Switch system and method for the monitoring of virtual optical paths in an optical burst switched (obs) communication network

Field of invention

The present invention relates to monitoring the presence of a communication channel or optical path in a wavelength address optical burst exchange communication network.

Background to the invention

There are many layers associated with the supply of highly available, robust and scalable telecommunications infrastructure. Information exchange and transportation are two of the atomic functions that such networks need to support in order to provide efficient service over geographical distances. Traditionally, these features have used individual exchanges (Ethernet silicon switches, ATM switches, SDH cross-connects, etc.) and high bandwidth point-to-point transport pipes (eg, 10G and 40G WDM, as well as DWDM technology). Is being done.

With the advent of tunable lasers as a replacement technique for traditional distributed feedback lasers, the ability to support flexible optical circuits enables a more dynamic supply-centric means of supplying physical channels within known fiber capacitances. I have to. In fact, in next-generation optical burst switched (OBS) networks, the ability to deliver optical packet switch and transport (OPST) performance is virtually connected in optical transmission networks. It is an efficient and effective means of exchanging and transporting information between endpoints, and such networks typically use ring topologies and wavelength address bursts. In such a configuration, the physically connected rings constitute a distributed switch in which the wavelength of light emitted from the source laser creates a virtual path between the ports of the virtual switch. "The Optical Switching Revival: A recent paper entitled "Rebuilding Optical Networks for Packets" (Heavy Reading, Volume 7, Issue 3, March 2009) discloses this virtual switch concept and is the next generation of optical burst exchange (OBS). ) The network is heading towards it.

Traditional circuit-exchanged telecommunications systems operate on fixed-only communication channels that are permanently available for transmitting and receiving information. The completeness of such communication paths is monitored in several different schemes, for example SONET \ SDH telecommunications standards J0, J1, J2 used to generate Trace Identifier Mismatch (TIM). G709 has Trail Trace Identifier (TTI), IP \ MPLS has Bidirectional Forwarding Detection (BFD), and / or Ethernet has the new 802.1ag \ Y1731. In these schemes, information quanta are emitted through specific probe packets into the system as a means of ensuring channel continuity for the purpose of exchanging information, where the channels are the terminal equipment of the network. A messaging or higher level entity for checking continuity and connectivity between higher level communication functions embedded within.

OBS communication networks typically access physical channels in an asynchronous temporal manner. That is, access to the communication channel is not aligned with any system timing boundaries, but such access is derived by dynamic traffic load. In a real OBS network, due to the high aperiodic nature of the proposed access mechanism for the physical transportation infrastructure, the high level control functions of the network are all linked and before using such infrastructure. There is a need to ensure that there is continuous visibility into the state of the physical plant for which the availability of routes is fully known. No alarm failure occurs in an asynchronous OBS network where long periods of inactivity can exist before a given route is used in the network, and asynchronous, aperiodic OBS routes are time-limited. Probing on a base is a significant requirement for high system availability. An unalarmed failure indicates a situation in which a path failure is not detected due to data loss due to possible long cycle inactivity on the OBS path. Without such a deterministic mechanism, as proposed, any OBS-based transmission network is prone to unalarmed failure events and lack of resilience to carrier-class deployments.

Optical communication systems based on the Dense Wavelength Decision Multiplex technology exhibit wavelength-dependent performance. In wavelength address optical burst networks, services that overlay such infrastructure provide the lowest level of performance to ensure that all wavelengths in the network remain colorless throughout the virtual switch path. It is necessary to maintain. Colorless operation exhibits a wavelength-independent property of each path in such a virtual switch, i.e. any color can be assigned to perform the task of carting data on any path of the switch. is there. Each path is assigned a specific wavelength and each node or switch in the network is configured to transmit / receive a specific wavelength.

In a ring network, data is generally scheduled asynchronously on such virtual wavelength address switches so that all nodes have fair access to the switch fabric. By default, the route monitoring function is performed at the highest scheduling priority level and in a cyclical manner. This type of scheduling is disclosed by international application PCT2004 / 040722 filed by the same applicant as this application, Intune Networks Limited.

Therefore, in order to provide flexible network services, physical route monitoring of virtual routes in optical burst exchange networks is required to provide a highly available network infrastructure that overcomes the above-mentioned problems associated with OBS networks. There is a need to do.

