EP1952584A1

EP1952584A1 - Method for establishing a loop-free tree structure in a data transmission network and associated network element

Info

Publication number: EP1952584A1
Application number: EP06807694A
Authority: EP
Inventors: Bernd Stilling; Thomas Treyer
Original assignee: Nokia Siemens Networks GmbH and Co KG
Current assignee: Nokia Solutions and Networks GmbH and Co KG
Priority date: 2005-11-16
Filing date: 2006-10-31
Publication date: 2008-08-06
Also published as: CN101366244A; US20100220633A1; US20080304428A1; WO2007057301A1

Abstract

The invention relates to, among other things, a method during which a network element (58, 60) of a data transmission network (50) is automatically integrated into a method for establishing a loop-free tree structure or is automatically removed from such a method. By taking preset criteria into account, it is ensured that no loops can arise in the data transmission network (50) when removing a network element from the method for establishing a loop free tree structure.

Description

description

Method for determining a loop-free tree structure in a data transmission network and associated network element

The invention relates to a method for determining a loop-free tree structure in a data transmission network. A tree structure consists of so-called nodes, which correspond to network elements, and branches that lie between each two nodes and the connections between

Network elements correspond. The network elements are, for example, switches or bridges. A loop-free tree structure exists if there is no ring structure within the tree structure. Loop-free tree structures are, for example, one above the protocol layer 1, i. the physical layer protocol layer 2, i. the link layer, see, for example, IEEE (Institute of Electrical and Electronics Engineers) 802. ID, 1998, especially Chapter 8, where a so-called spanning tree algorithm and associated protocol are described.

However, even at higher protocol levels, there is the demand for loop-free tree structures, see, for example, protocol level 3, d. H. the so-called network layer. The known algorithms converge very slowly, in particular in the case of very many network nodes in a data transmission network. On the other hand, a manual or static shutdown of the inclusion in the procedures for determining the loop-free tree structure is very critical, which is why, for example, the company Cisco is required to maintain the STP protocol, if it is unnecessary ("keep STP even if it is unnecessary ").

For example, a Carrier Ethernet network consists of more than 70 or even more than 100 network nodes. The network topology usually provides redundancy through the use of ring structures and mesh structures. Methods such as STP (Spanning Tree Protocol) are used to manage these topologies. The use of STP is very complex and critical. STP and its newer versions, such as RSTP (Rapid STP) and MSTP (Multi STP) are limited in their scalability. With the standard STP parameters, the so-called network diameter is limited to 7 so-called hops. By optimizing the STP parameters, a network diameter of up to 19 hops can be achieved, but this is still not sufficient for the requirements of a Carrier Ethernet network.

The basic functionality of STP or other protocol layer 2 (L2) methods is not just resiliency. The basic functionality is also to keep the Layer 2 network free of loop under all circumstances. Loops are to be avoided for the following reasons in a layer 2 network: so-called broadcast or broadcast frames are transmitted infinitely long in the loops, so-called multicast or multi-call frames are transmitted infinitely long in the loops, the duplication of the Broadcast and multicast frames continue until the maximum data transfer rate of the Ethernet network is reached, all connections are completely filled with broadcast data transfer traffic, most of the queues in the network nodes are filled, the control processors in the Network nodes are overloaded, payload traffic is transmitted at a very high frame loss data transmission rate, user networks are flooded with broadcast and multicast messages, and - so-called in-band management of the network is no longer possible. In corporate networks, it has been learned that a misplaced patch cable, ie a cable with a length that is, for example, less than ten meters, or the addition of a new switch, can inadvertently cause a loop and break down the entire network. For these reasons, STP is an absolute requirement in corporate networks, even if layer 2 (L2) fault tolerance (resiliency) is not used in the company.

A carrier network or operator network has to guarantee a loop-free operation. All network elements should guarantee this with the standard parameters. When the network nodes and their parameters are reconfigured, loop-free operation should be guaranteed, even in the event of a misconfiguration. An increase or a change of the network should not lead to loops, even for a short time.

Thus, without a change, STP is not suitable for the required network sizes and can not support technologies in which a large number of access points are connected in a ring structure.

Nevertheless, it is an object of the invention to provide a simple and improved method for determining a loop-free tree structure. In addition, an associated network element should be specified.

