WO2020048643A1 - Method for reserving at least one transmission path, network node, computer programme, and computer-readable medium - Google Patents

Abstract

The invention relates to a method for reserving at least one transmission path (P) for transmitting data packets from a stream initiator (T) to at least one stream participant (L) in a network, in particular an industrial network, which comprises a plurality of network nodes (B1-B9, 12, 13), wherein the stream initiator (T) issues a notification message (2, A1, A2) for a stream, wherein at least one item of forwarding information (3), which represents at least one predefined transmission path (P) or transmission tree (SBn) for propagating the notification message (2, A1, A2) or from which at least one such message can be derived, is provided by being inserted in the notification message (2, A1, A2) by the stream initiator (T) and/or by at least one network node (B1-B9, 12, 13), and/or by being transmitted, together with a stream identifier, in particular a stream ID, associated with the stream and contained in the notification message (2, A1, A2), to at least one network node (B1-B9, 12, 13) in a forwarding-information message which is separate from the notification message (2, A1, A2). The invention also relates to a network node (B1-B9, 12, 13), a computer programme, and a computer-readable medium.

Description

description

Method for reserving at least one transmission path, network node, computer program and computer-readable medium

The invention relates to a method for reserving at least one transmission path for the transmission of data packets from a stream initiator to at least one

Stream participants in an industrial network, in particular, which comprises several network nodes, the stream initiator issuing an announcement message for a stream. In addition, the invention relates to a network node, a computer program and a computer-readable medium.

In the IEEE, extensions known as AVB (Audio Video Bridging) have been defined for the protected transmission of a stream (periodic data transmission). In particular, the CBSA shaper introduced here (CBSA is the abbreviation for: Credit Based Shaping Algorithm) together with access control using the so-called Stream Reservation Protocol (SRP, see in particular IEEE802.1Q) Secure data transmission with determinable maximum latency is possible in the network. The protected data is transmitted in the form of periodic streams.

In the successor group of AVB entitled TSN (Time Sensitive Networking), further optimizations and additions were added to cover the requirements of other applications and thus to be able to use Ethernet in other markets. A variety of standards are part of TSN. In this context, time synchronization (IEEE802. LAS-Rev), frame preemption (IEEE802. LQbu) and reservation (IEEE802. LQca, IEEE802. LQcc) may be mentioned purely by way of example. The stream reservation protocol is dependent on path information from the network for the establishment of a stream. By distributing an announcement message for the stream, for example a talker advert along the path defined by the network (with AVB: all active ports) and the subsequent registration of at least one recipient, the path of the data and the necessary configuration of the shapers on the connection between sender and Receiver (s) set up. With TSN in particular, a transmitter is also referred to as a talker and a receiver as a listener.

It should be noted that in addition to transmission of stream data packets from one talker to exactly one listener or from one talker to several listeners, it is also possible for several talkers to send to exactly one listener. This scenario, which can also be referred to as a "multiple talker per stream", is described in the European patent application, which also goes back to the applicant, with the file number 18 154 319.0.

The announcement message for the stream is expediently sent either by one sender (talker) or - in the case of "multiple talker per stream" - by one receiver (lister). The stream initiator, i.e. the participant who initiates a stream, can thus be both a transmitter / talker and a listener / receiver. The stream announcement message can accordingly be a talker or listener advertise.

In order for a stream participant to be able to log on to a stream announced by a stream initiator, a participation message is sent by one (or more than one). In the event that one or more listeners log on to a stream announced by a talker, this listener in particular sends a listener ready or list ner join. In the event that one or more talkers log on to the stream, it will send or send it in particular a Talker Ready or Talker Join (see European patent application with the file number 18 154

319.0).

As far as path selection is concerned, AVB, since it focused on the home network area as its target market, has opted for RSTP, i.e. the Rapid Spanning Tree Protocol (see for example

IEEE802.1w).

There is a need for improved path selection for streams, especially for AVB and TSN. Because by using the shortest paths in a network, shorter transmission times can be achieved. In the

IEEE802.1Qca mechanisms are defined to enter a forwarding for the stream address in the network. The

Stream reservation has already considered restrictions in the distribution of the announcement message (stream description). Control of the forwarding is accordingly possible by restricting the forwarding of the stream data address. The corresponding restriction is that some ports of the network nodes between the stream initiator and stream subscriber (s) are hidden.

When using so-called seamless redundancy, in particular according to IEEE802.1CB, two paths that are disjoint to each other are required for each stream. According to IEEE802.1Qca, this can be achieved by setting up paths for the stream addresses for two different VLAN IDs. This enables automatic configuration of seamless redundancy / CB, as described in WO 2019/001718 A1, which also goes back to the applicant.

For the targeted forwarding, it is necessary to set up the forwarding address for each stream. With a new stream registration, a separate tree must be calculated for each stream and all network nodes, such as bridges, must be configured accordingly before the necessary configuration for the stream can take place. Furthermore, for the protection of streams or the periodic transmission of stream data packets, local forwarding must be configured in each network component between the sender (talker) and receiver (listener) of a stream. In addition, an internal resource check must be carried out and a list of the resources already required (reservation) must be kept. According to IEEE802.1, the network components or network nodes are bridges in Layer 2 Ethernet networks.

Subsequent, secure transmission of data packets is only possible through a sensible configuration of the mechanisms for streams.

Various hardware mechanisms and different models for configuring the mechanisms have arisen in the IEEE. The configuration models are standard

IEEE802.1Qcc. The aforementioned Stream Reservation Protocol has been introduced at AVB in order to carry out an automatic configuration. There are different approaches for the different use of TSN mechanisms, from the central configuration, i.e. setting up every network node, for example every bridge, through a control center to an extended protocol for the automatic setup, the standardization of which is just beginning (see especially the IEEE project "P802.1Qdd - Resource Allocation Protocol").

The mechanisms defined in the AVB group are limited to streams with one stream class. By using a special address in the Ethernet network, the so-called Stream Destination Address (Stream-DA), control of the data transfer is possible. With SRP there is an automatic configuration in the network and after a successful reservation the path from the listener to the talker is activated. If resources are insufficient, however forwarding is not activated and data transmission is prevented.

With AVB, it is not possible to select the transmission path in the network when making a reservation. With RSTP, only one loop-free tree is active in the network.

With TSN, several algorithms are described in IEEE802.1Qca, how a path definition can be made in the network. However, the processes involved in the reservation are not dealt with in more detail.

As an additional hardware mechanism, time-based control of the queues available for the different priorities in a network node is described in the IEEE. The creation of a configuration for this mechanism, also known as a schedule, is not further explained. In order to get the information for the schedule, a special variant of SRP was described, which stores the registrations from the end device (listener) in the network node and supplies configuration information to the end device. The network node is not configured by SRP. Only additional information that the network node received through a configuration is delivered to the terminal.

In the opinion of the applicant, an expansion of SRP is necessary for a more convenient automatic network configuration of a network with AVB_ or TSN mechanisms, which is associated with less effort. So far, as far as the applicant is aware, a reservation is only possible in a Layer 2 network due to the restriction to Ethernet. IP connections or MPLS networks (MPLS stands for Multi Protocol Label Switching, see in particular IETF RFC 3031 - Multiprotocol Label Switching Architecture) cannot be transmitted across router boundaries due to the direct communication between SRP and the neighboring network node (bridge) . The extended descriptions of the special Variant of SRP for the additional information for end devices in the IEEE802.1Qcc standard already contains IPv4 and IPv6 addresses for local configuration of the end device interfaces.

