 3. Alternatively, the common MAC address could be assigned using suitable software, or the like, prior to coupling the end station 3 to the network. The MAC address may also be assigned by a controller within the network. As will be appreciated by a person skilled in the art, a further alternative is for a copy of the common MAC address to be stored in the respective memories 13,14 of the NICs 10,11. Again, this is not important to the implementation of the present invention, as long as each NIC 10,11 is able to determine the common MAC address when it is required.

As mentioned above, when the modified end station 3 is connected to the LAN 1, an IP address is determined. This is achieved by having the primary NIC 10 (or the back-up NIC 11 should the primary not be operational) transfer the common MAC address, which is stored in the memory 19, to the DHCP device 4. The DHCP device 4 assigns an IP address, based on the common MAC address, which is transferred back to the NIC 10. This common IP address is then stored in the memories 13,14 for use by both NICs 10,11.

Again however, the common IP address may be stored in only one of the NIC memories 13,14 or alternatively in the host memory 19. In these cases, the respective NICs 10,11 will simply be adapted to access the common IP address from the location in which it is stored, as required. Generally, the common IP address is however at least stored in the memory 13 of the primary NIC 10. In this case, the back up NIC 11 is adapted to either obtain the common IP address from the memory 13 of the NIC 10, or alternatively directly from the DHCP device 4, as required.

Communication between the end station 2a and the modified end station 3 is achieved by having the end station 2a generate an initialisation data packet, in the normal way, containing the IP address of the end station 3 in the destination IP field 32. This is transmitted to each end station coupled to the network, as described above . In the case of the modified end station 3, the data packet will be received by both the NICs 10,11. As described above however, the NIC 10 is the primary NIC and accordingly, as long as the NIC 10 is fully operational, the NIC 11 will not take any action. Thus, the NIC 10 will detect that the IP address in the destination IP field 32 matches the common IP address and operate to process the data packet as required. This includes transferring any data contained within the data packet to the host processor 18, via the bus 17 as required.

The card processor 15 then generates the response data packet, as described above, including its own individual MAC address in the source MAC field 35 of the response data packet. This is then transferred to the end station 2a in the normal way.

Accordingly, any further communication between the end station 2a and the modified end station 3 is carried out by having the end station 2 generate data packets including both the common IP address and the individual MAC address of the NIC 10, in the destination IP and MAC fields 32,33 respectively. Thus the data packet is transferred directly to the NIC 10, across the network 1, as required. The manner in which a system detects and then operates, should a fault occur to the NIC 10, will vary depending on the current operating procedures and the circumstances in which the fault occurs. Four examples of operation of the present invention are described below, although it will be understood that other circumstances may arise, depending on the particular network in question, in which case the operation of the invention may vary from that described in these specific examples.

Method A

This occurs when the routing information for the common IP address has timed-out since the end station 2a last communicated with the modified end station 3. As a result, when the end station 2a attempts to communicate with the end-station 3, it will determine that time-out has occurred and that the initialisation procedure must be repeated to allow the routing information to be determined. Accordingly, the end station 2a will determine the common IP address from the network, in the normal manner, generate an initialization data packet, as described above, and transmit this to each end station coupled to the network. In this case, because the NIC 10 has failed, the NIC 11 will detect the initialisation data packet and generate the response data packet, as described above, which is transferred back to the end station 2a. As the response data packet will contain the individual MAC address of the NIC 11, subsequent communication between end stations 2a and the modified end station 3 will be via the NIC 11.

Method B

In this case the end station 2a attempts to communicate with the modified end station 3 prior to the time out of the routing information for the common IP address occurring, but following the failure of the NIC 10. As a result of this failure, any response expected by the end station 2a, from the NIC 10, will not be received.

The end station 2a continues to wait for this response until the routing information for the common IP address has timed-out, causing the routing information stored within the end station 2a to be erased.

Accordingly, because the end station 2a is still attempting to communicate with the end station 3, the end station 2a sends out a new initialisation packet as described above under Method A. A response data packet is then received from the NIC 11 establishing new routing information such that subsequent communication between the end stations 2a, 3 will be carried out via the second NIC 11.

Method C

In this case, when the NIC 11 determines that a fault has occurred to the NIC 10, the NIC 11 generates a reset data packet which is transferred to the end station 2a. This causes the end station 2a to reset its communication setup with the common IP address by erasing the individual MAC address of the NIC 10.