The object related to the method is achieved by the method steps indicated in claim 1. Further developments are specified in the dependent claims.

In the method according to the invention, the following step is carried out in a network element or in each network element: in a network element of a data transmission network, automatic incorporation of the network element into a method for determining a loop-free tree structure or automatically extracting the network element from such a method depending on at least one or all of the following:

the number of further network elements directly connected to the network element,

the detection of the arrival or the detection of the

Lack of data to determine the loop-free tree structure, and

- The function assigned to the network element in the loop-free tree.

Considering the specified criteria ensures that loops are guaranteed even when reconfigured and incorrectly repackaged.

In the method according to the invention are all

Network elements according to their basic configuration switched so that they are included in STP. Alternatively, the network elements in the basic function are excluded from the STP method. Thus, the basic configuration is not important because it can be determined relatively quickly whether a network element is to be included in the STP method or whether it should be excluded from the STP method. A network element automatically detects whether or not an active STP instance is required for that particular network element. If a network element need not be involved in the STP process, this network element does not participate in the STP process, which is referred to as STP pruning. Only when incorporation of the network element into the STP process is required will it be included in the STP process. In this way, the number of network elements involved in the STP process can be reduced. This greatly increases the scalability of STP. The STP protocol itself is not changed.

The effectiveness of the method according to the invention also depends on the network topology. The inventive method is particularly effective in topologies in which a large number of network elements or network nodes are connected in a ring, in particular at the periphery of the network,. Two examples will be explained below with reference to the figures.

In a further development, the method is carried out in different network elements in the same way. The various network elements can both have the same structure as well as a different structure. For example, a program or hardware can be created once and used multiple times for different network elements. In this way, the maintenance of the program or the hardware is reduced.

In another development of the method according to the invention, the number of network elements directly connected to the network element is determined for the relevant network element. With a number greater than two, the subject network element is included in the method for determining the loop-free tree structure. If, on the other hand, the number is equal to two or, in one embodiment, less than three, then the network element is excluded from the method for determining the loop-free tree structure. This training is based on the consideration that at

Network elements in rings loop freedom can be ensured in other ways, for example, by the fact that only a network element of the ring structure is included in an STP process.

In another development of the method according to the invention, the network element is initially excluded from the method for determining the loop-free tree structure. It defines the beginning and the end of a test period. Within the test period, after the

Exempting the network element from the method for determining the loop-free tree structure detects the arrival or the absence of data that is used to set a serve loop-free tree structure. These data are contained, for example, in so-called BPDUs (Bridge Protocol Data Unit). If such data is received within the PrüfZeitspannen, the network element remains exempted from the process, because it is ensured that a network element that has sent the data, the STP performs and thus ensures the loop freedom in the ring. If, on the other hand, no such data is received within the test period, the network element is automatically included in the method after the end of the test period. In this way, it is ensured that at least one network element, for example in a ring, performs the STP method. Further methods can ensure that only exactly one network element in a ring structure performs the STP process, even if the ring structure is not connected to any other network structure.

In a next development, the network element is first included in the method for determining the loop-free tree structure. After inclusion, it is determined that the mesh element forms the root of the loop-free tree structure. After this determination, the network element remains involved in the process. If, on the other hand, it is ascertained after incorporating the network element that the network element is not the origin of the tree structure, the network element is excluded from the method again. By means of this procedure, it is possible to ensure, for example, that in a ring structure exactly one network element carries out the STP method, namely the network element which has been defined as the root of the loop-free tree structure in the ring structure. The development is particularly suitable for ring structures that are not connected to any other network structures of a data transmission network, d. H. for isolated ring structures.

In another development, the data for setting the loop-free tree structure according to the so-called Ethernet protocol is transmitted, see IEEE 802.3. However, the method according to the invention can also be applied to other transmission protocols.

In a development, at least one network element is a multiplexer for broadband connections or at least one network element is an optical multiplexer. In this case, a broadband connection is a connection with a data transmission rate greater than 500 kilobits / s in a transmission direction as used in connection with the x Digital Subscriber Line (xDSL) method, where x indicates a special DSL method, e.g. ADSL (Asymmetrical DSL).