Starting from the prior art, it is an object of the present invention to provide a method of the type mentioned, which provides protected transmission with low configuration effort and optimal use of the available network topology.

This object is achieved in a method of the type mentioned at the outset in that at least one forwarding information which represents at least one predetermined transmission path or transmission tree for the dissemination of the announcement message or from which at least one can be derived is provided by: is inserted into the announcement message by the stream initiator and / or by at least one network node, and / or by inserting at least one network node in a forwarding information message separate from the announcement message together with a message associated with the stream in the announcement message contained stream connoisseurs, in particular

Stream ID being transmitted.

Through the provision of forwarding information according to the invention either directly in the announcement message or alternatively or additionally in another way, it is possible for the distribution of a stream announcement message and thus the data transmission following the establishment of the stream to flexibly select an optimal path hold true. The forwarding information provided in the manner according to the invention represents a predetermined transmission path or transmission tree in the network which can or will be used for the dissemination of the announcement message or at least one such can be derived from the forwarding information. The announcement message for a stream preferably comprises or is given by a stream description. The announcement message can be a unique ID (StreamID) assigned to the advertised stream, a designated one

Stream address (StreamDA), the specification of a virtual network (VLAN), a traffic specification (English: traffic specification, TSpec), which preferably include the specification of a maximum frame size and / or a maximum number of frames per interval , an accumulated latency and possibly error information. The latter two details can change or change during the transfer in the network. The latency is accumulated when nodes are passed on to nodes, and error information is, among other things. recorded when there are insufficient resources available for the stream at a node or port. The parameters or information mentioned are those that are provided for a talker advert in IEEE 802. IQ, in particular clause 35.

If the forwarding information is inserted into the announcement message, this provides in particular an addition to a stream identifier, for example a StreamID, and / or in addition to a stream address, for example a StreamDA, and / or in addition to the specification of a virtual network , such as a VLAN ID. Preferably, the forwarding information, when inserted into an announcement message, represents information contained therein in addition to all of the above parameters / information available according to the standard.

Via the announcement message (in particular talker or listener advertise) and / or in response to the receipt of such a participation message sent by a stream subscriber (in particular listener or talker join or

Ready) can in the manner known from the prior art at the ports of network nodes on the transmission path resources for a transmission of stream data packets between the stream initiator and the at least one stream subscriber are reserved or, if available, are reserved.

Following the reservation of one or more transmission paths while performing the method according to the invention, data packets are preferably transmitted between at least one transmitter and at least one receiver using fourth resources. Resources, in particular network resources that are reserved for transmission via stream to participating network nodes, can be, for example, address table entries and / or frame buffers and / or transmit time slices and / or bandwidth and / or shapers settings, jitter and / or limit latency.

It has proven to be particularly suitable if an expert is used as forwarding information, which is assigned to the network node that is closest to the stream initiator sending the announcement message. This network node can also be viewed and designated as a marginal node (from the point of view of the stream initiator). According to a particularly preferred development, the bridge ID of the edge node is provided as forwarding information. A bridge ID is preferably to be understood as a system ID assigned to the edge node in accordance with a routing protocol. An identifier assigned to a node can be a MAC address assigned to it.

The basic idea of this particularly preferred embodiment is not to calculate and set up a separate network path for each stream, as is made possible by the expansions in IEEE 802.1 Qca, but rather the network path from the point of view of the network node, in particular the bridge use, which is the first to receive an announcement message sent by a stream initiator. This node is the edge node on the stream initiator side. By using the forwarding routes of nodes in the network for streams, the need for a calculation for each stream is eliminated when a stream is newly announced by its stream initiator. The trees / paths can be calculated in the network regardless of the stream reservation. In the event of a topology change, it is no longer necessary to recalculate all stream paths and enter the forwarding routes for the stream addresses, as required by using IEEE 801. Qca, before the reservation can be updated on the new paths. It is sufficient to recalculate the network paths on the basis of the new topology for the nodes, such as bridges, and to automatically update the reservation with the stream reservation protocol.

By using the path of a bridge, the stream registration and the path are independent of the stream initiator (for example a terminal device which represents a talker or listener for a stream), which sends the announcement message, in particular triggers the stream registration. So that the other network nodes know the path to be used, the announcement message from this first network node, that is to say the edge node, can advantageously be expanded by an expert, in particular a ID or the ID of the edge node.

Alternatively or additionally, the assignment of the stream ID and the edge node that received the announcement message first is possible in a separate protocol, which corresponds in particular to the variant that a forwarding information message separate from the announcement message is sent. When querying the forwarding through the stream reservation, the assignment of the stream identifier and forwarding information collected or provided via the forwarding information message, in particular via another protocol, can then be evaluated and the forwarding of the announcement message can be restricted accordingly . The calculation of the forwarding paths for streams is therefore only dependent on the network when using the paths from the network edge nodes. A stream registration can take place at any network node. The forwarding of the announcement message is then preferably carried out in a modified or by means of a modified stream reservation protocol on the basis of the already existing forwarding path or tree for the connoisseur, in particular the ID of the peripheral node that was the first to receive the announcement message.

For each network node, at least one calculated tree that can be used for forwarding is expediently calculated and known in the network, for example by IS-IS, in particular using SPB (Shortest Path Bridging) and / or that in IEEE 802.1 Qca referenced

MRT (maximally redundant trees) algorithm. As far as the term "paths / trees known in the network" is concerned, it is generally the case that not all nodes know the course of a complete tree / path, but only each node knows how it receives announcement messages received from a neighboring node at a receiving sport / Forward user data, specifically via which or which send ports, which - viewed globally - results in the complete paths / trees via the network.

The use of additional information from the Edge Bridge ID to determine the forwarding path of a stream enables the use of only one VLAN identifier (VID) in the entire network for all paths - since the paths of the streams are configured by the reservation protocol.

The ISIS protocol may be mentioned as an example of a routing protocol via which or according to the edge node or all nodes in the network preferably / are assigned a system ID. The ISIS or IS-IS protocol is, in particular, the "Intermediate System to Intermediate System Protocol", preferably in accordance with ISO / IEC 10589: 2002 (E). In the frame As a rule, it is provided that the network nodes each send out information at least about their direct neighbors and that each network node receives "knowledge of the topology" on the basis of the messages received from its nodes and the information about its neighbors, for example builds up a database or list with topology information, in particular path information.

Another preferred embodiment of the method according to the invention is characterized in that the edge node identifier is inserted by the edge node in the announcement message, which has proven to be particularly expedient.

The insertion by the edge node is possible in a simple manner and the correspondingly expanded announcement message can then be disseminated in a targeted manner using the expert.

In a further advantageous embodiment, it can be provided that at least one set of predefined trees is set up in the network, the or each set comprising at least one predefined tree for several, in particular each network node, via which the respective network node as the root has several others, in particular can reach any other network node in the network.

Then it can further be provided that two or more sets of predefined trees are set up, each set comprising at least one predefined tree for several, in particular each network node, via which the respective network node as the root has several others, in particular each the other network node in the network, and the trees of different sets have been calculated using different algorithms. Then it can also apply that each set of predefined trees is assigned its own virtual network, and preferably a virtual network is provided in addition to the forwarding information, and / or a separate class is set up for each set of predefined trees , and preferably too in addition to the forwarding information, the specification of one of these classes is provided.