According to the present invention, as shown in the appended claims, an optical burst switch in an optical burst exchange type (OBS) communication network is provided, and the optical burst switch is described. A virtual optical path from one communication node to multiple other communication nodes that allows traffic data to be transmitted and / or received between nodes over a physical optical path located in a ring network. Means for monitoring and maintaining the mesh, In the OBS communication network, it is a means for transmitting a data packet probe on a virtual route from one node to each other node, and the data packet probe information received by the switch transmits traffic data between the nodes. Means and means that provide information about the availability of physical optical paths for transmission To be equipped.

The network is a high-priority, asynchronous time-limited network that probates physical channels and provides a guarantee of availability of physical optical paths in virtual optical circuits, as seen in optical burst interchangeable networks. It uses the periodic emission and / or reception of the burst information. The present invention is particularly applicable to communication networks that use time bursts of optical wavelength division multiplexing and source wavelength addressing to destinations to achieve a complete N × N virtual switch path in a ring topology network. ..

In one embodiment, the data packet probe is sent using the periodic emission and reception of high priority asynchronous time-limited burst information.

In one embodiment, the probe information is generated on the source side of the virtual switch to allow source side protection exchange.

In one embodiment, a virtual path is formed by tuning the wavelength of a source laser with an N × N full mesh switch formed by such a path.

In one embodiment, the physical path monitoring function is in the band due to the wavelength address optical burst and has monitoring performance along the entire burst path.

In one embodiment, the insertion and filtering of the highest priority packet data periodically scheduled on the virtual switch is determined based on the data packet probe information.

In one embodiment, the invention comprises a second ring network that enables a bidirectional ring topology.

In one embodiment, the switch comprises means of scheduling the highest priority maintenance message throughout the distributed light exchange fabric, which is based on the fabric switch tuning the wavelength of the source tunable laser. The protection mechanism allows bidirectional route exchange to occur within a certain time limit if a failure is detected on one ring.

In one embodiment, each logical port of the virtual switch comprises an indication of full connectivity available for the transfer of data.

In one embodiment, each data packet probe contains a unique identifier.

In one embodiment, the data packet probe provides at least one parameter of quantitative information about the availability of physical optical paths between nodes before transmitting data from one node to another. Including.

In one embodiment, the quantitative information contains one or more of the following parameters: optical received power, error rate, behavioral and instrumentation data related to the currently or already probed virtual path. It provides a means of messaging high priority information from the data plane and / or slow physical channel degradation via so-called "clear" channels.

In one embodiment, the switch is a plurality of logical ports, where each logical port of the distributed switch collects a virtual path maintenance vector for the distributed switch in a cyclical manner. A target port and a means for creating a virtual path maintenance matrix for a complete N × N virtual switch, the matrix being a combination of all logical port vectors.

In one embodiment, each maintenance vector consists of a binary connectivity route map between each source and each destination of the switch so that it is addressed based on the source burst wavelength, where each entry is 0. Either (no available route) or 1 (available route).

In one embodiment, a physical path monitoring function that is within the band due to the wavelength address optical burst and has monitoring performance along the entire burst path is provided.

In one embodiment, a means is provided for equitable access to each of the virtual routes maintained by the distributed scheduling system, the scheduling system being inserted into the virtual circuit for route maintenance. Guarantee delay and jitter limits for data packet probe information.

In one embodiment, a means is provided for defining the order in which scanning will be performed by the data packet probe.

In one embodiment, the scanning order depends on the priority and load of traffic between the ports of the switch.

In one embodiment, the probe wavelength address packet is periodically and deterministically sourced by probing other optical paths in the system for "no alarm" in such a data plane nonexistent path. Ensure failure detection. A route that does not have such a data plane is a route that has not yet started service.

In a further embodiment of the invention, a method for transmitting and / or receiving data in an optical burst exchange (OBS) optical network is provided. A virtual optical path from one communication node to multiple other communication nodes that allows traffic data to be transmitted and / or received between nodes over a physical optical path located in a ring network. Steps to monitor and maintain the mesh, In an OBS communication network, a step of transmitting a data packet probe on a virtual route from one node to each other node because the data packet probe information received by the switch transmits traffic data between the nodes. Provides information about the availability of physical optical paths, steps and including.

In a further embodiment, an optical switch in an optical communication network is provided, the optical switch. Monitors and maintains a mesh of virtual optical paths from one communication node to multiple other communication nodes that allows traffic data to be transmitted and / or received between nodes via physical optical paths. Means to do and In the OBS communication network, it is a means for transmitting a data packet probe on a virtual route from one node to each other node, and the data packet probe information received by the switch transmits traffic data between the nodes. Means and means that provide information about the availability of physical optical paths for transmission To be equipped.