In a next development of the method according to the invention, the method for determining the loop-free tree structure is a so-called spanning tree method, in particular:

the method according to IEEE 802. ID (STP),

the method according to IEEE 802. Iw (RSTP), or the method according to IEEE 802.1s (MSTP).

However, the inventive method can also be used in other methods for determining loop-free tree structures, especially at higher protocol levels.

In a next development, the data is transmitted in accordance with an optical transmission method. For example, data in optical data transmission networks can also be transmitted in accordance with the Ethernet protocol.

The invention also relates to a network element, in the operation of which the inventive method or a further development thereof are performed. Thus, the above-mentioned technical effects also apply to the network element.

In the following, exemplary embodiments of the invention will be explained with reference to the accompanying drawings. Show: FIG. 1 shows method steps for automatic removal of a network element from a method for

Determining a loop-free tree structure or automatically incorporating a network element in such a method,

2 shows the structure of an access data transmission network, FIG. 3 shows the topology of the data transmission network according to FIG. 2, FIG. 4 shows an optical data transmission network or an optical carrier data network (CDN), FIG. 5 shows the optical data transmission network according to FIG View of a data communication network

(DCN), and FIG. 6 shows the topology of the data transmission network according to FIG. 4 determined by means of the method according to the invention.

FIG. 1 shows method steps for automatic removal of a network element from a method for determining a loop-free tree structure or for automatic inclusion of a network element in such a method for determining a loop-free tree structure. The method begins in a method step S10. In method step S10, STP is switched off for the relevant network element or STP is switched off in method step S10. In an immediately following method step S12, it is determined in the network element in which the method steps are carried out how many other network elements are directly adjacent to the relevant network element. This number will also be referred to as degrees below.

In a method step S14, it is checked whether the degree determined in method step S12 is equal to two. If this is not the case, a method step S17 follows immediately after method step S14. In step S17, it is checked whether the degree is greater than two. If this is the case, follow the procedure S17 immediately a method step S18, in which the STP method is turned on in the relevant network element, so that this network element is included in the determination of the loop-free tree structure for the data transmission network. After method step S18, the

Procedure in a step S28 for the time being ended until e.g. there is a change in the topology of the data transmission network. If, on the other hand, it is determined in step S17 that the degree is not greater than two, i. the degree is 0 or 1, the method step S17 immediately follows

Method step S19, in which the STP is switched off for the relevant network element, in which the method steps shown in Figure 1 are performed.

If, on the other hand, it is determined in method step S14 that the degree determined in method step S12 is equal to two, method step S14 is followed directly by a method step S16 in which STP is switched off for the relevant network element. This is followed, as indicated by an arrow 2, the method step S28, in which the method is terminated, so that in the relevant network element, the STP method is not performed. Thus, the network element in question is disregarded when determining a loop-free tree structure.

The method illustrated in FIG. 1 is used, for example, for all network elements or all network nodes of a network

Data transmission network performed. However, for example, it is also possible to manually set individual network elements that are not included. As a result of a first variant V1, all network elements are included that have more than two adjacent network nodes. In contrast, all network elements are excluded from the process having only two, only one or no adjacent network element, in particular network elements having only two neighbors in a ring. In another exemplary embodiment of the variant V1, the query of the method step S17 is performed instead of the query in the method step S14, with "yes" to Step S18 is branched. If no, a branch is made to method step S16. The method step S19 is not required in this other embodiment.

In other words, each network element counts the number of active Network Network Interface (NNI) ports. An active NNI port is an NNI port, with the connection status "up and running." The role of the network element concerned and its properties are determined according to the number of detected NNI: -> 3 NNI: if the number of NNI > 2 is performed on the relevant network element STP / RSTP - 2 NNIs: if the number of NNI is exactly two, then these two ports are treated as ring ports by the network element: the STP protocol is turned off for that network element,

BPDUs (Bridge Protocol Data Unit) received on one ring port are transparently forwarded to the other ring port. These BPDUs are preferably forwarded with a higher priority than other frames. For example, the goal is to redirect these BPDUs to less than 5 milliseconds under peak load. The learning of MAC addresses (medium access) is completely switched off at the ring ports. From now on, the network element is no longer one

Bridge or bridge in the ring. It only works as a hub or as a distribution unit of data packets. As a result, each frame coming from a user is forwarded in both ring directions simultaneously.