For a given network, different paths can be calculated using different path calculation algorithms. These can be distinguished and selected via assigned virtual networks and / or "availability classes", similar to the stream classes already provided according to the prior art. For example, paths can be calculated for a given network once according to ISIS, in particular with SPB and possibly automatic switching, and once according to MRT, in particular for redundant disjoint trees, and one of the two algorithms can be selected by one of then two possible VLANs, in particular an associated VLAN ID, or one of two Availability classes introduced for this purpose is selected. The availability classes are preferred, as described in the older international patent application with the application number PCT / EP2018 / 086124, selection aids for choosing the right class in the network with the same latency properties but different redundancy mechanisms to meet the requirements regarding the availability of stream transmission in the case of a network error and the associated necessary reconfiguration time. The introduction of an availability class makes sense to reserve only the necessary resources in the network, since, for example, with seamless redundancy, two active reservations are always set up in the network for each stream, but there are applications that can cope with a temporary loss of communication. The availability classes then deal in particular with the selection of a tree or path or serve this by selecting the required redundancy mechanisms in the network. The forwarding routes in the network are thus determined by the availability class, the latency in the network is defined by the stream class. A different combination of both classes is possible in a network. The configuration of the network nodes, for example bridges, and the forwarding of the announcement message to the next node can then preferably take place together with availability (for example whether CB, that is to say seamless redundancy is activated in a given network or not). End devices can be connected to each node. For each node, the pre-calculated tree or several pre-calculated trees are available for forwarding a stream and are known in the network. At CB, both MRT trees are known for the selected edge node expert, in particular the edge node ID and thus for the stream in the network.

In a further particularly advantageous embodiment of the method according to the invention, provision can be made for the at least one forwarding information to be provided by the stream initiator and / or at least one network node in the type-length-value format (TLV format). is inserted into the announcement message.

 This way of providing the forwarding information has proven to be very advantageous and offers, among other things, a great deal of flexibility and was already used for the additional IP information from IEEE 802.1 Qcc in the reservation protocol.

The TLV structure in particular makes it possible to select or support any desired combinations, for example of Ethernet, IP, MPLS and / or other transport variants and / or tunnel techniques. Accordingly, in a development of the method it can be provided that the at least one forwarding information comprises or is formed by at least one Ethernet forwarding information and / or at least one IP forwarding information and / or at least one MPLS forwarding information . MPLS forwarding information can be, for example, one or more MPLS labels that can be inserted, for example, by one or more network nodes in received announcement messages. Provision can also be made for at least one forwarding information from at least one network node to be inserted into an announcement message received by the latter, which is available on the basis of a local configuration of the network node.

Another embodiment of the method according to the invention is further characterized in that at least one network node that receives an announcement message with at least one forwarding information contained therein changes at least one forwarding information contained in the announcement message and / or at least forward one Information is removed from the announcement message, the change and / or the removal preferably taking place in accordance with a local configuration of the at least one network node.

Provision can also be made for at least one forwarding information to be inserted into at least one participation message for a stream from the stream subscriber who is issuing it and / or from at least one network node, and / or that at least one network node is used, who receives a participation message with at least one forwarding information contained therein, changes at least one forwarding information contained in the participating message and / or at least one forwarding information contained in the participating message is removed from the participating message. The modification and / or removal can then preferably take place in accordance with a local configuration of the at least one network node.

The local configuration can in particular be an existing local configuration for forwarding incoming data packets, preferably depending on the header of these. In other words, the locally configured forwarding is then used for the forwarding of stream announcement messages and the possible addition of new information for subsequent nodes. The participation message is, in particular, a message which is sent and returned in a manner known per se from at least one terminal device which wishes to participate in a stream announced via an announcement message, for example a talker or listener advertise (stream subscriber) is transmitted to the stream initiator. In particular, it can be a so-called taler or listener join or ready.

In an MPLS (sub) network in particular, a (MPLS) label can be set at the entrance to the network, for example, which determines the path through the network. The MPLS network nodes then preferably check in each case what is to be done with the label, the procedure preferably being carried out in accordance with a local configuration of the respective node. Options for dealing with such a label are, for example, that it has to be removed, changed or passed on unchanged. Provision can also be made for a new label to be added or added, in particular a new label for subsequent network nodes. Purely by way of example, it should be mentioned that, following the establishment of MPLS paths, a local rule at a node is that packets with "Label 3" with "Label 4" are forwarded or removed from packets "Label 3" and the next label is inserted. Such a local rule leads to a change in the respective packet header in the forwarding and is dependent on the locally configured forwarding for the current headers of the packets.

In a preferred development of the method according to the invention, it is provided that

Announcement messages and / or stream participation messages are processed, for example at least one network node, in particular an MPLS network node at least inserts a label in such a message and / or at least changes a label in such a message and / or at least one label from such a message Message is removed. Like the respective knot If the stream announcement messages and / or stream participation messages are treated or processed, it will then preferably be based on a local configuration of the network node. The local configuration of (MPLS) network nodes can be obtained, for example, by a central network management and / or local settings by a terminal.

The decision as to which forwarding information (s), for example which label (s), is added to a stream announcement message and / or a stream participation message, and / or the decision as to whether any forwarding information contained is removed , and / or whether and if so, how a forwarding information contained therein is changed can also be based, for example, on or taking into account a stream address contained in the announcement or participation message, such as StreamDA, and / or based on or taking into account of the receiving sport of the respective message.

Reservation and subsequent protected transmission of stream data can be done flexibly in a wide variety of networks and using any network components.

In particular, in the event that the forwarding information is provided in a separate forwarding information message, it can be provided that at least one, preferably each network node that has received a separate forwarding information message and then does so receives an announcement message for a stream, a stream identifier contained therein and, based on the stream identifier, determines the forwarding information for this announcement message from the separate forwarding information message. Since the connection information or the assignment to the stream does not initially exist when the forwarding information is provided separately from the announcement message, a mapping of the stream and forwarding Useful information. If forwarding information is not (or not only) transmitted directly in the announcement message, but (possibly also) separately from it, the network nodes in a network can, for example, each be provided with a list or table which contains a plurality of streams -IDs each assign the associated forwarding information. This table / list can then be used to determine which forwarding information belongs to a received announcement message and is to be consulted for it. This table can also be configured in the respective network components during the configuration of the network. This can be done, for example, when connecting two networks by specifically configuring two connection network components, such as connection nodes, at only these two locations in the network by configuring the table / list with stream IDs and the associated forwarding information.

It is also expediently provided that at least one, preferably each, network node determines on the basis of the forwarding information, via which or which transmission port of the network node the announcement message is forwarded.

In a further very particularly advantageous embodiment of the method according to the invention, the determination is made about which or which transmission port of the network node the announcement message is to be forwarded, separately from evaluation and / or processing of the announcement message and / or storage of information from it, in particular for receipt a resource reservation at at least one port, in particular at the receiving port and at least one sending port for the announcement message. It has proven to be ideal if the determination, via which or which transmission ports the forwarding is to take place, is made in a separate layer from the reservation processes. Then in particular a forwarding layer and a reservation layer separate therefrom are provided. The two separated layers can be layers in the sense of the OSI reference model, in particular according to IS07498-1.

This preferred embodiment according to the invention is characterized by the use of the forwarding information, before given in the form of TLVs, in a layer separated from the layer of the reservation protocol, for example SRP or an extended version of SRP, and one independent of the forwarding information Reservation at the physical send ports of the stream. This makes it possible for the resources required for secure data transmission to be reserved in any network component, regardless of the transport layer used.

