INTERMEDIATE NETWORK IN A RING TOPOLOGY, AND METHOD FOR SETTING UP A NETWORK CONNECTION BETWEEN TWO NETWORK DOMAINS

Abstract

The invention relates to an intermediate network (1) in a ring topology for setting up a connection between two network domains (2a, 2b), having a first marginal node (6a) and a second marginal node (7a), which are marginal nodes in a first network domain (2a) and which are connected to one another by means of a first network connection (9a) within the first network domain (2a), a third marginal node (6b) and a fourth marginal node (7b), which are marginal nodes in a second network domain (2b) and which are connected to one another by means of a second network connection (9b) within the second network domain (2b), a first virtual network connection (4a), which connects the first (6a) and the third (6b) marginal nodes via an intermediate network (3), and a second virtual network connection (4b), which connects the second (7a) and the fourth (7b) marginal nodes via the intermediate network (3). The first network connection (9a), the second network connection (9b), the first virtual network connection (4a) and the second virtual network connection (4b) have a ring topology on which a connection redundancy protocol is implemented.

Description

The invention relates to an intermediate network in a ring topology, in particular in Ethernet/Layer 2 networks, and a method for setting up a failsafe network connection between two network domains.

PRIOR ART

Ethernet-network domains are frequently connected to one another via virtual Ethernet connections via an intermediate network, in order to facilitate network traffic between otherwise disjoint network domains. Particularly in industrial applications, for example in factory or process automation, the requirements for reliability and failsafe operation of such Ethernet links are very high, in order to guarantee the security of the networked components.

These Ethernet connections between network domains are often implemented via virtual private networks (VPN), for example a virtual private wire service (VPWS) or a specially preconfigured virtual local network (VLAN). These virtual private network connections are established via an intermediate network, the operation of which is often outside the sphere of influence of the network domain operator.

Firstly, it is now necessary for network traffic between two network domains connected via such an intermediate network with the help of virtual private network connections to be controlled in such a way that there are no unwanted data traffic loops; secondly, network security is to be established in the network traffic between the network domains, which may capture failures of network components in the intermediate network without being dependent upon the security measures of the intermediate network operator.

It is possible, for example, to integrate the two network domains into a joint network with cross-domain redundancy protocols such as RSTP, for example. However, this requires a high implementation cost. It also requires a large amount of planning in order to keep network-domain-internal network traffic inside the domain, in other words to prevent network traffic that is supposed to take place within a domain from being routed via the intermediate network, thus causing unnecessary costs and latencies.

Alternatively, proprietary protocols may be used that permit additional information to be exchanged via individual connections between the network domains and thus to respond to any faults that may occur. However, this solution lacks flexibility and requires a high level of adaptation to the existing infrastructure.

There is therefore a need for easier and effective solutions for connecting network domains with one another via an intermediate network.

ABSTRACT OF THE INVENTION

One concept of the invention is therefore to connect to one another two disjoint network domains, which are interconnected via an intermediate network and exchange network traffic via the intermediate network, in such a way that, in the event of a failure in any component in or at the edge of the intermediate network, the network connection between the network domains remains guaranteed. For this purpose redundant virtual network connection paths are established through the intermediate network, the management and control of which is independent of the network connection actually implemented by the intermediate network. This offers the advantage that the network traffic between two network domains can be protected by an intermediate network on the part of the network domains, without being dependent upon the security mechanisms in the intermediate network, which are often outside the sphere of influence of the network domain operators.

One embodiment of the present invention therefore consists of an intermediate network in a ring topology as claimed in claim 1 for setting up a connection between two network domains, having a first marginal node and a second marginal node, which are marginal nodes in a first network domain and which are connected to one another by means of a first network connection within the first network domain, a third marginal node and a fourth marginal node, which are marginal nodes in a second network domain and which are connected to one another via a second network connection within the second network domain, a first virtual network connection, which connects the first and the third marginal node via an intermediate network, and a second virtual network connection, which connects the second and the fourth marginal node via the intermediate network, the first network connection, the second network connection, the first virtual network connection and the second virtual network connection having a ring topology on which a connection redundancy protocol is implemented.

According to an advantageous embodiment the first network domain and/or the second network domain may be Ethernet/Layer 2 domains.

According to a further advantageous embodiment the first virtual network connection and the second virtual network connection may be virtual private network connections on Layer 2.