Therefore, the network element no longer needs to evaluate the "topology changed" notifications of the STP protocol Transmission direction, ie from the network to a user, the bridge function, however, is still effective. A downlinked frame to a local user will only be forwarded to that user and will not be forwarded in the ring.

- 1 NNI: if the number of NNIs is exactly 1, the network element itself classifies itself as a so-called leaf node or edge node in the access network. No BPDUs are created or interpreted on the NNI port. All BPDUs received on the single NNI port are discarded without processing.

- 0 NNI: - the network element has no connection to others

Network elements, eg. Only a data transfer between two connected subscribers or users is possible.

Any event that connects or disconnects from an NNI port results in a recalculation of the number of NNI ports. As the number of NNI ports changes, the role and properties of the network element change accordingly.

In a variant V2, in addition to the method steps explained with reference to FIG. 1, the following method steps are carried out, wherein the jump performed by the dashed arrow 2 is not executed. If it is determined in method step S14 that the degree determined in method step S12 is not greater than two, then immediately after method step S14 there follows again a method step S16 in which the STP is deactivated for the relevant network element. After method step S16, variant V2 is followed immediately by a method step S20 in which the relevant network element checks whether BPDUs are being received. If this is not the case, then immediately after the method step S20 follows Step S22. In method step S22, the STP is activated for the relevant network element. After method step S22, in method variant V2, the method is terminated in method step S28, see dashed arrow 4.

If, on the other hand, it is determined in method step S20 that the relevant network element receives BPDUs, method step S28, d follows immediately after method step S20. H. the process is terminated, leaving the STP switched off for the relevant network element.

The variant V2 is carried out in another embodiment, even without the method steps of the variant Vl, whose function is then met by other methods.

The variant V2 and also a variant V3 explained below are used in particular when all network elements of the data transmission network are in a ring structure. Namely, if all the network elements forming the ring perform the method according to variant Vl (STP pruning), the ring would no longer be loop-free. At least one network element in the ring should perform STP. In typical networks, such as access networks, such a topology need not be considered. An access ring has at least one network element connected to the core data transmission network, resulting in at least one network element having three NNI ports. If the connection to the core is lost, all services will be interrupted, regardless of whether there is a flood of broadcast messages or not.

Nevertheless, the variants V2 and V3 are explained, which also allow, for example, in such cases, a loop-free operation of the network. In the variant V2 already explained, each network element with exactly two NNI ports will suppress the STP (STP pruning). In this mode of operation, each of these network elements checks to see if STP BPDUs are present in the ring. A timer is reset with each BPDU received at an NNI Ingress. However, if the timer reaches its final value without a BPDU being received, STP will be turned on for that particular network element. For example, the end time is five times the so-called "hello time" of BPDUs, which is, for example, two seconds.

This procedure ensures that at least one network element in the ring structure STP is performed. However, coincidentally, it can also be several network elements. In order to ensure that only exactly one network element STP performs, a variant V3 is carried out, which is explained below.

In variant V3, the method steps explained on the basis of variant V1 and variant V2 are carried out, but the jumps represented by arrows 2 and 4 are not carried out. In variant V3, immediately after method step S22, a method step S23 is followed in which the system waits until the root network element in the data transmission network has been determined. This is followed by a method step S24. In method step S24, the relevant network element determines whether it has become the so-called root of a loop-free tree structure. If this is the case, then immediately after method step S24 follows

Method step S28, in which the method is ended, wherein the STP remains switched on for the relevant network element.

If, on the other hand, it is determined in method step S24 that the relevant network element has not become the root of the loop-free tree structure, it follows immediately after Method step S24 is a method step S26. In step S26, the STP is turned off for this network element. Subsequently, the method is terminated in method step S28.

In other words, if the network element can assume that the method for selecting the root bridge is completed, it checks if it has become the root bridge or not. If the network element has not become the root bridge and still does not have more than two NNIs, then the network element will disable STP again. In particular, the so-called "forward delay timer" of the STP indicates the time required to select a bridge.