In another embodiment, an optical burst switch in an optical burst exchange (OBS) communication network is provided, said optical burst switch. Monitors and maintains a mesh of virtual optical paths from one communication node to multiple other communication nodes that allows traffic data to be transmitted and / or received between nodes via physical optical paths. Means for In the OBS communication network, it is a means for transmitting a data packet probe on a virtual route from one node to each other node, and the data packet probe information received by the switch is traffic data between the nodes. Means and means that provide information about the availability of physical optical paths for transmitting To be equipped.

In a further embodiment, a controller for an optical burst switch in an optical burst exchange (OBS) communication network is provided, said controller. Monitors and maintains a mesh of virtual optical paths from one communication node to multiple other communication nodes that allows traffic data to be transmitted and / or received between nodes via physical optical paths. Means for In the OBS communication network, it is a means for transmitting a data packet probe on a virtual route from one node to each other node, and the data packet probe information received by the switch is traffic data between the nodes. Means and means that provide information about the availability of physical optical paths for transmitting To be equipped.

It also provides a substantially faster, self-sustaining programmable hardware logic system that is further configurable via slower computer programs and merges / filters probe packets from / from the main data path. , The program performing this method can be implemented on a suitable recording medium or read-only memory.

The present invention will be more clearly understood from the following description of embodiments given by way of example only, with reference to the accompanying drawings.

<figref num="1">FIG. 6 illustrates a known optical ring communication network architecture using source wavelength exchange and ring topology.</figref><figref num="2">It is a figure which shows the distributed wavelength address switch in the optical ring network which shows a virtual and physical path.</figref><figref num="3">FIG. 6 shows virtual and physical paths probed in a bidirectional ring network architecture.</figref><figref num="4">It is a figure which shows the operation of the switch in the optical ring network which uses the virtual and physical paths by this invention.</figref><figref num="5">It is a flow chart which shows the route availability for each probe.</figref><figref num="6">It is a flow chart which shows the route availability for each probe.</figref><figref num="7">It is a figure which shows one Embodiment which showed how the route availability monitoring function is applicable to source exchange type protection.</figref>

Detailed description of the drawing

Then, referring to the drawing first, FIG. 1, a known optical ring with several external network interfaces 100, 101, 102, 103 connected to several ports 104, 105, 106, and 107. Communication network architectures have been demonstrated, but preferably these include fast tuneable lasers that are wavelength exchanged to address a given destination port and are located in a ring network. Ports 104, 105, 106, and 107 are connected from the transmitter to the receiver in a clockwise manner, eg, using optical fiber, by a physical optical path link 108. Each port 104, 105, 106, and 107 is capable of communicating with each other via a wavelength address optical virtual path. As an example, paths 109, 110, and 111 are set up between 107-104, 107-105, and 107-106 based on the different burst wavelengths transmitted from the fast tunable lasers present within 107. It is a route.

With reference to Figure 2, some such ports 200 and 201 on the ring architecture are shown and all supportable on such distributed wavelength address switches (4 ports in this case). Logical and virtual switch mesh paths are shown by dotted lines. 203 and 204 circle a group of virtual routes transmitted from or received by 201 and 200, respectively. In the example shown, λ₁And λ₂Is assigned to receive the 200 and 201 sides so that only these wavelengths are dropped on the receiver at each of these. All other wavelengths pass through 200 and 201, respectively, so that all traffic bursts from one port to another on the switch "virtually" travel directly to the exact wavelength-assigned receive port. This is to construct a virtual path in the distributed switch.

With reference to FIGS. 3 and 4, FIG. 3 shows an optical burst exchange network with a bidirectional ring topology. FIG. 4 shows the architecture of an optical burst switch in an optical ring network using virtual and physical paths according to the present invention. An optical burst switch, as shown in FIG. 2 or 3, is located in a ring network from a node that allows data to be transmitted and / or received between nodes via a physical optical path. Controls the mesh of virtual optical paths to multiple other communication nodes.

During operation, the switch merges the data packet probe 401 from the optical module 402 (watchdog) with the client service packet data 400. This data is transferred via a switch to wavelength allocation block 404 and forward to fast tunable laser 405, which, in the case of a bidirectional protection ring configuration, is temporal on fiber optic 412 or 413. Emit a burst limited to. On a periodic basis, the 401 probe packet is forwarded, assigned different wavelengths (404) to different virtual ports on the switch, and transmitted.