The task of the (resource) reservation layer is then, in particular, the internal configuration of a shaper and the reservation of storage space within the real-time queue for the secure transmission of the stream data between the port in the direction of the stream initiator and one or more ports in the direction of Stream participants who log in or have logged in to an announced stream, for example by sending a listener / talker join or ready message. Such ports are given in particular by ports with a valid listener / talker reservation. As a rule, the available bandwidth on the link must also be checked

(Feedback of the possibility of a transmission) and a list of the bandwidth already used is kept internally by the network node. This usually requires the source port, the destination port (s) and the amount of data (contains the frame size dependent on the forwarding). After successful reservation, especially after registration by one or more listeners or talkers, the forwarding of the stream data can be activated. The task of path determination, in particular of the forwarding layer separate from a reservation layer, is to determine possible send ports for forwarding the announcement message (stream description). SRP, for example, distributes incoming announcement messages via these ports - after successfully checking whether the stream could be transmitted with the currently existing reservation, or together with an error code for insufficient resources. For forwarding, the forwarding information additionally inserted into the announcement message and / or provided separately from it can be used for forwarding.

This also makes it possible for the

Change stream addresses. A use of, for example, MPLS labels or IP frames for forwarding the

Stream data becomes possible.

For MPLS in particular, the forwarding information can also be given, for example, by a stack of labels or include one on the basis of which the forwarding is defined in an MPLS network. According to the applicant, several IPv4 and Layer 2 information can also be used for future networks with network virtualization techniques. This makes it possible, for example, to transmit Layer 2 streams in an IP network via a connection with IP headers. This type of encapsulation can exist multiple times in larger networks and enables a multi-stage encapsulation of the streams and the transmitted stream data.

It can further preferably be provided that after receiving an announcement message, the send port or ports are determined in a first step, and in a second step the evaluation and / or processing of the announcement message and / or the storage of information therefrom, in particular to obtain a Reservation of resources at at least one port, in particular at the receiving port and at least one sending port for the announcement message, follows. In particular, it can be checked whether the transmission port (s) meet one or more requirements according to the announcement message.

It should be noted that the evaluation and / or processing of the announcement message and / or the storage of information from it, in particular to obtain a resource reservation on at least one port, is to be understood in particular to mean the previously known, possibly standardized processes, which are also in accordance with the known ones Reservation protocols, such as SRP, occur when an announcement message for a stream is received from a network node, such as a bridge. This preferably also applies to checking whether the send port (s) meet one or more requirements according to the announcement message.

In a further embodiment of the method according to the invention, it can further be provided that the overhead caused by a transmission of stream data packets via the respective send port or ports is determined and in particular when evaluating and / or processing the announcement message and / or storing information therefrom for receiving a resource reservation on at least one port is taken into account. The determination of the overhead caused is particularly preferably carried out in a forwarding layer which is separate from the reservation layer which handles the reservation process.

In SRP, for example, the source port (receiving sport of each announcement message) is stored internally, that is, in the respective network node, so that a registration by one

Stream participants can be carried back to the stream initiator. If the send option is successful, the announcement message is forwarded to the possible destination ports / send ports. For this purpose, a check at each potential target port with the overhead caused by the transmission to this port is very useful. In the TSpec of an announcement message, for example, is according to Standard only contain the data volume of the application - however, the frame size required for the data transmission results from the respective transport method, which is described in one embodiment of the present invention by the TLVs contained in the announcement message, which

Stream transmission at the target port. When receiving a registration of a stream participant, all resources are checked and reserved at the port of their registration, in particular the listener / talker join or readys. After successful reservation, the forwarding is activated. This is activated, for example, by SRP in the forwarding.

Layer 2 addresses usually change in a router. Due to the preferred separation of wide line and reservation, resources can be reserved as for pure Layer 2 streams. When forwarding, the IP information can be used to forward the announcement message to the correct port. The local handling of the data in the queue is then again preferably defined by the configured mechanisms, for example IEEE802.1Q (if it is Ethernet). For other transmission technologies, local configurations can be carried out analogously in order to enable protected data transmission.

It is also possible to use MPLS information for forwarding. With MPLS a stack of labels is defined, which can be expanded, shortened or only read in every MPLS router. This may change the number of TLVs used and the overhead for later transmission of the stream data, since the labels are dynamically removed or added in the MPLS network. The addition, removal or modification of MPLS Labein is defined in the respective MPLS node and is comparable to the previously separated information of the forwarding in the node in the form of a list or table. The labels are always in each frame along the path defined by MPLS are transmitted in the MPLS network and represent an additional overhead.

It should be noted that in virtualized networks, networks are often implemented as a purely virtual function using overlay techniques. A layer 2 connection (ethernet) can be formed via an IP connection on physical Ethernet. This is also made possible by the preferred use of TLVs.

It is also expedient for the implementation of the method according to the invention that the reservation is carried out in every network node along the data transmission. Appropriately, an activation of the forwarding and a configuration of shapers takes place at each port, as is known per se from the prior art.

When the announcement message is forwarded, for example via SRP, the updated forwarding information is preferably added in the form of TLVs in the stream announcement message from the stream initiator. The structure of the stream announcement message has been expanded in IEEE 802.1 Qcc to the terminal by TLVs. In an analogous manner, within the scope of the present invention, the announcement messages of an extended stream reservation protocol, which are exchanged between the network components, can be expanded with TLVs.

The invention also relates to a network node with a plurality of ports which is designed and / or set up to carry out the method according to one of the preceding claims.

The network node according to the invention has proven to be particularly suitable for carrying out the method according to the invention. The network node according to the invention is preferably designed and / or set up to receive an announcement message for a stream and to insert at least one forwarding information therein and / or to at least one forwarding information contained in the announcement message, in particular at least one Label, to change and / or to remove at least one forwarding information contained in the announcement message, in particular at least one label.

It can also be provided that the network node according to the invention is designed and / or set up to

to insert an identifier assigned to the network node, in particular its bridge ID, as forwarding information into the announcement message, the bridge ID preferably being a system ID assigned to the network node in accordance with a routing protocol, such as the ISIS protocol . This is particularly preferred if the network node according to the invention takes on the role of an edge node via which one or more terminals are connected to an existing network.

Another embodiment is characterized in that the network node is designed and / or set up to receive a participation message for a stream and at least insert forwarding information therein and / or to at least forwarding information contained in the participation message , in particular at least one label, to change and / or to remove at least one forwarding information contained in the participation message, in particular at least one label.

Alternatively or additionally, the network node according to the invention can be characterized in a further advantageous embodiment in that it is designed and / or set up to receive an announcement message and forwarding information for it, the forwarding information both contained in the announcement message can be transmitted as well as separately, and to determine on the basis of the forwarding information, via which or which of its ports it has to forward the announcement message (send or destination ports), this determination being separate from an evaluation and / or Processing of the announcement message and / or storing information from it, in particular to obtain a resource reservation at at least one port, takes place, in particular in a separate layer of the network node.

In other words, the network node according to the invention is preferably characterized in that it has a forwarding layer and a reservation layer, which is separate therefrom. The reservation layer can be used to process a reservation protocol, such as SRP or an extended version of SRP. The forwarding layer takes over - separately from the reservation - preferably the determination of possible send ports for the stream announcement message.

In a further development, the network node according to the invention can be designed and / or set up to carry out the preferred embodiment described above in connection with the method according to the invention.

The network in which the method according to the invention is carried out or in which the network node according to the invention takes part is expediently an AVB or TSN network or an AVB or TSN-capable network. In particular, the network nodes of such a network are AVB or TSN capable, support one or more AVB or TSN standards, for example the reservation of resources for forwarding with guaranteed latency. This applies accordingly preferably to the network node according to the invention. Another object of the invention is a computer program which comprises program code means for carrying out the steps of the method according to the invention.