This results in the following processes in a ring:

1. a ring without STP is formed, d. H. the last patch cable is plugged in,

2. a large number of broadcast messages (broadcast storm) is triggered

3. no BPDUs are generated,

4. One or more network elements decide that STP is not yet activated in the ring and activate STP itself. The ring ports on these network elements are blocked in consideration of the typical STP timing values such as learning delay or forwarding delay.

5. a root bridge is selected

6. All ring ports go in the so-called forwarding mode except one port. 7. Disable all network elements except the root bridge (bridge) STP.

8. a single network element executes STP in the ring, namely the network element, which is also the root of the loop-free tree structure.

The variant V3 is in another embodiment without the method steps of the variant Vl or executed without the method steps of the variants V1 and V2.

Figure 2 shows the structure of an access data transmission network 50. The data transmission network 50 includes at its periphery a plurality of data transmission rings 52, 54 and 152 and 154 and other ring structures, not shown. In the data transmission ring 52, two collection units 56, 58 and five multiplexers 60 through 68 are interconnected using Ethernet lines 70 through 82 into a ring. The collection units 56, 58 are also referred to as Aggregatorswitch. For example, collection units of the type SURPASS hiD 6650 from Siemens AG can be used, which have been expanded by units with the aid of which it is possible to carry out the method steps explained with reference to FIG. The multiplexers 60 to 68 are also referred to as DSLAM (Digital Subscriber Line Access Multiplexers). For example, units SURPASS hiX 5630 or 5635 from Siemens AG can be used.

Alternatively, however, units of other companies can be used for the collection units 56, 58 and for the multiplexers 60 to 68. The data transmission ring 52 also includes other unillustrated multiplexer units. The data transmission rings 54 are also connected to the collection units 56 and 58.

In the data transmission ring 152 also a plurality of multiplexer units and two collecting units 156 and 158 are connected in a ring by means of Ethernet lines. The data transmission rings 154 are also connected to the collection units 156 and 158. In each data transmission ring 52, 152, 54, 154, two collection units are arranged for the sake of redundancy.

The data transmission network 50 also contains two collection units 160 and 162, for example units SURPASS hiD 6650 or 6670 from Siemens AG. The collection unit 160 is connected to the collection unit 56 via an Ethernet line 164 and to the collection unit 156 via an Ethernet line 158. The collecting unit 162 is connected to the collecting unit 58 via an Ethernet line 166 and to the collecting unit 158 via an Ethernet line 170. The communication network 50 also includes other network elements connected to the collection units 160 and 162.

Instead of the multiplex units 60 to 68, optical line termination units can also be used, i. OLTs (Optical Line Terminator). In an access network, there are a large number of multiplexers (DSLAMs) and OLTs that serve to aggregate traffic from thousands of users to, for example, an IP backbone. For redundancy reasons, the DSLAM / OLT are interconnected to ring structures. The access rings are, for example, each with two collection units 56, 58, 156, 158 connected to the core of the collection network (aggregation network). From the point of view of the standard STP, in the topology shown in FIG. 2 with respect to the data transmission rings 52 and 152, there are sixteen so-called hops or forwarding units. This has already reached the limit of scalability of STP. A special feature of the topology, however, is that the DSLAMs 60 to 68 each have only two ring ports as a connection to the access network. Therefore, it is possible to turn off STP in these DSLAMs 60 to 68 without affecting the redundancy or looping freedom. In this procedure, the topology shown in Figure 3.

FIG. 3 shows for the data transmission network 50 the topology that results when the method shown in FIG. 1 is executed for each network element. From the point of view of the STP, the DSLAMs 60 to 68 are no longer bridges, but so-called hops, ie distribution units 180 and 182, respectively. The collection units 56 and 58 are now connected to the same hub 180 from the point of view of the STP. However, this is a valid topology for STP. The number of forwarding units has dropped from sixteen so-called hops to six hops. Thus, the STP method converges faster or it can only ensure a secure convergence.

As can be seen in Figure 3, STP is performed only in the collection units 56, 58, 156, 158, 160 and 162, respectively. in network nodes having at least three ports to adjacent network devices. In contrast, STP is not performed in the multiplexers 60-68, as they each have only two adjacent network elements.

FIG. 4 shows an optical data transmission network 200 operated by a network operator. The data transmission network 200 contains two fiber optic

Data transmission rings 202 and 204 and a variety of other fiber links not shown.