According to a further advantageous embodiment the connection redundancy protocol may be a Spanning Tree Protocol (STP), in particular a Rapid Spanning Tree Protocol (RSTP), a Media Redundancy Protocol (MRP), an Ethernet Ring Protection Protocol (ERP) or an Ethernet Automatic Protection Switching Protocol (EAPS). Such protocols are tested redundancy protocols, which can be easily integrated into the existing network domain infrastructure due to standardization.

According to a further advantageous embodiment a domain-internal connection redundancy protocol can be implemented in the first network domain and/or the second network domain respectively. By linking a connection redundancy protocol in the intermediate network and connection redundancy protocols in the network domains themselves, it is possible to respond to failures of virtual network connections in the intermediate network particularly efficiently and quickly.

The invention furthermore creates a method for setting up a network connection between two network domains as claimed in claim 6, with the following steps: establishment of a first virtual network connection between a first marginal node of a first network domain and a second marginal node of a second network domain via an intermediate network;

establishment of a second virtual network connection between a third marginal node of the first network domain connected with the first marginal node, and a fourth marginal node of the second network domain connected with the second marginal node via the intermediate network; andimplementation of a connection redundancy protocol in a ring topology formed by the first, the second, the third and the fourth marginal node.

This has the advantage that loops and ring closures within the ring topology of the marginal nodes may be avoided by the intermediate network.

The implementation of a connection redundancy protocol may advantageously comprise the following steps:

designation of a marginal node as a master node;blocking of the virtual network connection assigned to the designated master node for network traffic between the first and the second network domain; andrelease of the respective other virtual network connection for network traffic between the first and the second network domain.

In this way network-domain-internal data traffic can still take place within the respective network domain between the associated marginal nodes, without rerouting of data traffic via the virtual network connections being necessary. In particular, data traffic via the virtual network connections is restricted in this case only to the necessary network traffic between the two network domains.

The blocked virtual network connection may advantageously be released in the event of a fault occurring in the released virtual network connection. By means of this redundancy, it is possible to respond on the network domain side to a failure in a virtual network connection, regardless of the type and duration of the fault. The interrupted virtual network connection is therefore advantageously protected by the initially blocked and redundantly held virtual network connection.

Further modifications and variations will emerge from the features of the dependant claims.

BRIEF DESCRIPTION OF THE FIGURES

Various embodiments and developments of the invention will now be described in more detail with reference to the accompanying drawings, in which

FIG. 1 is a schematic diagram of a network domain architecture with an intermediate network according to an embodiment of the invention;

FIG. 2 is a schematic diagram of an implementation of a connection redundancy protocol in an intermediate network according to a further embodiment of the invention;

FIG. 3 is a schematic diagram of a network connection protection in the event of a connection fault in an intermediate network according to a further embodiment of the invention; and

FIG. 4 is a schematic diagram of a method for setting up a redundantly protected network connection between two network domains according to a further embodiment of the invention.

The described developments and further developments may be combined with one another as required, where appropriate. Further possible developments, further developments and implementations of the invention also include not explicitly mentioned combinations of features of the invention described above or below with regard to the exemplary embodiments.

The accompanying drawings are intended to provide a further understanding of the embodiments of the invention. They illustrate embodiments and, in combination with the description, serve to clarify the principles and concepts of the invention. Other embodiments and many of the advantages mentioned will emerge with regard to the drawings. The elements of the drawings are not necessarily drawn to scale. The same reference characters designate the same or similarly operating components.

DETAILED DESCRIPTION OF THE INVENTION

For the purposes of the present invention, network domains are self-contained networks with network nodes and network connections between the network nodes. These network domains have marginal nodes, which—in addition to connections to the network domains themselves—are able to provide connections from the network domains. Network domains may for example be production cells or control stations in a production plant or system, machine cells in factories, automation components in an automation plant or similar components provided with an internal network. Such network domains may be operated with a communication protocol on communication Layer 1 and/or 2 (physical layer or data link layer), for example an Ethernet protocol, e.g. in a PROFINET framework.

Connection redundancy protocols for the purposes of the present invention may be network protocols which deactivate redundant paths in networks in order to avoid unwanted network traffic ring closures and activate the deactivated paths in the event of network failures, in order to safeguard network traffic security in the network. Such connection redundancy protocols may for example be Spanning Tree Protocols (STP) such as the Rapid Spanning Tree Protocol (RSTP), the Media Redundancy Protocol (MRP), the Media Redundancy Real-Time Protocol (MRRT), the Ethernet Ring Protection Protocol (ERP), the Ethernet Automatic Protection Switching Protocol (EAPS), the High-Availability Seamless Redundancy Protocol (HSR) or the Parallel Redundancy Protocol (PRP). Other redundancy protocols on communication Layer 1 or 2 may of course also be used.