Finally, the subject of the invention is a computer-readable medium which comprises instructions which, when executed on at least one computer, cause the at least one computer to carry out the steps of the method according to the invention.

The computer-readable medium can be, for example, a CD-ROM or DVD or a USB or flash memory. It should be noted that a computer-readable medium should not only be understood as a physical medium, but also, for example, in the form of a data stream and / or a signal which represents a data stream.

Further features and advantages of the present invention who the clearing from the following description of embodiments of the invention with reference to the accompanying drawings. It shows in a purely schematic representation

1 shows an industrial network of an automation system;

FIG. 2 shows the network from FIG. 1 with the tree from the point of view of the node B5;

FIG. 3 the network from FIG. 1 with the tree from the point of view of the node B6;

FIG. 4 shows another industrial network to which a large number of end devices are connected;

Figures 5-13 another industrial network, each de Figure shows three trees from the perspective of one of the nine network nodes;

FIG. 14 shows an exemplary embodiment of a network node according to the invention with a separate forwarding and reservation layer in an enlarged representation;

FIG. 15 the network node from FIG. 14, the overhead calculated in the forwarding layer being transferred to the reservation layer;

FIG. 16 shows the layers of the network node from FIGS. 14 and 15;

FIG. 17 the network node from FIGS. 14 and 15, which receives and forwards a listener join; and

18 shows three different exemplary embodiments of network nodes according to the invention.

FIG. 1 shows a purely schematic partial representation of an industrial network via which data packets can be exchanged between terminals of an industrial automation system not shown in the figures. In the exemplary embodiment described, the network comprises six network nodes in the form of bridges B1-B6, which are connected to one another via cable 1 - also referred to as links. It should be noted that in FIG. 1 both the solid and the dashed lines each represent a cable or a link 1 between two nodes B1-B6.

One or more terminal devices can be connected to each of the six nodes B1-B6 shown, or can log on in order to enable data communication.

In Figure 1, purely as an example, only two terminals are Darge. Of these, one forms a talker T, from which data packets periodically to the other, represent a listener L. lend end device to be transmitted. A secure, periodic transmission of data packets from the talker T to the listener L as a stream should take place in particular with guaranteed latency.

According to the AVB or TSN, it is known that for the establishment of a corresponding stream, an announcement message 2 is issued by the talker T and distributed in the network.

It should be noted that a scenario with a talker T and a listener L is shown as an example in the figures. It goes without saying that data packets can also be sent periodically from one talker T to several listeners L. It is also possible for a plurality of talkers T to send one listener L. This scenario, which can also be referred to as a “multiple talker per stream”, is described in the European patent application with the file number 18 15 43 19.0, which also goes back to the applicant. An announcement message 2 can accordingly originate from both a talker T and a listener L. It can are in particular a talker or listener advertise. In both possible 1: N scenarios, the announcement message 2 is expediently issued by the individual, that is to say the one talker T or the one listener L.

One or more stream participants (Talker T or Listener L) can then log on to a stream announced by a Talker T (or Listener L). For this purpose, a participation message is usually sent by the stream participant (s). In the prior art, these participation messages are also referred to as joins or readies. The associated processes, including the reservation of network resources required for the secure transmission of data at participating network nodes B1-B6 between the talker (s) T and the listener (s) L of a stream are sufficiently well known from the prior art (see, inter alia, IEEE 802. IQ and other standards), which is why they are no longer described in detail here. A reservation protocol, such as SRP, is preferably used to process the reservation. Such a protocol relies on path information from the network to set up the stream. By distributing the announcement message 2 along a path defined by the network (in the case of AVB: all active ports) and the subsequent registration of a listener L, the path of the data and the necessary configuration of the shapers on the connection between Talker T and Listener L are set up .

With AVB, it is not possible to select the transmission path or tree in the network when making a reservation. With RSTP (al so the Rapid Spanning Tree Protocol) only one free tree is active in the network. AVB has limited itself to RSTP because it focused on home networks as the target market. In FIG. 1, a loop-free tree S is drawn in for the node B1 as a root with a bold line. This is to be used for all streams between end devices that could be connected to any bridges B1-B6. There is an active path P for the two terminals T, L shown in FIG. 1, which extends over B5, B3, B1 and B2.

There is also a need for improved path selection for streams - also with AVB and TSN. By using the shortest paths in the network, shorter transmission times can be realized. Mechanisms are therefore defined in IEEE802.1Qca to enter a forwarding for the stream address (Stream DA) in the network. The stream reservation already takes into account restrictions in the distribution of the announcement messages 2, which is why control of the forwarding is made possible by restricting the forwarding of the stream data addresses. The corresponding restriction is that some ports of the network nodes between stream initiator T and stream subscriber L are hidden. In order to enable protected transmission with at the same time low configuration effort and ideal use of the available network topology (in particular the shortest paths), it is provided according to the present invention that to reserve a transmission path P for the transmission of stream data packets from the the talker T forming the stream initiator to the one stream subscriber, the listener L in the network with a plurality of network nodes B1-B6 represents forwarding information 3 (see, for example, FIG. 2), which is a predetermined transmission path P or transmission tree S for distribution the announcement message 2 is represented or from which one can be derived, is provided by inserting it into the announcement message (here the talker advertise) 2 by the stream initiator T and / or by at least one network node B1-B6.

As an alternative or in addition to the fact that forwarding information is inserted into the announcement message 2, this can also be done in at least one of the network nodes B1-B6 in a forwarding information message separate from the announcement message 2 together with one in the announcement message associated with the stream to be set up 2 contained stream connec- tors are transmitted or have been previously.

In the context of the preferred exemplary embodiment described here, the identifier 3 of the talker-side edge node is used as forwarding information, that is to say the network node B5 which is closest to the talker T. Of course, it is possible that, as an alternative to an edge node expert 3, other forwarding information is used.

The edge node identifier 3 in the present case is the bridge ID of the edge node B5, which is an IS-IS system ID and is formed from the MAC address of the edge node B5 (according to ISO 10589: 2002). This has previously been assigned to the edge node B5 by the IS-IS routing protocol. The remaining nodes B1-B4, B6 are also bridge IDs in the form of IS-IS system IDs have been assigned, by means of which each node B1-B6 can be uniquely identified and distinguished from the remaining nodes B1-B6. At least one tree S _Bn (with n = 1, 2, ..., 6) has been calculated as root for each node B1-B6 and is known in the network by the IS-IS protocol.

Known in this context means that not all nodes B1-B6 know the course of a complete tree / path S _Bn , but only each node B1-B6 knows how it receives data or data received from a neighboring node B1-B6 on a receiving port To forward messages sent by one of the further nodes B1-B6, specifically via which or which transmission ports. From a global perspective, these neighborhood relationships result in the complete paths / trees or they can be derived. According to the IS-IS protocol, the network topology is available through the collected neighborhood information and the forwarding information can be calculated for all bridge IDs of the network and the result of the local forwarding is stored in the node. Based on the network topology, the forwarding path is calculated according to the algorithms in IEEE 802.1 Qca, or the local forwarding is calculated depending on the respective source of the data (nodes B1-B6) by determining the valid receiving port and the forwarding ports to be used.