Referring to the lower part of FIG. 4, the detection of a failed virtual route is obtained from a data packet probe due to a physical link failure or a decrease in communication channel (eg, an increase in the number of bit errors). It is collected in a source-side manner for each virtual route, allowing source-side protection exchanges to be supported at the physical route level, and data from fault routes in the ring network is bidirectional route switching ring. Transferred to the protection path in the configuration. That is, the data packet probe information 410 and 411 emitted from other nodes in the network received at the switch provide information about the availability of physical optical paths for transmitting data between the nodes. Effectively, the data packet probe 401 provides a mechanism for transporting equivalent means of managing the fiber infrastructure, away from higher layers in the network and maintaining or monitoring endpoint-to-endpoint connectivity. The data packet probe can be configured to include parameters such as, but not limited to, optical continuity, optical power loss, accurate addressing, burst loss, etc., thereby ensuring accurate optical tuning. It is possible.

Each physical optical path is used by data traffic when what the client service needs is forwarding between ports through the optical path. Each node / port has a special traffic generator, the detector is that each node sends traffic to all other nodes independently of the client traffic flow and all nodes are independent of the client traffic flow. It is provided to receive special traffic from all other nodes. The operation of the switches 200 and 201 will be described in more detail later.

Unlike the case where individual optical channels are used to monitor and detect link failures or degradations, this path monitoring feature is within the band of wavelength address optical bursts and is complete (electrical and optical). It has monitoring performance along the burst path.

As an example, referring to FIGS. 3 and 4, a virtual route is available to ensure that the route is available to ensure a 50 ms protection exchange performance that is protected in the event of a route failure. Within the time limit cycle time, it is possible to scan around the physical ring node-by-node clockwise or counterclockwise (linearly port-by-port in an equivalent virtual switch). The time between such virtual route maintenance packets scheduled on the switch fabric ensures that all source-destination signals can ensure a 50 ms out-of-service metric on a given virtual switch scale. It is for doing. In the example of an 80 port (80 lambda) switch, the scanning time for 80 such route maintenance packets is preferably 5 ms.

The scanning mechanism described above is completely self-sustaining and independent of those with port-to-port traffic loads and persists for zero load cycles to ensure detection of "no alarm" failures. An additional advantage of this scanning mechanism between virtual switch ports is that the scanning order, port configuration size, and rate are fully programmable, and such parameters are within the virtual switch itself. It can be configured based on different parameters (for example, a slightly loaded port can scan at a slower rate). This allows the rate of detection performance to increase based on heavy traffic load or traffic priority between ports.

With reference to FIGS. 5 and 6, a flow chart showing the route availability for each probe is shown. The flow diagram outlines the logic for the inlet and exit sides of each route (ie, on the transmit and receive sides), as well as these closures for source exchanges.

In FIG. 5, the steps are as follows on the transmitting side. 1. Create a frame by creating a route availability frame for each destination with an accurate error rate. 2. Label the frame for the data path and mark it as priority 1. 3. Merge frames into the data path and buffer to the correct destination with the correct priority. 4. Route availability frames are forwarded to the destination node throughout the Burst fabric.

In FIG. 6, the steps are as follows on the receiving side. 1. Frames are received from all sources or destinations. 2. Detect the route availability frame in DataStream and transfer it to the monitoring function. 3. A route monitoring function exists for each source that sends traffic to this destination. 4. If route availability is lost, an interrupt will occur in the system.

Figure 7 shows how the route availability monitoring feature can be applied to source exchange protection. This application example shows a link between the reception monitoring function on each node and the transmission path function. The route availability frame transmitted from A to B is indicated as B if the route availability frame from which B is the source is received by A. Therefore, node B has information about the availability of the route from node B to node A, which can be used as a base for protection exchanges, i.e., using a second optical burst fabric.

A method for monitoring the physical path of a distributed packet switch formed by a mesh of virtual paths for failure or degradation provides full N × N mesh connectivity between nodes based on the burst wavelength. Will be understood. The switch allows the insertion and filtering of the highest priority packets that are periodically scheduled on the virtual switch, so that there is no traffic forwarded across the virtual route. It provides a means of route failure or degradation. Route maintenance information on the source side of the switch enables source-side protection exchange.

