WO2000011848A1

WO2000011848A1 - Method and apparatus for a multiprotocol switching hub

Info

Publication number: WO2000011848A1
Application number: PCT/GB1999/002692
Authority: WO
Inventors: Robert Hylton Madge; Nicholas Ian Moss; Richard John Knight; Trevor John Coleman
Original assignee: Madge Networks Limited
Priority date: 1998-08-18
Filing date: 1999-08-16
Publication date: 2000-03-02

Abstract

The invention relates to a communications network for transferring data between end stations (5, 6, 7) coupled to the network. The data is transferred in the form of data packets generated in accordance with either a first or a second transfer protocol. Each data packet also inlcudes a destination address indicating the end station to which the data is to be transferred. The network has a number of ports (2) for coupling to respective end stations, and a number of switching nodes (4) for interconnecting the number of ports (2). The switching nodes are adapted to determine the destination address of each data packet and transfer the data packet to the respective port (2).

Description

METHOD AND APPARATUS FOR A MULTIPROTOCOL SWITCHING HUB

The present invention relates to a communications network capable of transferring data in accordance with either a first or second transfer protocol and a method of coupling end stations thereto.

Currently, the majority of local area networks (LANs) operate using one of two types of protocol. These are "Carrier Sense Multiple Access Collision Detection" type networks (CSMA/CD) , such as Ethernet, or token passing networks, such as Token Ring.

In CSMA/CD networks, multiple end stations are connected to each other via a single bus. If an end station is to transmit to another end station, it monitors the bus to determine whether there is any traffic present, and if not sends the data as required. If two end stations begin to transmit simultaneously, there is a collision of the data on the bus which causes corruption of the signals. The end stations are configured to detect this and if it occurs, cease sending the data. The end stations then wait for a random time before monitoring the bus again and transmitting the data if possible.

In contrast to this, token passing networks operate by passing a token around each end station of the network in turn. End stations are then only able to transmit data if they are in possession of the token. After the transmission, or if it has no message to send, the end station passes the token on to the next end station.

More recently, networks have been introduced that operate on a point-to-point principle, rather than using shared media. However, these networks have maintained the use of the CSMA/CD and token passing type protocols in their existing forms.

Accordingly, whilst both of the above mentioned protocols work well, they are currently incompatible and both suffer from certain disadvantages. In accordance with a first aspect of the present invention, we provide a communications network for transferring data between end stations coupled to the network, the data being transferred in the form of data packets generated in accordance with either a first or a second transfer protocol, each data packet including a destination address indicating the end station to which the data is to be transferred, the network comprising: a number of ports for coupling to respective end stations; and, a number of switching nodes for interconnecting the number of ports, wherein the number of switching nodes are adapted to determine the destination address of each data packet and transfer the data packet to the respective port. Thus the present invention provides a communications network that is able to handle data packets generated in accordance with both CSMA/CD and token passing or other protocols such as FDDI . This is achieved by ensuring that the switching nodes are able to determine the routing information if present and the destination address of each type of data packet no matter which format the packet is in.

It will be understood by one skilled in the art that there is no limitation in principle to the number of different protocols which may be accommodated, so that, for example, three or more different protocols may be implemented within the same network. For the sake of simplicity, the description which follows is of a network implementing only two different protocols. Furthermore, whilst any protocols may be used, the present invention will be described with respect to CSMA/CD and token passing protocols and more specifically Ethernet and Token Ring.

Typically the apparatus further comprises a controller, wherein when a new end station is connected to the network, the controller is adapted to generate a first data packet in accordance with the first transfer protocol, transfer the first data packet to the new end station, monitor for a response from the end station, and determine whether the end station is configured to operate in accordance with either the first or the second transfer protocol depending on the result of the monitoring step. This allows the network to automatically determine the protocol which an end station will use during operation, thereby preventing data of the wrong format being sent to the end station.

In accordance with a second aspect of the present invention, we provide a method of coupling an end station to a communications network, the communications network being capable of transferring data in accordance with either a first or second transfer protocol, the method comprising the steps of : connecting the end station to the network; causing the network to generate a first data packet in accordance with the first transfer protocol; transmitting the first data packet to the end station; monitoring for response from the end station; and, determining whether the end station is configured to operate in accordance with either first or second transfer protocol depending on the results of monitoring step.