FIG. 1 shows a schematic diagram of a network domain architecture with an intermediate network 1. The intermediate network connects two disjoint network domains 2a and 2b. The network domains 2a and 2b each have a domain-internal network or core network 5a, 5b, which are connected with marginal nodes 6a, 6b or 7a, 7b via connections 8a, 8b. The marginal nodes 6a, 6b or 7a, 7b provide connection options outside the network domains 2a, 2b and are connected to one another respectively via marginal connections 9a, 9b. The marginal connections 9a, 9b may of course comprise other components which are not illustrated.

These marginal nodes 6a and 6b are connected via a first virtual network connection 4a via the intermediate network 3, and the marginal node 7a and 7b via a second virtual net-work connection 4b via the intermediate network 3. The virtual network connections 4a, 4b are each tunneled connections on the basis of virtual private network connections (VPN), i.e. the virtual network connections 4a, 4b are transparent with regard to possible interposed (not shown) network components of the intermediate network 3. The virtual network connections 4a, 4b are each independent of one another, i.e. node- and edge-disjoint. Since the intermediate network is based on virtual network connections 4a, 4b, it may also be described as a virtual intermediate network.

FIG. 2 shows a schematic diagram of an implementation of a connection redundancy protocol in the intermediate network 1. The marginal nodes 6a, 7a, 6b, 7b, together with the network connections, 9a, 9b and the virtual network connections form a ring topology for the intermediate network 1. A connection redundancy protocol is implemented in this ring topology, which prevents unwanted network traffic loops occurring. The connection redundancy protocol is designed initially to select one of the marginal nodes 6a, 7a, 6b, 7b as the so-called master node, i.e. as the node which manages and controls the forwarding of data via the available network connections 4a, 4b, 9a and 9b. By way of example, FIG. 2 shows the marginal node 6a selected as the master node; it is however of course also possible for one of the other marginal nodes 6b, 7a or 7b to be selected as the master node. In the example shown in FIG. 2 the marginal nodes 6b, 7a, 7b are slave nodes subordinate to the master node 6a, that are subject to control by the master node 6a.

Within the connection redundancy protocol, the master node 6a prevents unwanted network traffic loops from forming. For this purpose a ring closure via the network connections 4a, 4b, 9a and 9b is prevented, whereby for example one of the network connections, in FIG. 2 the network connection 4a, is blocked for normal network traffic. Thus provision may for made, for example, for a port 11 of the master node 6a, which is used for communication with the marginal node 6b, to be blocked. Provision is advantageously made for the blocked network connection to be one of the two virtual network connections 4a, 4b, so that network-domain-internal network traffic, for example network traffic within the network domain 2a, may also continue to be routed via the network connection 9a, without having to carry out a costly rerouting at high latency via the intermediate network 3.

Network traffic generated in the core network 5a of the network domain 2a and to be transferred to the core network 5b of the network domain 2b, is labeled by way of example with the reference character 10. This network traffic 10 may be routed initially to the master node 6a via the network connection 8a. At the master node 6a the network traffic is then not routed via the blocked virtual network connection 4a through the intermediate network 3, but is transferred to the marginal node 7a via the marginal connection 9a. From there the network traffic 10 is transferred via the prioritized virtual network connection 4b via the intermediate network 3 to the marginal node 7b, from where the network traffic 10 is transferred via the network connection 8b to the core network 8b.

FIG. 3 shows a schematic diagram of a network connection protection in the event of a connection fault 13 in an intermediate network 1.

The intermediate network 1 in FIG. 3 corresponds to the intermediate network 1 in FIG. 2, in which the marginal node 6a has been selected as the master node by way of example. FIG. 3 furthermore illustrates the case in which a fault 13 has occurred in the intermediate network 3. The fault 13 may for example be a node failure, a connection failure, a connection interruption, a power failure or other permanent or temporary fault, which prevents the smooth and faultless operation of network traffic transmission via the prioritized virtual network connection 4b. Provision may be made, for example, for the fault 13 to be detected by the absence and/or non-receipt of a predefined number of “hello” messages from marginal node 7b to marginal node 7a. As soon as a fault 13 has been detected in the virtual network connection 4b, countermeasures may be initiated within the connection redundancy protocol of the intermediate network 1.

