Multiplex communications – Data flow congestion prevention or control – Control of data admission to the network
Reexamination Certificate
2000-03-31
2004-05-11
Kizou, Hassan (Department: 2662)
Multiplex communications
Data flow congestion prevention or control
Control of data admission to the network
C370S351000, C370S465000
Reexamination Certificate
active
06735170
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates generally to communications networks and, more particularly, to methods and systems for establishing high-capacity connections in such networks.
BACKGROUND OF THE INVENTION
In a communications network, traffic generally travels from a source to a destination though one or more intermediate nodes, also known as cross-connection hubs, which may be spaced hundreds of kilometers apart. Different geographical regions of a large network are usually owned and run by entities known as network operators, network administrators or telcos and several such network operators may join forces to establish a communications network which spans an entire country or continent.
Usually, the network operators serve to provide capacity which is then loaned to smaller entities known as service providers. Each such service provider may be desirous of establishing a relatively permanent, high-capacity connection between two points in the large network. Each such high-capacity connection may be intended to provide communications services for multiple end users, also known as subscribers. Thus, the high-capacity connection associated with a service provider may be referred to as a composite connection.
Each composite connection generally has a capacity that is still considerably less than the switching capacity of a single network node. Thus, several composite connections associated with different service providers and having their own respective sources and destinations may nevertheless be processed by the same switching node if it so happens that these composite connections share the same portion of the route which passes through the switching node in question. Therefore, it is often the case that individual switching nodes end up processing traffic streams belonging to multiple composite connections that are associated with different service providers.
In today's networks, traffic commonly consists of digital frame-based signals which abide by a synchronous standard such as SONET (Synchronous Optical Network), which is described in Bellcore document GR-253-CORE, Issue 2, December 1995, Revision 2, January 1999, hereby incorporated by reference herein in its entirety. The elementary unit defined by the SONET standard is a synchronous transport signal having a line rate of 51.84 Mbps, commonly referred to as an STS level 1 (or STS-1) signal.
As is known in the art, each frame of an STS-1 signal has a relatively small number of bytes in a portion known as the “transport overhead” and a significantly larger number of bytes in a portion known as the “synchronous payload envelope” (SPE). The transport overhead contains the so-called section overhead and line overhead, and it performs the functions needed to transmit, monitor and manage the SPE over a fiber system. It also contains an STS-1 payload pointer, which indicates the start of the SPE. For its part, the SPE consists of a path overhead and a payload. The path overhead supports the monitoring and management of the payload, while the payload carries the voice or data signals being transported.
In many cases, it is desirable to transmit signals at a rate higher than the relatively low basic SONET line rate of 51.84 Mbps. The SONET standard provides for at least two ways of transmitting higher-bandwidth signals at line rates that are a multiple of (e.g., N times) the basic SONET line rate of 51.84 Mbps. Both techniques involve mapping a high-rate data stream into the SPE of a high-rate SONET signal. Specifically, both techniques require initially chopping up the high-rate data stream into N SPE segments per frame.
Using a first technique, the SPE segments are then multiplexed on a byte-by-byte basis and transported by an STS-N signal. The individual SPE segments travel independently across the network and may reach their destination with slight relative delays. Hence, it is beneficial to employ this first technique when it is desired to transmit N independent STS-1 signals in parallel.
Using a second technique, the SPE segments are not multiplexed but rather are transmitted in strict order by an STS-Nc signal (known as a “concatenated” signal). The fact that the signal is concatenated is indicated by a special arrangement of bytes in the transport overhead of the STS-Nc signal which identify one of the portions of the SPE as being the last such portion in the frame. The STS-Nc signal carries the individual SPE segments, in order, until they reach their destination. Hence, it is beneficial to employ this second technique when it is desired to transmit a high-rate data stream which cannot be broken down into smaller components.
The hardware needed to process an STS-N signal is different from the hardware required to process an STS-Nc signal. For example, it is noted that in order to process an STS-N signal at a switching node, it is feasible to de-multiplex the STS-N signal into its N STS-1 components and to process each component separately. This is an advantage in situations where N is high and where a single ASIC (application specific integrated circuit) is incapable of processing the entire STS-N signal. In such a case, an appropriate number of additional ASICs can be provided to help process the N de-multiplexed STS-1 signals in parallel.
In contrast, an STS-Nc signal may not be broken up (e.g., through de-multiplexing) because the sequence of bytes in the SPE is important. In order not to break up the signal at each step of the way along its route, the same application-specific integrated circuit (ASIC) must usually be employed to process the entire STS-Nc signal.
Since the two types of higher-capacity signals have their own specific advantages and useful applications, it is usually the case that a composite connection to be established by a service provider will consist of both STS-N signals (for different values of N) and STS-Mc signals (for different values of M). In order to allow a particular combination of STS-N signals and STS-Mc signals to be transmitted, it is necessary for the service provider to verify with each network operator that suitable hardware resources are available along each segment of the route.
There are two main limitations which can affect the establishment of a composite connection for accommodating multiple lower bandwidth signals (including some STS-N signals and some STS-Mc signals). The first is strictly a capacity limitation, i.e., the hardware at each intervening node must have sufficient available resources to process the totality of signals in the composite connection. This can easily be verified by comparing the total available capacity of a node to the required capacity of the composite connection. The total available capacity will, of course, be affected by existing traffic (e.g., other composite connections) currently being handled by the node.
The second limitation is more subtle and is one which is imposed by the specific hardware configuration of each node. For instance, the available hardware must be distributed in such a way that it is capable of processing the concatenated signals forming part of the composite high-capacity connection. Some distributions of hardware will permit the processing of signals in one category (e.g., STS-N) while not allowing the processing of signals in another category (e.g., STS-Nc). Moreover, the manner in which the available hardware resources are distributed depends largely on the way in which the presently occupied resources were allocated for processing the signals belonging to the other composite connections currently passing through the node in question.
To consider a simple example, if a node is equipped with 4 ASICs, each of which has just enough residual capacity to process three STS-1 signals, then processing of an STS-12c signal cannot take place at the node in question, even though the total available capacity is sufficient to process an STS-12c signal. This is due to the fact that the entire STS-12c signal must be processed by a single ASIC, none of which is available in this case. Thus, it would not be p
Kizou Hassan
Nortel Networks Limited
Swickhamer Christopher M
LandOfFree
Method and system for establishing content-flexible... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and system for establishing content-flexible..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and system for establishing content-flexible... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3228197