In the present case, the edge node ID 3 is inserted by the edge node B5 as forwarding information into the announcement message 2. This procedure is indicated schematically in FIG. 2 for the edge node B5 and in FIG. 3 for the edge node B6. From the talker T, the announcement message 2 is transmitted in a manner known per se to the first node from the point of view of the talker 2, that is to say its edge node B5, and this inserts the bridge ID 3 in the form of its IS-IS system ID into the announcement message 2 a. On the basis of this forwarding information 3, the announcement message 2 can then be transmitted in a targeted manner in an optimized way in the network, specifically along the tree S _B s of the edge node B5. It should be noted that the announcement message 2 represents or comprises a stream description for the stream. This contains the parameters or information known from the prior art, specifically a unique ID (StreamID) assigned to the stream, a designated stream address (StreamDA), the specification of a virtual network (VLAN), a traffic specification (TSpec ) with the specification of a maximum frame size and a maximum number of frames per interval, an accumulated latency and possibly, in particular if required resources are not available, required error information. The latter two details change in a manner known per se when passed on in the network. In connection with the previously known standardized content of stream announcement messages, reference is also made to IEEE802.1Q.

The forwarding information 3 added to the announcement message 2 in the form of the edge node ID represents additional information contained in the parameters / information provided in accordance with the prior art or standard.

Each network node B1-B6, which receives an announcement message 2 with the bridge ID 3 of the peripheral node B5, uses this forwarding information, that is to say the bridge ID 3, to determine which or which send ports the announcement message 2 is forwarded to.

The advantageous use of the forwarding information, in particular in the form of the bridge ID 3, makes it possible for a separate network path or

Tree must be calculated and set up, but the already existing known path P or tree S _B s can be used here from the point of view of the edge node B5. By using the forwarding paths of nodes B1-B6 in the network for streams, there is no need for a calculation for each stream. The trees / paths in the network can be calculated independently of the stream reservation. With a topology change, not all stream paths have to be new calculated and the forwarding routes for the stream addresses entered before the reservation can be updated. It is sufficient to recalculate the network paths using the new topology for nodes B1-B6 and to update the reservation, which can be done automatically using the SRP stream reservation protocol in accordance with IEEE 802.1 Q.

Due to the use of the tree S _B s or path P of a bridge B5, the stream registration and the tree S _B s or path P are also independent of the talker T, which triggers the stream registration. Because the bridge ID 3 is added to the announcement message 2 by the edge node B5, the following nodes (in the scenario according to FIG. 2 the node B3) know which path P is to be used. The tree S _B s from the perspective of the talker T is indicated by arrows in FIG. 2. The path P from the talker T to the listener L goes here via B5, B3 and B2.

FIG. 3 shows which tree S _B6 would or would result if, alternatively, a talker T located at node B5 announces a stream, this is done by another talker T located at node B6, and the bridge ID 3 of the node B6 then forming the edge node is inserted. The path from talker T to listener L is via B6, B4 and B2.

In the exemplary embodiment described here, the edge bridge ID 3 is added in the type length value format (TLV format) from the edge node B5 to the announcement message 2. Subsequent to the addition, this represents an additional TLV in the announcement message 2 compared to the standard content of a talker advert. In particular, the TLV structure makes it possible to also use any combination of Ethernet, IP, MPLS and / or to choose or support other transport variants and / or tunnel techniques. To further illustrate the distribution of announcement messages from stream initiators and participation messages from stream participants, FIG. 4 shows a network with a plurality of connected terminals E1-E8 and a scenario with two streams. As can be seen, the network also comprises six nodes B1-B6, which, however, are connected somewhat differently than the nodes B1-B6 from FIG. 1. Here, at least one terminal E1-E8 is connected to each node B1-B6.

A first stream is announced from the terminal El, for which the device El represents the talker T. As can be seen, El is located at the node Bl, which means B1 represents the marginal node for El. The announcement for the first stream is made by issuing a first announcement message in the form of a first talker advert Al. The tree S _Bi for the node Bl as an edge node is indicated in FIG. 4 by solid lines with the greatest line width.

A second stream is also announced by the terminal E6 by issuing a second announcement message in the form of a second talker advert A2. The tree S _B6 for the node B6 as an edge node is indicated in FIG. 4 by solid lines of medium thickness. Solid lines of the smallest line width represent in Figure 4 the connections of the terminals E1-E8 to the respective edge nodes B1-B6.

The spread of the announcement message A1 for the first stream, for which El represents the talker T, is indicated in FIG. 4 by arrows with a solid line. The announcement message A2 for the second stream, for which the E6 represents the talker T, however, is represented by arrows with a dashed line.

In the scenario shown in FIG. 4, the terminals E3 and E5 register as listeners on the first stream, where they each issue a listener join J1. The way the listener joint J1 from the respective listener E3, E5 back to the talker El is shown in FIG. 4 by arrows with a dash-dotted line.

Furthermore, the terminals E2 and E3 register as listeners on the second stream, which is done by issuing listener joins J2. It should be noted that the terminal E3 represents a listener L of both streams.

According to the exemplary embodiment described here, the announcement message A1 for the first stream, which is sent from E1, is added by the node B1 forming the edge node for this terminal to its bridge ID 3. The distribution of the announcement message A1 takes place in the network on the basis of this forwarding information along the tree S _Bi for or from the edge node Bl. In an analogous manner, in the announcement message A2, which is issued by the terminal E6, the edge node B5 whose bridge ID 3 inserted in the announcement message A2 and the distribution of this in the network takes place along the tree S _B s of the node B5.

In other words, in both cases, that is to say for both streams, the announcement message A1, A2 issued by the stream initiator E1, E6 as forwarding information is each added to an identifier which is assigned to that network node B1, B5 that corresponds to the respective one Stream initiator El, E6 is closest and forms the edge node for this El, E6 and is distributed along the edge node tree S _BI , 5

In the context of the preferred exemplary embodiment described here, not only is a tree S _Bn set up for each node Bl-B6, but a set of predefined trees Sl _Bn -S3 _{Bn is} set up in the network. Each set of trees for each network node B1-B6 comprises at least one predefined tree S1 _Bn -S3 _Bri , via which the respective network node B1-B6 can reach as a root every other network node B1-B6 in the network. Specifically, three sets of predefined trees Sl _Bn -S3 _{Bri are} set up, one set for each of the network nodes B1-B6 includes a tree Sl _Bn as a root, which was calculated in accordance with SPB and the second and third sets each include a tree S2 _Bn , S3 _Bn for each of the network nodes B1-B6, the maximum of which together represent the maximum disjoint trees according to the MRI algorithm.

For the sake of clarity, FIG. 4 does not show all these trees S1 _Bn -S3 _Bri , but only one tree S _Bi , s for the edge nodes B1 and B5 can be seen. However, the principle can be seen in FIGS. 5 to 13 for a slightly different network topology, specifically the arrangement shown there with nine network nodes B1-B9. It should be noted that the trees Sl _{Bn of} the first set therein are represented by arrows with a solid line, and the trees of the second and third sets S2 _Bn , S3 _{Bn are represented} by arrows with a dashed or dash-dotted line. Furthermore, it should be noted that each of FIGS. 5 to 13 each shows the three trees S1 _Bn -S3 _Bri from the perspective of one of the nodes B1-B9, specifically FIG. 5 shows the three trees S1 _Bi , S2 _BI , S3 _BI for B1, the figure 6 the three trees S1 _B2 , S2 _B2 , S3 _B2 for B2, the figure 7 for B3, ... and the figure 13 for B9. The respective node B1-B9 forming the roots of the three trees S1 _Bn -S3 _Bri is shown with a bold outline.