In the data transmission ring 200, for example, five optical multiplexer units 210 to 218 become one

Ring structure connected. The multiplexers 210 and 212 are duplicated for redundancy and serve to redundantly couple the two data transmission rings 202, 204 and the redundant access of a network management system (NMS). If the multiplexers 210 and 212 are considered to be a multiplexer, there are in the data transmission ring 202 between each two adjacent multiplexers 212 to 218, for example, two or more than two amplifier units 230 to 244, which are interconnected by means of optical transmission lines 250 to 272. An optical

Transmission line 274 of data transmission ring 202 lies between multiplexers 210 and 212. In addition, a transmission line of data transmission ring 202 lies between multiplexers 210 and 212.

The multiplexer units 210 to 218 are, for example, SURPASS hiT 7300 multiplexer units of the company Siemens AG. These multiplexer units are also referred to as add-drop multiplexers. The amplifier units 230 to 244 are, for example, SURPASS hiT 7300 amplifier units from Siemens AG. However, it is also possible to use units of other companies for the multiplexers 210 to 218 or for the amplifier units 230 to 244.

The data transmission ring 204 is similar in construction to the data transmission ring 202, see, for example, the multiplexers 210, 212 and other multiplexers 220, 222 and 224.

The multiplexers 210 and 212 form a core data transmission network, also referred to as a backbone. The multiplexers 214 to 218 and the multiplexers 220 to 224, in contrast, are connected to further units, not shown, from which they collect data or to which they distribute data. In a data transmission ring 202, for example, more than 50 transmission channels, in particular 80 transmission channels, are transmitted at a data transmission rate of more than 20 Gbit / s, in particular of 40 Gbit / s. Such data transmission methods are also referred to as DWDM (Dense Wavelength Division Multiplexing). In another embodiment, Wavelength Division Multiplexing (WDM), Synchronous Digital Hierarchy (SDH), SONET, or any other suitable method is used instead of the DWDM method.

However, a data transmission channel in the data transmission rings 202 and 204 is for the management of the multiplexers and amplifier units. A gating unit 300 is connected to the multiplexer 212, for example via a line 314. Likewise, the multiplexer 214 is connected to a gating unit 302 via a line 316. From the meshing unit 300 or the tide unit 302 feeds a line 310 or 312 to a network management system NMS.

A transmission channel of the optical data transmission network 200 is used in each data transmission ring 202 or 204 for the control of the network. In this

Data transmission channel data are transmitted, for example, according to the Ethernet protocol.

FIG. 5 shows the optical data transmission network 200 from the perspective of the control network which operates on an Ethernet basis. From the point of view of the control network, the multiplexers 210 to 224 and the amplifier units 230 to 244 are so-called switches or bridges (bridges), which is illustrated in FIG. 5 by reference characters with trailing lowercase letters b, see, for example, Multiplexer 214b, which corresponds to the multiplexer 214 ,

Thus, Figures 4 and 5 show a typical DWDM network with two redundantly connected data transmission rings 202, 204. The network elements are: Optical Add-Drop Multiplexers (OADM) 210 through 224 and Optical Line Repeaters (OLRs) 230 through 244. For redundancy For example, the network management system NMS is connected to the DWDM network via two gateways (GW) 300 and 302, respectively. The gateways 300, 302 separate the internal data communication network (DCN) from the external Carrier Data Network (CDN). The gateways 300, 302 hide the internal IP addresses of the internal DCN, provide a so-called firewall and have additional functions. The

Carrier data network transmits user data, such as music data, video data, voice data and program data. By contrast, the DCN mainly transmits control data.

FIG. 5 shows the data transmission network 200 from the perspective of the DCN. In contrast to a routing network, the DCN is realized as a "switched network." In order to enable the data transmission network 200 shown in FIG default STP must be enabled on all network elements so that there are then 24 STP instances in this example. So many STP instances would dramatically increase the convergence time of STP.

However, due to the ring-based topology of the network 200, a large number of network elements have only two ring ports. Therefore, it is again possible to turn off STP in these network nodes without affecting the redundancy or the avoidance of loops. From the point of view of the STP, the network shown in FIG. 6 then results.