This allows end stations to be automatically connected to a communications network which transfers data generated in accordance with either the CSMA/CD or a token passing protocol . Accordingly, this allows a network to connect end stations which are designed to operate in accordance with for example, Ethernet, as well as end stations which are designed to operate in accordance with, for example, token passing protocols.

As mentioned above, there is no limitation in principle to the number of different protocols which may be accommodated, so that, for example, three or more different protocols may be implemented within the same network. This is achieved by determining the nature of the station when it is initially coupled to the network. The network is then able to ensure that only data packets having the correct format are passed to the end station.

Typically, the network determines that the end station is configured to operate in accordance with the first transfer protocol, if no response is received from the end station within a predetermined time period.

An alternative to this is for the network to determine that the end station is configured to operate in accordance with the first transfer protocol if an error signal is received from the end station within a predetermined time period.

In the first case, if the end station is configured to operate in accordance with the second transfer protocol, the method typically further comprises causing the end station to generate a data packet having the second transfer protocol, and transferring the data packet to the network. This allows the network to quickly and easily determine that the end station operates in accordance with the second transfer protocol . In this instance, the first data packet would typically be a modified Ethernet frame. This means that the network is configured to assume that each end station coupled to the network will operate preferably in accordance with the Ethernet protocol. However, it would be realised by a person skilled in the art that the default setting could be a token passing protocol or another protocol such as FDDI, depending on the default setting of the LAN 1.

Typically the modified Ethernet frame has a modified preamble. This allows the end station to detect the frame as a Ethernet frame, if it is an Ethernet end station.

An alternative however is to provide additional information inserted into the Ethernet transmitted frame. This would typically be added immediately after the preamble. As a result of this the Ethernet frame is effectively in an illegal format. Accordingly, if the end station is designed to operate in accordance with the Ethernet protocol, it will therefore fail to respond to the frame. From this, the network determines that the end station is to operate in accordance with the Ethernet protocol . In accordance with a third aspect of the invention, we provide a method for transferring data from a first computing device to a second computing device, the first computing device being capable of operating in accordance with either a first or second transfer protocol, the method comprising the steps of: causing the first computing device to generate a first data packet in accordance with the first transfer protocol; transmitting the first data packet to the second computing device; monitoring for a response from the second computing device; and, generating a number of subsequent data packets in accordance with either the first or second transfer protocol depending upon the result of the monitoring step. This aspect of the invention allows a computing device, such as a personal computer or the like, which is capable of operating either the Ethernet or token passing protocol to determine which protocol is operated by a second computing device to which it is attached. Accordingly, this allows an automatic reconfiguration of the first computing device so that it can operate with a computing device operating in accordance with either the first or second protocol .

Examples of the present invention will now be described with reference to the accompanying drawings, in which: -

Figure 1 shows in schematic form a LAN operating according to the present invention;

Figure 2 shows a representation of a modified Ethernet frame generated by the controller of the apparatus of Figure 1; and, Figure 3 shows in schematic form a connection formed between two end stations in accordance with the present invention.

Figure 1 shows a schematic view of a LAN 1 which is set up according to the method of the present invention. The LAN 1 includes a number of switching nodes 4 having a number of ports 2. Some of the ports 2 are used to couple the switching nodes 4 together via connections 3, whilst the remainder are used for coupling end stations to the LAN ₁. Each port 2 will include a port controller (not shown) which is used to configure the port, as will be described in more detail below.

Coupled to the ports 2 of the switching nodes 4 are a number of Ethernet end stations 5, which operate in accordance with the Ethernet protocol, a number of Token Ring end stations 6, which operate in accordance with a token passing protocol and a number of "Multigig" end stations 7, which are capable of operating in accordance with either the Ethernet or Token Ring protocols . Each end station 5,6,7 includes a respective end station port 5a, 6a, 7a having a respective end station controller (not shown) which is used to configure the respective end station port 5a, 6a, 7a for communication with the ports 2 of the switching nodes 4. In each case, the end station may take the form of a personal computer, a file server, a communications network or the like.

The LAN generally also includes a router 8 which determines which end station is permitted to communicate with which other end station. This is coupled to one of the switching nodes by a port 11 which operates in a similar manner to the ports 2 of the switching nodes .

