IMPROVEMENTS IN OR RELATING TO NETWORKS

TECHNICAL FIELD

This invention concerns improvements in or relating to networks and in particular, but not exclusively, to optical networks using Ethernet, in particular OAM (Operation, Administration, Maintenance) enabled Ethernet and more particularly, in accordance with the ITU Y.1731 Ethoam standard.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 is a diagrammatic view of an embodiment of a network in accordance with the invention;

Figure 2 is a diagrammatic view of an embodiment of a maintenance end point in accordance with the invention;

Figure 3 is a timeline of two schedules in accordance with the invention;

Figure 4 is a flowchart showing a method in accordance with an embodiment of the invention; and

Figure 5 is a flowchart showing a method in accordance with an embodiment of another aspect of the invention.

DETAILED DESCRIPTION

Referring to Figures 1 and 2, a network according to an embodiment of the invention comprises a plurality of nodes 10 to 17 interconnected by communication links 20 to 28, such as fibre optic cables. The network uses

Ethernet, in particular OAM (Operation, Administration, Maintenance) enabled Ethernet in accordance with the ITU Y.1731 Ethoam standard.

The plurality of nodes 10 to 17 comprise pairs of nodes 10 and 17, 11 and 12, 13 and 14 and 15 and 16 configured as maintenance end points (MEPs).

Each MEP 10 to 17 of each pair transmits continuity check messages

(CCMs) to the other MEP 10 to 17 of the pair. The CCMs are dedicated messages used for checking that the network is working, for example, each

MEP 10 to 17 may be able to determine faults in the network as a result of the absence of an expected CCM or from data contained in the CCM.

The network may be divided into sections 1 to 4 of the network (as indicated by the dotted lines) and each section 1 to 4 may be controlled/maintained by different operators. Each section 1 to 4 of the network may comprise one or more pairs of MEPs that the operator can use to determine whether that section 1 to 4 of the network is operating as it should be. In some instances, one operator may have some control/maintenance responsibilities over a section of the network that encompass other sections of the network controlled/maintained by other operators (in this instance section 1 includes sections 2 to 4). MEPs that operate for sections of the network that encompass other sections having dedicated MEPs are said to operate on a higher level and the CCMs sent by MEPs of the higher level sections pass through MEPs of lower level sections. For example, CCMs sent by MEPs 10, 17 pass through MEPs 1 1 to 16. The operator of higher level section 1 considers MEPs 1 1 to 16 to be maintenance intermediate points (MIPs). MIPs 1 1 to 16 do not initiate CCMs that are sent to MEPs 10, 17 that operate on a higher level but can listen to the higher level CCMs that pass therethrough.

Referring to Figure 2, each MEP 10 to 17 comprises a processor 50 and memory 51 and can be programmed (configured) to operate as desired.

The network may also comprise a management system 30 for configuring and controlling the MEPs 10 to 17. In this embodiment, the management system 30 is shown as a single device and communicates with the MEPs along the communication links 40 and 20 to 28. It will be understood however that the management system may comprise more than one device and may communicate with the MEPs over a dedicated communication link.

In accordance with the invention, each MEP 10 to 17 is arranged to transmit a CCM in accordance with a schedule such that an impact of transmission of the CCMs on network resources is reduced. In particular, to reduce network traffic at any one instant as a result of the transmission of CCMs, each MEP 10 to 17 is configured to transmit a CCM at a different time to the transmission of a CCM by another of the MEPs 10 to 17. Ideally, the transmission of CCMs by the MEPs 10 to 17 should be scheduled such that, as far as possible, the times of transmission of CCMs by the MEPs 10 to 17 are evenly distributed.