The switch is a high-priority, asynchronous time-limited switch that probates physical channels and provides a guarantee of availability of physical optical paths in virtual optical circuits, as found in optical burst interchangeable networks. It uses the periodic emission and reception of the burst information. The present invention is particularly applicable to communication networks that use time bursts of optical wavelength division multiplexing and source wavelength addressing to destinations to achieve a complete N × N virtual switch path in a ring topology network. ..

The route monitoring function can be performed by scheduling the highest priority packet throughout the wavelength address optical burst exchange network having a bidirectional ring topology, the operation of which will be described in more detail below.

In this embodiment, the data packet probe has the ability to generate the bandwidth of a free channel upon termination of communication, so that the probe can be used for control information in the event of a control channel failure. Will be understood.

Scheduling the highest priority maintenance message across the distributed light exchange fabric, which is based on the fabric switch tuning the wavelength of the source tunable laser. The method signal to the protection mechanism allows bidirectional route exchange to occur within a certain time limit, Each logical port on the distributed switch has an indication of full connectivity available to it for data transfer. It consists of two sides.

Each port of the distributed switch autonomously collects the virtual path maintenance vector itself for the distributed switch, and the virtual path maintenance matrix for the complete N × N switch is all logical. A combination of port vectors. The maintenance vector consists of a binary connectivity route map between each source and each destination of the switch so that it is addressed based on the source burst wavelength, where each entry is 0 (there are routes available). Either not) or 1 (available routes).

As an example, the following is a virtual route connectivity matrix, where the rows indicate the source (transmit) of the switch, the columns indicate the destination (receive) of the switch, and each destination port has a specific wavelength. Is. One entry at position (S1, D1) indicates a transmit, or loopback, that communicates with itself. With reference to the four port distribution switches in Figure 2, where all optical paths are available, the following matrix<tables num="1"><img id="000002" he="23" wi="149" file="JP5699397B2_D0001.tif" img-format="tif" img-content="drawing" /></tables>Represented by.

If there is a path failure between ports 2 and 3, the resulting path connectivity matrix will be:<tables num="2"><img id="000003" he="23" wi="149" file="JP5699397B2_D0001.tif" img-format="tif" img-content="drawing" /></tables>Is.

Fair access to each of the virtual routes is maintained by a distributed scheduling system. Such a scheduling system is implemented to guarantee delay and jitter limits for route monitoring burst information inserted into the virtual route for route maintenance. Fairness on the switch fabric is not affected by the inclusion of such pathway maintenance features.

The present invention supports fast and slow scanning mechanisms across different logical parts of the switch to optimize availability bandwidth performance and maintain availability metrics based on load and traffic type. It will be understood that it will bring about the performance to do.

The number of wavelengths on the virtual switch, and thus the number of ports on the virtual switch, is theoretically unlimited, but in preferred embodiments, the number of wavelengths is routinely found in ITU DWDM-enabled networks. In addition, it has 40 or 80 wavelengths.

The present invention also provides means for defining the order in which scanning will be performed or repeated within a given scanning cycle. As an example, this scanning order can be determined based on the priority and load of traffic between the ports of the switch.

In another embodiment of the invention, probe wavelength address packets replicated at each passive optical splitter in the photonic subsystem probe other optical paths in the system with packets that are periodically and deterministically sourced. Ensure detection of "no alarm" failures in routes where such a data plane does not exist.

Suitably, such probe packets provide a means of probing wavelength-dependent paths in the anti-collision subsystem of a wavelength address optical burst switch system.

Preferably, the probe packet contains information related to the type of packet, ie, some fields containing behavioral and instrumentation data related to the currently or already probed virtual path, or each distributed switch port. Provides a means of messaging high priority information from the data plane. Such messaging channels allow destination information (eg, received optical power, channel bit error rate) to be messaged back to the source, and source-side optimization of the transmit burst (eg, increased transmit power, etc.). Or increase the source-side pre-charp of the data burst).

The embodiments of the present invention described with reference to the drawings include processing performed in a computer device and / or a computer device. However, the invention also extends to computer programs, in particular computer programs stored on or within carriers that are configured to carry out the invention. The program may be in the form of source code, such as a partially compiled form, in the form of object code, or in the form of intermediate code between the source and object code, or according to the invention. It may be any other form suitable for use in the implementation of the method. The carrier can also include a storage medium such as a ROM, such as a CD ROM, or a magnetic recording medium, such as a floppy disk or hard disk. The carrier may be an electrical or optical signal that can be transmitted via an electrical or optical cable, wireless or other means.

The present invention is not limited to the embodiments described herein, but can vary in both structure and detail.