When an end station 5,6,7 is initially coupled to one of the ports 2 of the switching nodes 4, the respective port controller determines the protocol with which the end station 5,6,7 can communicate. This allows the port 2 of the switching node 4 and the end station port 5a, 6a, 7a to be correctly configured. In order to do this, the port controller generates a data packet, which is transferred to the end station 5,6,7.

In the present example, the default operation of the LAN 1 is the Ethernet protocol and accordingly, this packet will take the form of a modified Ethernet packet. It will however be realised that if the default communication protocol is different, an alternative packet could be used.

Figure 2 shows an example of a suitable Ethernet packet 20 which includes a header 21, including a destination address 21a, a payload 22 and a preamble 23.

The Ethernet packet 20 is transferred via the respective port 2 to the respective end station port 5a, 6a, 7a of the newly added end station 5,6,7. From this packet, the end station controller can determine which protocols may be operated by the network, allowing the end station port 5a, 6a, 7a to be suitably configured, if necessary. The end station controller then generates a response (which may include a failure to respond) depending on whether it is an Ethernet, Token Ring or Multigig end station.

The response generated will depend on how the end station is configured and the modification made to the Ethernet packet. It will be realised by a person skilled in the art that many modifications and suitable responses are possible, however, two possible examples are explained in more detail below.

From this response, the respective port controller is able to determine the format of data packets which the end station is able to handle allowing the port 2 to be correctly configured.

An indication of the format of data which the end station 5,6,7 is able to handle is also passed on to and stored in the router 8 allowing communications between end stations to be coordinated. Thus, in the case of an Ethernet end station 5, the router 8 is able to determine that the end station must receive data packets generated in accordance with the Ethernet protocol. Similarly, if the end station is a Token Ring end station 6, the controller determines that the data packets must be generated in accordance with the token ring protocol . In the case of a Multigig end station 7, communication can be achieved using either protocol, however where a choice exists, communication generally occurs in accordance with the token ring protocol as this is usually more efficient.

The router 8 can then ensure that only data packets generated in accordance with the desired protocol are to be sent to the respective end station. This may be achieved by sending an indication of the destination address of the new end station, along with an indication of the protocol to be used to each end station. Alternatively, the respective port 2 of the switching node 4 can be instructed to discard any data packets having the wrong format before they are transferred to the port 5a, 6a, 7a of the respective end station 5,6,7.

The modification of the Ethernet packet will now be described.

In a first example of the present invention, the preamble 23 of the Ethernet packet 20 is modified. The preamble was generally used to aid the synchronization of shared media type networks and, in general, is no longer essential to the operation of point-to-point type networks. Accordingly, the preamble may be altered without adversely affecting the LAN 1.

Furthermore, any Ethernet end stations 5, which handle Ethernet format data packets are unable to detect the modification to the preamble. This effectively means that the first end stations handle the data packet as a standard Ethernet packet, making the first end stations effectively blind to this modification of the protocol.

In this case, if the new end station is an Ethernet end station 5, it will simply receive the packet, ignore the modified preamble and accept the Ethernet packet. No response will be generated. It will be realised that in this example in which the default operation of the LAN 1 is the Ethernet protocol, no end station controller will be required in the end station ports 5a. This is as a result of the port being permanently correctly configured for Ethernet operation.

If the new end station is a Token Ring end station 6, it will be unable to receive the packet and accordingly an error signal will be generated as the response. If the new end station is a Multigig end station 7, it will detect the modified preamble and generate a response packet in accordance with the token ring protocol .

An alternative possibility is to modify the transmitted frame. When an Ethernet frame is transmitted, for every 4 bits of the frame, 5 bits worth of bandwidth are used in the transmission of the frame. In this case, it is possible to add in extra information in the vacant bandwidth. This information would be most sensibly located between the preamble 23 and the destination address 21a in the data packet 20 and allows a message to be sent to the end station. Again the end station can be easily configured to respond depending on which type of end station it is.

It is also possible that the present invention may be used to connect two end stations together using a passive communications medium and an example of this will now be described with reference to Figure 3.