For this purpose provision may be made for the master node 6a to release the hitherto blocked, redundant virtual network connection 4a for normal network traffic. This may be effected permanently or until such time as the fault 13 within the intermediate network 3 has been eliminated.

It is furthermore possible for the master node 6a itself to fail. In this case provision may be made within the connection redundancy protocol for one of the remaining marginal nodes 6b, 7a, 7b to be defined as the new master node, and for network traffic to be diverted via the remaining connections 8a, 4b, 8b and 9b to the exclusion of the failed marginal node 6a.

FIG. 4 shows a schematic diagram of a method 40 for setting up a redundantly protected network connection between two network domains, in particular between two network domains as shown in FIGS. 1 to 3. A marginal node is designated as the master node in a first step 41. In a second step 42, the virtual network connection which is assigned to the designated master node is blocked for network traffic between the first and the second network domain. In a third step the respective other virtual network connection is released for network traffic between the first and the second network domain.

Claims

1. An intermediate network (1) in a ring topology for setting up a connection between two network domains (2a, 2b), having: a first marginal node (6a) and a second marginal node (7a), which are marginal nodes in a first network domain (2a) and which are connected to one another via a first network connection (9a) within the first network domain (2a);a third marginal node (6b) and a fourth marginal node (7b), which are marginal nodes of a second network domain (2b) and which are connected to one another via a second network connection (9b) within the second network domain (2b);a first virtual network connection (4a), which connects the first (6a) and the third marginal node (6b) via an intermediate network (3); anda second virtual network connection (4b), which connects the second (7a) and the fourth marginal node (7b) via das intermediate network (3), wherein the first network connection (9a), the second network connection (9b), the first virtual network connection (4a) and the second virtual network connection (4b) have a ring topology on which a connection redundancy protocol is implemented.

2. The intermediate network (1) as claimed in claim 1, wherein the first network domain (2a) and/or the second network domain (2b) are Ethernet/Layer 2 domains.

3. The intermediate network (1) as claimed in one of claim 1 or 2, wherein the first virtual network connection (4a) and the second virtual network connection (4b) are virtual private network connections on Layer 2.

4. The intermediate network (1) as claimed in one of claims 1 to 3, wherein the connection redundancy protocol comprises a Spanning Tree Protocol, a Media Redundancy Protocol, an Ethernet Ring Protection Protocol or an Ethernet Automatic Protection Switching Protocol.

5. The intermediate network (1) as claimed in claim 4, wherein a domain-internal connection redundancy protocol is implemented in the first network domain (2a) and/or the second network domain (2b).

6. A method for setting up a network connection between two network domains (2a, 2b), with the following steps: establishment of a first virtual network connection (4a) between a first marginal node (6a) of a first network domain (2a) and a second marginal node (6b) of a second network domain (2b) via an intermediate network (3);establishment of a second virtual network connection (4b) between a third marginal node (7a) of the first network domain (2a) connected to the first marginal node (6a), and a fourth marginal node (7b) of the second network domain (2b) connected with the second marginal node (6b) via the intermediate network (3);implementation of a connection redundancy protocol in a ring topology formed by the first, the second, the third and the fourth marginal nodes (6a, 6b, 7a, 7b).

7. The method as claimed in claim 6, wherein the implementation of a connection redundancy protocol comprises the following steps: designation of a marginal node as the master node (6a); blocking (11) of the virtual network connection 4a) assigned to the designated master node (6a) for network traffic between the first and the second network domain (2a, 2b); andrelease of the respective other virtual network connection (4b) for network traffic between the first and the second network domain (2a, 2b).

8. The method as claimed in claim 7, furthermore with the following step: release of the blocked virtual network connection (4a), in the event of a fault (13) occurring in the released virtual network connection (4b).

9. The method as claimed in one of claims 6 to 8, wherein the first network domain (2a) and/or the second network domain (2b) are Ethernet/Layer 2 domains.

10. The method as claimed in one of claims 6 to 9, wherein the first virtual network connection (4a) and the second virtual network connection (4b) are virtual private network connections on Layer 2.

11. The method as claimed in one of claims 6 to 10, wherein the connection redundancy protocol comprises a Spanning Tree Protocol, a Media Redundancy Protocol, an Ethernet Ring Protection Protocol or an Ethernet Automatic Protection Switching Protocol.

12. The method as claimed in one of claims 6 to 11, furthermore with the following step: implementation of a domain-internal connection redundancy protocol in the first network domain (2a) and/or the second network domain (2b).