FIG. 6 shows the topology of the data transmission network 200 defined with the aid of the method explained with reference to FIG

Data transfer ring 202 STP in the multiplexers 214, 216 and 218 and in the amplifier units 230 to 244 off. These units are in terms of the STP method as a distribution unit 320, which is connected via the optical data transmission lines 250 to the multiplexer 212b and via the data transmission line 272 to the multiplexer 210b.

In the data transmission ring 204, on the other hand, STP has been deactivated in the multiplexers 220, 222 and 224 as well as in the amplifier units of the data transmission ring 204, so that these units represent, as regards the STP method, distribution unit 322 or hubs. The distribution unit 322 is over the optical

Data transmission line 278 connected to the multiplexer 212b and via the optical data transmission line 280 to the multiplexer 210b.

In the multiplexers 212b and 210b, however, the STP method was activated, in particular to avoid loops for the transport of data packets in the data transmission ring 202 or in the data transmission ring 204. The in FIG. 6 The topology displayed has only two network nodes or so-called hops. This also significantly reduces the convergence time of the STP process.

In other embodiments:

1.) During the change of the network in transition states, the number of NNIs of a single network element may change, so that STP is activated or deactivated. If the transition reduces the number of NNIs, the transition is not critical. If the number of NNIs is increased from two NNIs to three or more than three NNIs, the newly activated port should be blocked first. In a next step, STP is activated on the network element. If the loop-free tree has been calculated, the newly activated port will be unlocked according to the STP.

2.) Disabling MAC address learning in the ring ports has the effect of sending each uplink frame in both ring directions. The unnecessarily generated traffic is transmitted through the ring up to a blocked port of a collection unit where STP has disconnected the ring to avoid loops in the transmission of data. Depending on the traffic, the uplink traffic of one ring node may overlap with the downlink traffic of another ring node. In rare cases, this can lead to a reduction in the available bandwidth. However, a much greater influence on the bandwidth in the ring is the fact that the ring is not separated at an optimal location by STP, for example.

3.) A protection time for RSTP can be determined empirically 4.) For stable network operation, the root print should not be changed often. Therefore, the Wurzeibrucke and the redundant Wurzeibrucke should not undergo STP negative pressure. 5.) The proposed procedures can be activated or deactivated. The default value is activated. If the algorithm is deactivated, then the network element STP always goes through regardless of the current number of active NNIs relating to that network element. 6.) A so-called link aggregation can be used to increase the available bandwidth on a link. In these cases, the aggregated connection pays as an active NNI. To enable this, link aggregation and Link Aggregation Control Protocol (LACP) should be enabled by default on the ports. 7.) A network element may have a "subtending" interface to other network elements, in which case the subtending interface is paid as an NNI port, and cascaded interfaces are also paid as an NNI, because the topology is the same called dual homing for "subtended" network elements supported. This may be intentional or unintentional, so that a so-called plug-and-play method should treat a "subtending" interface as an NNI port.

The described methods avoid a dilemma that would occur in a static configuration: on the one hand, the network would not be loop-free without configuration. On the other hand, without a loop-free network, no configuration can be made by in-band management. On the other hand, the explained methods make it possible to ensure loop freedom even if a plug-and-play change of the network occurs.

The described methods also take into account the following considerations. After booting a network element, all its ports are blocked. In the next step, the network element captures the role of each of its ports. Two roles are significant: a so-called peripheral leaf port or leaf port is at the boundary of a network. A Loop can never originate over a leaf port because a leaf port is not connected to any other switch on the same network.

If a port is not a leaf port, it is called an NNI (Network Network Interface) port. An NNI port can be connected to other switches, thus risking looping.

The standard STP approach treats all ports as NNI ports to be sure. That's why STP works excellently in all topologies. RSTP adds the ability to set ports as leaf ports (operEdgePort is TRUE) through configuration. However, digital data switches or routers designed for specific applications may have additional capabilities to automatically detect if they have leaf ports without manual configuration.