Figure 3 illustrates two examples of schedules for the transmission of CCMs in accordance with the invention. The first example is for a network comprising two pairs of MEPs, 1 and 2. The transmission of a CCM by either one of the MEPs of pair 1 is indicated by MEP l and the transmission of a CCM by either one of the MEPs of pair 2 is indicated by MEP2. The top line indicates transmission of a CCM by an MEP of the pair in one direction in the network (from left to right or right to left in the network as shown in Figure 1) and the bottom line indicates transmission of a CCM by the other MEP of the pair in the other direction. As can be seen, the MEPs are configured to transmit the CCMs in a sequence that is repeated cyclically such no one transmission of a CCM coincides with another. In this way, traffic at any instant due to transmission of CCMs is reduced.

The second example of a schedule according to the invention indicates scheduling wherein it is not possible to have a unique transmission time for

each CCM. In this case, any two MEPs may transmit CCMs at the same time. In this way, the instantaneous traffic due to CCMs is kept to a minimum.

It will be understood that the above embodiment is for illustrative purposes only and in most practical situations, a network will comprise many more pairs of MEPs and the transmission time (time slice) for a CCM is likely to be much shorter. For example, in one embodiment, the cyclic period is of the order of milliseconds and is divided into a hundred or more time slices. In one embodiment, the cyclic period is 2ms and is divided into 500 time slices.

The schedule for the MEPs may be determined and/or managed by the management system 30. The management system 30 may be programmed to determine the schedule for the transmission of CCMs by the MEPs 10 to 17 and configure the MEPs 10 to 17 in accordance with this determination. This method carried out by the management system is illustrated in Figure 4.

Determining the schedule may comprise dividing a cyclic period, during which a sequence of CCMs are to be transmitted, into a set number of time slices, and allocating one of these time slices to the transmission of CCMs by a specified one of the MEPs 10 to 17 such that the transmission of CCMs is distributed, preferably as evenly as possible, across the cyclic period.

It will be understood that other methods of scheduling may be used in accordance with the invention such that transmission of CCMs is distributed across a cyclic period (rather than transmission of all CCMs in one cycle simultaneously).

In one embodiment, each CCM comprises sent count data on the number of data frames that have been sent from the MEP to the corresponding MEP of the pair, received count data on the number of data frames received by the MEP from the corresponding MEP of the pair and a copy of sent count data contained in the last CCM received by the MEP from the corresponding MEP of the pair.

The MEPs can use such information to calculate data frame loss. However, if a CCM message itself fails to be transmitted, this can result in the data frame loss calculated from the data contained in later received CCMs to be incorrect. In particular, the MEPs are programmed to include in a transmitted CCM a copy of sent count data contained in the last CCM received by the MEP from the corresponding MEP of the pair. In the prior art networks, an MEP has no way of knowing from the copy of sent count data contained in a received CCM whether one of its sent CCMs was lost. Therefore, the MEP has no way of knowing that the received count data does not tally with the copy of sent count data contained in a received CCM when one of its sent CCMs has been lost. Consequentially, data frame loss calculations based on such data in the received CCM may be presented as valid even though it is incorrect.

In accordance with one aspect of the invention, each MEP may be programmed to carry out the method shown in Figure 5. In this method, each MEP is arranged to store in memory the sent count data contained in CCMs transmitted from the MEP and compare the stored sent count data to the copy of sent count data contained in CCMs received from the corresponding MEP of the pair to determine if a CCM was lost. For example, if there is no corresponding value of sent count data in memory to the copy of sent count data contained in the received CCM then the MEP can assume that a CCM has been lost.

In order to provide efficient processing, each MEP may be arranged to search through the stored sent count data from oldest to newest to find the oldest stored sent count data with a value equal to a value of the copy of sent count data contained in a CCM received from the corresponding MEP of the pair and that is not older than any invalidated stored count data. If such a value if found then that data entry is invalidated (marked found). However, if no entry is found before an invalidated entry, then the MEP determines that a CCM has been lost and initiates appropriate action. For example, the MEP may be programmed to generate a signal indicating that any data frame loss calculation based on data contained in that CCM is invalid.