Figure 3 shows in schematic form a Multigig end station 50 capable of operating in accordance with either the Ethernet or a token ring protocol . The Multigig end station 50 is coupled via a transmission medium 51, to a new end station 52 which may operate in accordance with either a CSMA/CD or a token ring protocol, but whose status has not yet been determined. In this case, the operation of the system will proceed in one of two ways. In the first example, whenever the Multigig end station 50 is to transmit data to the new end station 52, the Multigig end station 50 must generate a modified Ethernet packet in the same manner as the port controller of Figure 1. The Multigig end station 50 then monitors for a response generated by the new end station 52 and, in a similar manner to the port controller of Figure 1, determines whether the new end station 52 would be capable of receiving data packets generated in accordance with the Ethernet or a token ring protocol .

Alternatively, the new end station 52 could be configured to generate a data packet which would be automatically sent to the end station 50 when the new end station is first added to the communications medium 51. In this case, the nature of the data packets sent would depend on the capabilities of the new end station. Thus an Ethernet end station would send a normal Ethernet packet, a Token Ring end station would send a normal Token Ring protocol type packet and a Multigig end station capable of handling either type of protocol would send a modified Ethernet packet and/or token ring protocol type packet .

It will be realised to a person skilled in the art that the present invention may be applied to any CSMA/CD or any token passing type protocols including, but not limited to, ATM and FDDI .

Claims

1. A communications network for transferring data between end stations coupled to the network, the data being transferred in the form of data packets generated in accordance with either a first or a second transfer protocol, each data packet including a destination address indicating the end station to which the data is to be transferred, the network comprising: a number of ports for coupling to respective end stations; and, a number of switching nodes for interconnecting the number of ports, wherein the number of switching nodes are adapted to determine the destination address of each data packet and transfer the data packet to the respective port.

2. Apparatus according to claim 1, wherein the apparatus further comprises a controller, wherein when a new end station is connected to the network, the controller is adapted to generate a first data packet in accordance with the first transfer protocol, transfer the first data packet to the new end station, monitor for a response from the end station, and determine whether the end station is configured to operate in accordance with either the first or the second transfer protocol depending on the result of the monitoring step.

3. A method of coupling an end station to a communications network, the communications network being capable of transferring data in accordance with either a first or a second transfer protocol, the method comprising the steps of: connecting the end station to the network; causing the network to generate a first data packet in accordance with the first transfer protocol; transmitting the first data packet to the end station; monitoring for a response from the end station; and, determining whether the end station is configured to operate in accordance with either the first or the second transfer protocol depending on the result of the monitoring step.

4. A method according to claim 3, wherein the network determines that the end station is configured to operate in accordance with the first transfer protocol if no response is received from the end station within a predetermined time period.

5. A method according to claim 4, wherein if the end station is configured to operate in accordance with the second transfer protocol, the method further comprises causing the end station to generate a data packet having the second transfer protocol, and transferring the data packet to the network.

6. A method according to any of claims 3 to 5, wherein the first data packet is a modified Ethernet frame.

7. A method according to claim 6, wherein the modified Ethernet frame has a modified preamble.

8. A method according to claim 3 , wherein the network determines that the end station is configured to operate in accordance with the first transfer protocol if an error signal is received from the end station within a predetermined time period.

9. A method according to claim 8, wherein the first data packet is a modified Ethernet frame.

10. A method according to claim 9, wherein the modified Ethernet frame includes an additional bit of information.

11. A method of transferring data from a first computing device to a second computing device, the first computing device being capable of operating in accordance with either a first or a second transfer protocol, the method comprising the steps of: causing the first computing device to generate a first data packet in accordance with the first transfer protocol; transmitting the first data packet to the second computing device; monitoring for a response from the second computing device; and, generating a number of subsequent data packets in accordance with either the first or the second transfer protocol depending on the result of the monitoring step.

12. A method according to claim 11, wherein the number of subsequent data packets are generated in accordance with the first transfer protocol if no response is received from the second computing device within a predetermined time period.

13. A method according to claim 11 or claim 12, wherein if the second computing device is configured to operate in accordance with the second transfer protocol, the method further comprises causing the second computing device to generate a data packet having the second transfer protocol, and transferring the data packet to the first computing device.

14. A method according to any of claims 11 to 13, wherein the first data packet is a modified Ethernet frame.

15. A method according to claim 14, wherein the modified Ethernet frame has a modified preamble.

16. A method according to claim 15, wherein the first computing device determines that the end station is configured to operate in accordance with the second transfer protocol if an error signal is received from the second computing device within a predetermined time period.

17. A method according to claim 16, wherein the first data packet comprises a modified Ethernet frame including an additional bit of information.