Two examples have been given above: an access network using DSLAMs or OLTs and

Collective units are. The collection units or aggregation switches have only NNI ports. The DSLAMs and OLTs, however, are to be evaluated more accurately. All user ports are by specifying leaf ports. It is safe to assume that there are no loops in the data transmission via user ports or subscriber ports. Even though there is a loop between two users, the effects of such a loop will be limited only to the particular user, for example by the application of filters and controlling functions such as e.g. MAC

Address restrictions, blocking of multicast and so-called address anti-spoofing. Thus, all user ports in a DSLAM or OLT can be treated as leaf ports. For a DSLAM or OLT, only the ports for downlink traffic have the role of an NNI. For an Ethernet access switch with many Fast Ethernet ports and some Gbit ports, such as: B. type hiD 6610 Siemens AG, the assumption Fast Ethernet ports are leaf ports and the Gbit port is NNI ports.

For DCN in WDM systems or DWDM systems, on the other hand, in addition to the NNI ports that connect one network element to another network element, there are ports where there are connections to the NMS and / or to a local configuration terminal (Local Craft Terminal). There are gateways between the external NMS / LCT Ethernet ports and the internal DCN. In this case, no loop can be decided over the external port, because there is the gateway. Without a configured gateway on the external NMS / LCT Ethernet port, it is up to an operator to avoid looping through this interface.

Claims

claims

Method for determining a loop-free tree structure in a data transmission network, comprising the steps of: in a network element (58, 60; 212, 230) of a data transmission network (50) automatically incorporating the network element (58) in a method for determining a loop-free tree structure or automatic Exempting the network element (60) from such a method depending on at least one of the following or all of the following: the number of further network elements (62, 160) directly connected to the network element (58, 60), the detection of the arrival or the Capture the

Lack of data to determine the loop-free

Tree structure, the function associated with the mesh element (58, 60) in the loop-free tree structure.

2. The method according to claim 1, characterized in that the method is performed in different network elements (58, 60) in the same manner.

3. The method according to claim 1 or 2, characterized in that for the network element (58, 60) the number of the network element (58, 60) directly connected network elements (58, 60) is determined (S14) that in a number greater than two, the mesh element (56) is included in the loop-free tree determination process (S18), and if the number is two or less than three, the mesh element (60) is excluded from the loop-free tree determination process (S16 ).

4. The method according to any one of the preceding claims, characterized in that the network element (58) from the A method for determining the loop-free tree structure is excluded (S16) in that a check period is set, and that either within the check period (S20) and after exempting the network element (58, 60) from the method for determining the loop-free tree structure data for determining the loop-free tree structure are received at the network element (60), and that after receiving the data, the network element (60) remains excluded from the process (S22), or alternatively within the test period (S20) and after removal of the network element (58, 60) from the loop-free tree determination process, no data for determining the loop-free tree structure is received at the network element (60), and after the expiration of the check period, the network element (58) is included in the loop-free tree determination process.

A method according to any one of the preceding claims, characterized in that the network element (60) is included in the method for determining the loop-free tree structure (S22), either after the inclusion is determined, that the network element (60) determines the origin of the tree structure (S24), and that the network element (60) remains involved in the process for determining the loop-free tree structure even after this determination or, alternatively, after incorporating the network element (60) in the loop-free tree determination process, it is determined that the network element (60) does not form the origin of the tree structure (S24) and that the network element (60) after this determination is excluded from the method for determining the loop-free tree structure (S26).

6. The method according to any one of the preceding claims, characterized in that the data is transmitted in accordance with Ethernet protocol.

7. The method according to any one of the preceding claims, characterized in that at least one network element (58) is a multiplexer for broadband connections or that at least one network element is an optical multiplexer (214).

8. The method according to any one of the preceding claims, characterized in that the method for determining a loop-free tree structure is a Spanning Tree method, in particular a method according to IEEE 802. ID, IEEE 802. Iw or IEEE 802.1s.

9. The method according to any one of the preceding claims, characterized in that the data is transmitted in accordance with an optical data transmission method.

10. Network element (58, 60), with a control unit which automatically incorporates the network element (58, 60) into a method for determining a loop-free tree structure or automatically excludes the network element (58, 60) from such a method, the control unit being dependent of one of the following criteria: the number of further network elements (62, 160) directly connected to the network element (58, 60), detection of the arrival or absence of data for determining the loop-free tree structure, the function comprising the Network element (58, 60) is assigned in the loop-free tree structure.

11. Network element according to claim 10, characterized in that the control unit contains a unit, in whose operation a method step according to one of claims 1 to 9 is executed.