The selection of which set of trees, or which tree Sl _Bn - S3 _{Bn should be used} from the point of view of the respective edge node B1-B9, which algorithm should be used accordingly, can be made by specifying a VLAN or a VLAN ID in the announcement message 2. A VLAN ID is assigned to each of the three sets of trees S1 _Bn -S3 _Bri , the trees S1 _Bn - the VLAN ID1, the trees S2 _Bn the VLAN ID2 and the trees S3 _Bn the VLAN ID3.

As an alternative or in addition to the use of VLAN IDs, it is also possible to insert an availability class, which then, like the current stream class, deals with the selection of a tree, for example, Shortest Path in particular with automatic switchover (in FIGS. 5 to 13 S1 _Bn ) or MRI for two redundant trees (in FIGS. 5 to 13 S2 _Bn and S3 _Bn )

Together with the availability, in particular whether CB is activated in the network or not (see also WO 2019/001718 A1 attributable to the applicant), the configuration of the bridges B1-B9 and the forwarding of the announcement messages 2 with the stream messages can then Information about the next bridge B1-B9 in the network.

It should be noted that the scenario according to FIGS. 5 to 13 also applies in turn to the fact that one or more terminal devices can be connected to each node B1-B9. Furthermore, it also applies here that each network node B1-B9, which receives an announcement message 2, A1, A2 with the bridge ID 3 of an edge node B1-B9, determines on the basis of this forwarding information via which or which transmission ports the announcement message 2 is forwarded .

In the context of the particularly preferred exemplary embodiment described up to here, the determination is made via which or which send port 7 of the respective node B1-B9 the announcement message 2, A1, A2 is to be forwarded, separately from an evaluation and / or processing of the announcement message 2, A1, A2 and / or a storage of information therefrom in order to obtain a resource reservation at the receiving port 6 and the respective sending port 7 of network nodes B1-B9 involved.

In the present case, the determination is made via which transmission ports of the respective node B1-B9 an announcement message 2, Al, A2 is to be forwarded in a forwarding layer 4 (see FIGS. 15 and 16) by a reservation layer 5, which the Processing of the reservation protocol, in the present case an extended version of SRP is used. The determination of the send ports takes place before the reservation is processed, in particular in one of these upstream steps. This is again illustrated purely schematically and purely by way of example in FIGS. 14 to 16. A network node B1-B9 is shown by way of example in FIGS. 14 and 15, two of the ports 6, 7 of the node B1-B9 being recognizable here. FIG. 14 shows that the announcement message 2, A1, A2 arrives at a reception port 6 which is connected to the talker T, on the basis of the forwarding contained in the announcement message 2, A1, A2 in the form of the edge node ID 3. Information is made about which or which destination or send ports 7 the announcement message 2, A1, A2 is to be forwarded to. The determination of the transmission port (s) 7 takes place in the forwarding layer 4.

Separately, specifically in the reservation layer 5, the reservation 8 for the physical ports 6, 7 is processed or carried out. Here, the announcement message is preferably evaluated and / or further processed and / or further information from it is stored, preferably as previously known in accordance with the prior art.

It should be noted that only two ports 6, 7 are shown by way of example in FIGS. 14 and 15. It is understood that a network node B1-B9 can have, in a manner known per se, more than two ports 6, 7 and the processes for one or more further possible send ports 7 for an announcement message 2, A1, A2 are alike.

In FIG. 16, the forwarding layer 4, the reservation layer 5 and the resource reservation 8 for the physical ports 6, 7 are again shown schematically as block elements. The reservation protocol processed in the reservation layer 5 supplies the source port 6 and the forwarding information 3 to the forwarding layer 4 and based on this determines the destination port (s) 7 as well as that with a transmission of stream data packets via or the respective transmission port 7 caused overhead 9. Send port (s) 7 and overhead 9 are transferred to the reservation layer 5, which is indicated by arrows in FIG. 16.

The reservation layer 5 or the one processed therein

The stream reservation protocol uses the information on the source port 6 for the resource reservation 8. In addition, the forwarding information 3 in the advertise message 2 and the local list or table with forwarding information make the forwarding via the destination ports 7 possible certainly. Depending on the respective forwarding, a new overhead 9 results. The overhead 9 is communicated to the reservation layer 5 so that the resources can be reserved accordingly for the new packet size. The resource reservation is based on the possibilities 10 of the physical ports.

If the necessary resources are available, the forwarding can be activated, which is indicated purely schematically in FIG. 16 by block elements provided with the reference number 11.

As described above, one or more listeners L generally register by issuing a listener join message J1, J2. This is not shown in FIGS. 14 to 16, but can be seen in FIG. 17, which in comparison to FIG. 14 also contains additional arrows which indicate the direction from the listener L to the talker T.

The fact that the forwarding information is provided in the form of TLVs, in particular in the announcement messages 2, A1, A2, in combination with a separation of the reservation and the forwarding, as provided in the exemplary embodiment described here, offers a particularly high degree of flexibility . In particular, a stream reservation in any network components B1-B9 is possible, regardless of the forwarding. This is also in Figure 18 illustrated. Specifically, a network node in the form of a bridge B1-B9 is shown on the left, in the middle a network node in the form of an IP router 12 and on the right in the form of an MPLS router 13. Under the respective network component B1- B9, 12, 13 is pure schematically the forwarding information 3, which is an Ethernet forwarding information 3 on the left, an IP forwarding information 3 in the middle and an MPLS forwarding information 3 on the right, and separately from that actually amount of data 8 indicated.

Purely by way of example, it should be mentioned that in a network, as shown in FIG. 5, different technologies are provided, for example in the three areas which comprise an upper area with nodes B4 and B5, a middle area with nodes B1 , B2, B3 and B6 and a lower area with the nodes B7 to B9. It is often the case that a “transport network” is equipped with one or more other technologies (in particular several levels) and at the ends of which Ethernet networks are connected. In FIG. 5, for example, the middle area with B1, B2 could be used , B3 and B6, such a transport network can be formed with a different technology, for example IP, whereas the upper and lower areas represent Ethernet networks or Ethernet subnetworks. Even in such a network with a heterogeneous technology structure (IP and Ethernet) the method according to the invention can be advantageously used.

For MPLS networks in particular, forwarding information 3 can also be inserted or removed from a stream announcement message 2, A1, A2 in the form of labels, in particular MPLS labels, preferably by at least one MPLS Network node 13. It is also possible for at least one forwarding information 3, which is contained in a received announcement message 2, A1, A2, in particular at least one label, to be changed. In an MPLS (sub) network, for example, at the entrance a (MPLS) label of the network is determined, which determines the path through the network. Then the MPLS network nodes 13 can check in each case what is to be done with the label, the preferred procedure being according to a local configuration of the respective node. Options for dealing with such a label are, for example, that it should be removed, changed or passed on unchanged. Provision can also be made for a new label to be added, in particular a new label for MPLS network nodes 13 still to follow. Furthermore, in particular in an MPLS network, provision can also be made for adding and / or removing and / or changing of at least one forwarding information, at least one or a label into or out of a stream participation message J1, J2. This is done in particular by at least one MPLS network node 13.

All network nodes B1-B9, 12, 13 (see in particular FIGS. 14, 15, 17 and 18) are designed and set up for carrying out the described method according to the invention and represent exemplary embodiments of network nodes according to the invention. They each comprise a forwarding layer 4 and a reservation layer 5 separate therefrom.

Although the invention has been illustrated and described in more detail by the preferred exemplary embodiment, the invention is not restricted by the disclosed examples and other variations can be derived therefrom by a person skilled in the art without departing from the scope of the invention.