Accordingly, the end station 2a once again generates an initialisation data packet which is transferred onto the network. Again, the NIC 11 responds to the data packet such that any subsequent communication is between the end station 2a and the NIC 11.

Method D

In this example, the NIC 11 detects the failure of the NIC 10 and generates an override data packet which is transferred to the end station 2a. This causes the end station 2a to replace the respective MAC address of the NIC 10, with the respective MAC address of the NIC 11. Accordingly, all subsequent communication occurs between the end station 2a and the NIC 11.

It will therefore be appreciated that in each of the above methods, the second NIC 11 provides an automatic back up which will operate in the event of the failure of the NIC 10. A second example of a system which provides a back-up communications device is shown in Figure 4. This consists of an end station group 7 having two end stations 20A,20B. Each end station has a NIC 21A,21B including a processor 22A,22B and a memory 23A,23B. The NICs are coupled to respective host processors 24A,24B and respective memories 25A,25B, via respective buses 26A,26B. The end stations are also coupled together via a connection 27. Each end station NIC 21A,21B is coupled to the network 5 via a bus 28.

In use, the operation of the group of end stations 7 will be similar to that of the modified end station 3, with the operation of the end stations being controlled by the host processors 24A,24B. Thus, one of the end stations 20A will be designated as the primary end station, with the other end station 2OB being provided as a back-up. Data indicating the status of each end station, will be transferred between the respective host processors via the connection 27 such that each end station 20A,20B is aware of the current operating status of the other end station.

A common MAC address will also be provided in the memory 25A and this is used by the primary end station 20A to determine a common IP address . The common IP address is then also stored in the memory 25A. Alternatively, the common MAC and common IP addresses may be stored in the memory 25B, in both memories 25A,25B, in respective memories 23A,23B in the NICs 21A,21B or in a common memory on the network.

During communication with an end station 2a, an initialisation data packet will be received by the two NICs 21A,21B. As the end station 20A is the primary end station, the NIC 21A is the primary NIC and accordingly, as long as the end station 20A is fully operational, the NIC 21B will not take any action. Thus, the NIC 21A processes the data packet as required and generates a response data packet including its own respective MAC address, which is stored in the respective memory 23A, in the source MAC field 35. This is transferred to the end station 2a, as described above .

Accordingly, any further communication occurs directly between the end station 2a and the end station 20A. This will continue until a fault occurs to the end station 20A. When this happens, the system is adapted to operate such that any further communication will be via the NIC 2IB. Again, this can be achieved in accordance with any of the methods A to D described above .

Thus, for example, in the case of methods A,B or C, the end station 2a may not detect an expected response from the end station group 7, the common IP address may have timed-out, or a reset data packet is received from the backup NIC 2IB. Any of these events cause the end station 2A to generate an initialization data packet, such that the MAC address of the NIC 2IB is transferred to the end station 2a. Alternatively of course the NIC 21B may generate a override data packet which simply causes the end station 2a to swap the respective MAC addresses currently stored for the respective MAC address of the NIC 2IB.

In any event, further communication occurs with the end station 20B.

It will be realised that in this example, a common connection 27 need not be provided if the two end stations 20A,20B are configured to transfer status information via the bus 28. In the example of a second modified end station 3, shown in Figure 5, the modified end station includes two standard NICs 110,111 coupled to the network 1 via a bus 112. The NIC 110,111 are also coupled to a host processor 118 and a host memory 119 via a bus 117, as shown. In this case, the host processor 118 is designed to monitor the operations of the NICs 110,111 to determine when a fault occurs. This allows standard NICs to be used as the NICs 110,111, with no modification being required. The host processor 118 therefore assigns one of the NICs 110 as the primary NIC with the other NIC 111 being the back-up NIC.

As in the example shown in Figure 3 , each NIC will generally store its own respective MAC address with the common MAC address being stored in the host memory 119. Similarly, the host processor 118 will cause one of the NICs 110,111 to request a common IP address from the DHCP device 4. Again this would generally be stored in the host memory 119.

However, as in the previous examples, the common IP address and the common MAC address may be stored in any suitable memory, such as a memory provided on the network 1 or a memory included in one or both of the NICs 110,111. Operation of the modified end station 3 shown in Figure 5 is substantially identical to that of the modified end station 3 shown in Figure 3. Accordingly, when the end station 2a generates an initialization data packet, the host processor 118 will cause the primary NIC 110 to transfer its own respective MAC address to the end station 2a, such that all subsequent communication is via the primary NIC 110. Should a fault occur with the NIC 110, in accordance with methods A or B, the end station 2a will generate a further initialization data packet. Accordingly, the host processor 118 will detect that the NIC 110 is not functioning and accordingly cause the NIC 111 to transfer its own MAC address to the end station 2a.

