Multiplex communications – Fault recovery – Bypass an inoperative station
Reexamination Certificate
2000-04-26
2004-09-21
Kizou, Hassan (Department: 2662)
Multiplex communications
Fault recovery
Bypass an inoperative station
C370S221000
Reexamination Certificate
active
06795394
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to data networks, to nodes for such networks, to methods of transmitting data, and to corresponding software. More particularly, it relates to data transmission networks having working paths and secondary paths for use by working traffic when there is a fault on the working path.
BACKGROUND TO THE INVENTION
A number of mechanisms are known for handling fault conditions in data transmission networks or systems. At higher levels of the well known OSI hierarchy, packets may be buffered and resent. At lower levels of the OSI hierarchy, if there is control over the route taken by the data, then an alternative route through the network can be tried. An example of an OSI layer
1
protocol, is SONET, and its equivalent outside North America, SDH. Such layer
1
transmission systems typically have secondary (usually termed protection) paths arranged such that working traffic is switched on to these secondary paths in the event of failure at a node or on a line between nodes of the system.
Typically, the nodes and lines are arranged to form rings. The protection paths are typically arranged either in the form of dedicated path protection rings (DPRing) or shared protection switched rings (SPRing). There are two kinds of SPRing, unidirectional line switched rings, or bi-directional line switched rings (BLSR). Bell Core Standard GR-1230-CORE Issue 2, defines BLSR for SONET. The standard includes protocols for extra traffic to use the protection paths when they are not in use by the working traffic. Paths are defined here as extending between neighbouring nodes, so typically data will pass over several such paths to reach its destination. Many paths can be multiplexed onto a single span, e.g. by time division multiplexing according to SONET/SDH standards, or by wavelength division multiplexing for example.
Where there is a ring configuration, and a fault is detected on a line on the working path, there are two possibilities in principle, for re-routing the working traffic to avoid the faulty line. Firstly, it could be switched on to the protection path on the same span as the faulty working path. This is known as a span switch. Secondly, the working traffic could be routed on the protection path the other way around the ring, using spans other than the faulty span. This is called a ring switch. In general, span switching is preferred since it occupies fewer of the protection paths available on the ring. This means several span switches may occur on the same ring, but only one ring switch.
There has long been a concern about the bandwidth efficiency of such systems, if 50% of the total bandwidth is reserved for the protection paths. Many efforts have been made to improve the efficiency. For example, extra traffic is carried on the protection channels. This extra traffic differs from working traffic in the treatment accorded to it at higher levels of the well known OSI hierarchy, such as priority level, charging rate, monitoring and so on. At the transport layer, it is treated the same as other traffic except for being given a lower priority under fault conditions, so it is unprotected and is automatically removed if a protection request is made, to switch the working traffic on to the protection path. In practice, it proves hard to derive much revenue from unprotected traffic.
Other attempts to improve bandwidth efficiency have involved efforts to share the protection path between several working paths. For example, where neighbouring rings overlap, a single protection path can serve both rings.
If the interconnections between nodes are arranged in a mesh rather than in one or more rings, then there may be many routes between any pair of nodes. In this case, in the event of a fault on one span, the traffic can be divided and sent over the protection paths for the various alternative routes to the destination node. Thus each protection path can have much less than 50% of the bandwidth of a given span between nodes, and the overall bandwidth efficiency can be much greater than 50% and still guarantee protection against any single fault. Again extra traffic can be sent over the protection paths. Compared to ring arrangements, the main drawback of mesh networks is the complexity of managing the network and deciding which protection paths to use in the event of a fault. Even if this is pre-calculated, changing traffic patterns may warrant complex re-configurations.
Yet other attempts at improving efficiency have involved removing completely unused working paths.
In summary, network operators perceive rings to be inefficient, and meshes to be efficient, yet difficult to manage and operate. Extra traffic to make use of the idle protection bandwidth is seldom used, because span switches occur often enough to cause too many outages on the extra traffic.
SUMMARY OF THE INVENTION
It is an object of the invention to provide improved arrangements which address the above mentioned problems.
According to the invention there is provided a communication network comprising a plurality of nodes linked by spans, to carry working paths between the nodes for use by working traffic and to carry protection paths, the nodes being arranged to use the protection paths for extra traffic, when the protection paths are not being used for working traffic, the nodes being arranged to use one or more of the protection paths for working traffic in the event of a fault on one of the working paths, and thus displace extra traffic from the protection path or paths used by the working traffic, the nodes further being arranged to use an alternative path to protect at least some of the displaced extra traffic.
Having at least some of the extra traffic protected rather than being discarded, in the event of a fault, can enable better bandwidth utilisation, regardless of how the working traffic is protected. Furthermore this way of achieving better bandwidth utilisation may be simpler to manage and operate compared to trying to achieve a corresponding bandwidth utilisation improvement in other known ways such as by providing more meshed paths for a mesh restoration of the working traffic. While this known method has the disadvantage of introducing more complexity into managing the switching decisions for the mesh restoration, the invention can be combined with such known methods or be regarded as an alternative.
Preferred Features
Preferably the alternative path comprises at least some of the protection paths not occupied by working traffic. This can be simpler and easier to manage and operate because these paths may be of the same type, same level of service and so on. Also this can improve levels of bandwidth utilisation, since more protected traffic can be accommodated.
Preferably at least some of the protection paths form a protection ring and the alternative path comprises the remainder of the protection ring to the part normally used by the respective extra traffic. This use of the other way around the protection ring provides a guaranteed alternative path and is easier to manage and operate than a mesh arrangement of protection paths. Each node need be aware only of the node in its own ring, not of the entire network. Rings inherently provide predetermined routes between nodes, making management and operation simpler and faster.
Preferably the nodes are arranged such that there is a predetermined configuration of which of the protection paths are used to protect respective ones of the working paths. This enables protection to occur more quickly.
Preferably the nodes are arranged to send the extra traffic simultaneously both ways around the protection ring. This simplifies the protection of the extra traffic because under single fault conditions, the extra traffic is already present on an alternative path and so the need for switching is avoided and the protection can occur more quickly. UPSR (unidirectional path switched ring) is an example of this.
Preferably the alternative path is formed by providing path loop back either side of the fault, the nodes further being arra
Dann Craig T
Langridge Dave
Swinkels Gerard L
Barnes & Thornburg LLP
Kizou Hassan
Mills Donald L
Nortel Networks Limited
LandOfFree
Data network having enhanced availability of extra traffic does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Data network having enhanced availability of extra traffic, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Data network having enhanced availability of extra traffic will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3206154