For example, even if a combination of adding forwarding information, in particular in the form of a marginal node ID, and treating forwarding and reservation separately in separate layers was described above, these two aspects are self-evident. can also be realized without the other aspect.

Claims

Claims 1. A method for reserving at least one transmission path (P) for the transmission of data packets from a stream initiator (T) to at least one stream subscriber (L) in an industrial network in particular, which has a plurality of network nodes (B1-B9, 12 , 13), the stream initiator (T) issuing an announcement message (2, A1, A2) for a stream, characterized in that at least one forwarding information (3) which contains at least one predetermined transmission path (P) or transmission tree (S _Bn) for the distribution of the announcement message (2, Al, A2) is represented or from which at least one of these can be derived, is provided by the stream initiator (T) and / or at least one network node (B1-B9, 12, 13) in the announcement message (2, Al, A2) is added, and / or by knowing at least one network (B1-B9, 12, 13) in a forwarding information message separate from the announcement message (2, A1, A2) together with one assigned to the stream , in the announcement message (2, Al, A2) contained stream identifier, in particular special stream ID, is transmitted. 2. The method according to claim 1, characterized in that an identifier is used as forwarding information (3), which is assigned to the network node (B1-B9, 12, 13) that is closest to the stream initiator (T) and marks the edge (B1-B9, 12, 13 ) for the stream initiator (T), in particular the bridge ID of the edge node (B1-B9, 12, 13) is provided as forwarding information (3), the bridge ID preferably being one of the edge node (Bl-B9, 12, 13) according to a routing protocol, assigned system ID. 3. The method according to claim 2, characterized in that the connoisseur from the edge node (B1-B9, 12, 13) is inserted into the announcement message (2, Al, A2). 4. The method according to any one of the preceding claims, characterized in that the at least one forwarding information (3) is made available by the stream initiator (T) and / or at least one network node (B1-B9, 12, 13) in type-length-value format in the announcement message ( 2, Al, A2) is inserted, and / or at least one forwarding information (3) is inserted, which is available due to a local configuration of the network node (B1-B9, 12, 13), and / or at least forwarding information comprises or is formed by at least one Ethernet forwarding information (3) and / or at least one IP forwarding information and / or at least one MPLS forwarding information. 5. The method according to any one of the preceding claims, characterized in that at least one network node (B1-B9, 12, 13) that receives an announcement message (2, Al, A2) with at least one forwarding information (3) contained therein, contains at least one in the announcement message (2, Al, A2) Forwarding information (3) is changed and / or at least one forwarding information (3) is removed from the announcement message (2, A1, A2), the change and / or the removal preferably being in accordance with a local configuration of the at least one Network node (B1-B9, 12, 13). 6. The method according to any one of the preceding claims, characterized in that the forwarding information (3) is provided in a separate forwarding information message, and at least one, preferably each network node (B1-B9, 12, 13) that receives a separate forwarding information message and then receives an announcement message (2, A1, A2) for a stream, one contained therein Stream connoisseur and the forwarding information (3) for this announcement message (2, A1, A2) from the separate forwarding information message is determined on the basis of the stream identifier. 7. The method according to any one of the preceding claims, characterized in that at least one, preferably each network node (B1-B9, 12, 13) determines on the basis of the forwarding information (3) via which or which transmission ports (7) of the network node (B1-B9, 12, 13) the announcement message (2 , Al, A2) is forwarded. 8. The method according to claim 7, characterized in that the determination of which or which send ports (7) of the network node (B1-B9, 12, 13) the announcement message (2, Al, A2) is to be forwarded, separately from an evaluation and / or processing of the announcement message (2, Al, A2 ) and / or storing information from this, in particular to obtain a resource reservation (8) at least one port (6, 7), in particular in a separate layer (4). 9. The method according to claim 8, characterized in that after receiving an announcement message (2, Al, A2), the transmission port (s) (7) are determined in a first step, and in a second step the evaluation and / or processing of the announcement message (2, Al, A2) and / or that Storage of information from this takes place in particular to obtain a resource reservation (8) at the respective transmission port (7). 10. The method according to any one of claims 6 to 9, characterized in that the overhead (9) caused by the transmission of stream data packets via the respective transmission port (7) and in particular when evaluating and / or processing the announcement message (2, A1, A2) and / or storing information from it to obtain a resource reservation (8) is taken into account at the respective sending port. 11. The method according to any one of the preceding claims, characterized in that at least one set of predefined trees (S 1 _Bn S3 _Bn ) is set up in the network, the or each set comprising at least one predefined tree (Sl _{Bn -} S3 _Bn ) for several, in particular each network node (B1-B9, 12, 13) , via which the respective network node (B1-B9, 12, 13) can reach several other, in particular every other network node (B1-B9, 12, 13) in the network as the root. 12. The method according to claim 11, characterized in that two or more sets of predefined trees (Sl _{Bn -} S3 _Bn ) are set up, each set comprising at least one predefined tree (Sl _Bn - S3 _Bn ) for several, in particular each network node (B1-B9, 12, 13) which the respective network node (B1-B9, 12, 13) can reach as a root several others, in particular every other network node (B1-B9, 12, 13) in the network, and the trees (S1 _Bn - S3 _Bn ) of different sentences were calculated by different algorithms, and in particular each set of predefined trees (Sl _{Bn -} S3 _Bn ) is assigned its own virtual network and / or its own VLAN ID, and preferably in addition to the forwarding information (3) the specification of a virtual network and / or a VLAN ID is provided, and / or a separate class is set up for each set of predefined trees, and preferably one of these classes is provided in addition to the forwarding information (3). 13. Network node (B1-B9, 12, 13) with a plurality of ports (6, 7) which is designed and / or set up to carry out the method according to one of the preceding claims. 14. network node (B1-B9, 12, 13) according to claim 13, characterized in that the network node (B1-B9, 12, 13) is designed and / or configured to receive an announcement message (2, Al, A2) for a stream and to insert at least forwarding information (3) into it and / or to to change at least one forwarding information (3), in particular at least one label, contained in the announcement message (2, Al, A2) and / or to change at least one forwarding information contained in the announcement message (2, Al, A2) (3), in particular at least one label. 15. network node (B1-B9, 12, 13) according to claim 14, characterized in that the network node (B1-B9, 12, 13) is designed and / or directed to an identifier assigned to the network node (B1-B9, 12, 13), in particular its bridge ID as forwarding information (3) into the Announcement message (2, Al, A2), the bridge ID preferably being a system ID assigned to the network node (B1-B9, 12, 13) in accordance with a routing protocol. 16. network node (B1-B9, 12, 13) according to one of claims 13 to 15, characterized in that the network node (B1-B9, 12, 13) is designed and / or set up to provide an announcement message (2, A1, A2) and forwarding information (3) for the latter, in particular in a type-length-value format, to receive, wherein the forwarding information (3) can both be contained in the announcement message (2, A1, A2) and can be transmitted separately, and to determine on the basis of the forwarding information (3) via which or which of its ports (6, 7), it has to forward the announcement message (2, Al, A2), where in this determination, separately from an evaluation and / or processing of the announcement message (2, A1, A2) and / or storage of information from it, in particular for obtaining a resource reservation (8) on at least one port (6, 7), in particular in a separate layer (4) of the network node (B1-B9, 12, 13). 17. Computer program comprising program code means for performing the method according to one of claims 1 to 12. 18. A computer readable medium comprising instructions which, when executed on at least one computer, cause the at least one computer to carry out the steps of the method according to one of claims 1 to 12.