In the case of method C or D, the end station 118 will determine that the NIC 110 has developed a fault and accordingly cause the NIC 111 to generate either a reset or a override data packet, which is transferred to the end station 2a. Again, as described with respect to Figure 3, this will cause all subsequent communication to occur via the NIC 111.

It will be appreciated that in the examples of Figures 3, 4 and 5 additional redundancy may be provided by the use of further NICs in the modified end station 3, or further end stations 20 in the group of end stations 7, again all of which use the common MAC and hence common IP address.

It will also be appreciated that the present invention can be implemented with a static DNS type method of IP address designation, in which the IP addresses are designated by an individual before the respective end stations 2,3 are coupled to the network.

Claims

1. A communications network which operates to transfer data between communications devices in accordance with network addresses, the communications network having a group of at least two communications devices coupled thereto, the communications devices in the group having a common network address, wherein the network transfers data destined for the common network address to a selected one of the communications devices in the group.

2. A communications network according to claim 1, the network further comprising an address designating device which assigns each communications device a network address based on a respective device address, wherein the communications devices in the group have a common device address, the common network address being assigned to each of the group of communications devices in accordance with the common device address .

3. A communications network according to claim 2, wherein the address designating device comprises a Dynamic Host

Configuration Protocol (DHCP) Device.

4. A communications network according to any of the preceding claims, wherein each communications device in the group has a respective individual device address, the data being transferred to the selected device in accordance with the respective individual device address and the common network address .

5. A communications network according to any of the preceding claims, wherein each communications device in the group cooperates so as to determine the selected one of the communications devices.

6. A communications network according to claim 5, wherein the selected communications device is determined in accordance with information representative of the operational status of each communications device in the group .

7. A communications network according to any of the preceding claims, wherein the network address comprises an Internet Protocol (IP) address.

8. A group of communications devices for use in a communications network, the communications network operating to transfer data to communications devices coupled thereto in accordance with corresponding network addresses, the group comprising: at least two communications devices having a common network address, the communications devices being arranged so that data destined for the common network address is transferred to a selected one of the communications devices .

9. A group according to claim 8, wherein each communications device of the group has a respective individual device address, wherein the data is transferred to the selected communications device in accordance with the respective individual device address.

10. A group according to claim 8 or claim 9, wherein each device address is a Media Access Control (MAC) address.

11. A group according to claim 10, wherein each individual device address is stored in a respective store in the respective communications device.

12. A group according to any of claims 8 to 11, wherein the common device address is stored in a common store accessible by each communications device.

13. A group according to any of claims 8 to 12, wherein each communications device in the group cooperates so as to determine the selected one of the communications devices.

14. A group according to claim 13, wherein the selected communications device is determined in accordance with information representative of the operational status of each communications device in the group.

15. A group according to any of claims 8 to 14, wherein the communications devices comprise respective network interface cards of a communications network end station.

16. A method of operating a communications network interconnecting a number of communications devices and including a group of at least two communications devices, the communications network operating to transfer data between communications devices in accordance with network addresses assigned to the communications devices, the method comprising: assigning a common network address to the communications devices in the group; and, transferring data destined for the common network address to a selected one of the communications devices in the group .

17. A method according to claim 16, wherein the communication devices of the group have a common device address, the common network address being assigned in accordance with the common device address .

18. A method according to claim 16 or claim 17, wherein each communication device of the group has a respective individual device address, the data being transferred to the selected one of the communications devices in accordance with the respective individual device address.

19. A method according to any of claims 16 to 18, wherein the method further comprises causing the communications devices to cooperate so as to select the selected one of the communications devices.

20. A method according to claim 19, wherein the method further comprises selecting the communication device in accordance with information concerning the operational status of each device.

21. A method according to at least claim 18, wherein the method of transferring data to the selected one of the communications devices comprises : transferring initialization data to each of the communications devices in the group in accordance with the common network address; causing the selected communications device to generate response data including its respective individual device address; and, transferring subsequent data to the selected communications device in accordance with the common network address and the respective individual device address.