JP2003524334A - Fault Tolerance of Multiple Networks by Redundant Network Control - Google Patents

Abstract

SUMMARY The present invention provides a method and apparatus for detecting and managing the state of a computer network with network nodes having redundant network connections and recovering from multiple network failures. In one embodiment, a network status table is used at each node that manages data related to network conditions between the node and other nodes in the network. In various embodiments, rerouting of data is independently managed such that communication paths for transmitting data from a node to a connected node and receiving data from the connected node are independently selected. Is done. In some embodiments, the invention is operable to route data through one or more intermediate nodes when a direct connection between a pair of nodes is not possible.

Description

Detailed Description of the Invention

FIELD OF THE INVENTION The present invention relates generally to computer networks, and more particularly to a method and method for providing a fault tolerant network with redundant connections to network nodes capable of detecting and recovering from multiple network failures. Regarding the device.

Notification of Co-pending Applications This application is related to the following co-pending applications, which are incorporated herein by reference:

“Fault Toleran Networking”
Networking), Serial No. 09 / 188,976; and Attorney Case No. 256.045usl Background of the Invention In an environment where computers are used for work, computer networks have become increasingly important to communication and productivity. . E-mail replaces paper mail and fax as a means of disseminating information in many situations, and the availability of large amounts of information on the Internet is very valuable for both many work tasks and personal tasks. It became a resource. The ability to exchange data over a computer network also enables sharing of computer resources such as printers in a business environment, allowing networked computers to be centrally managed based on the network. became.

For example, an office worker's personal computer runs software that is automatically installed and updated over a network that is networked and shared by people in several different offices. It is possible to generate data to be printed by the printer. Networks can be used to inventory software and hardware installed within each personal computer, greatly simplifying inventory management tasks. Similarly, the software and hardware configurations of each computer can be managed over a network to facilitate user support operations within the network environment.

Networked computers are also typically connected to one or more network servers that provide data and resources to the networked computers. For example, a server may store some software applications that can be executed by its networked computers, or store a database of data that can be accessed and used by its networked computers. be able to. Network servers are also typically capable of managing access to particular networked devices, such as printers, which can be used by any networked computer. Similarly, a server may facilitate the exchange of data such as email or other similar services between networked computers.

Connecting from a local network to a larger network, such as the Internet, provides a greater ability to exchange data, such as by providing Internet email access or access to the World Wide Web. These data connections have made it practical to do business over the Internet and have contributed to the growth in the development and use of computer networks. Internet servers that provide data, perform functions such as e-commerce, audio or video streaming, e-mail, or provide other content are available on local networks, and such data
Rely on the operation of the Internet to provide a path between the server and client computer systems.

However, like other electronic systems, networks are prone to failure. Misconfigurations, wiring faults, electronic component failures, and some other factors can
It can damage the network connection and can render the computer network inoperable. In critical network environments, such as process control, medical, or other critical applications, backup or redundant network components can be used to minimize such failures. One example is to use a second network connection connecting critical network nodes that provides the same functionality as the first network connection. However, managing network connections that facilitate operation in the event of a network failure can be a daunting task, which itself depends on the network system or user's ability to properly detect and correct network failures. is there. Moreover, when both the primary and redundant networks fail, the exclusive use of either network makes it impossible to get the network fully operational. What is needed is a method and apparatus for detecting and managing the state of a computer network that uses redundant communication channels.

SUMMARY OF THE INVENTION The present invention is a method and apparatus for detecting and managing the state of a computer network comprising network nodes having redundant network connections and recovering from multiple network failures. In one embodiment, at each node, a network status table is used that manages data related to network conditions between that node and other nodes in the network. In various embodiments, rerouting of data is managed independently so that communication paths for transmitting data from a node to a connected node and receiving data from a connected node are independently selected. To be done. In some embodiments, the present invention is operable to route data through one or more intermediate nodes when a direct connection between a pair of nodes is not possible.

DETAILED DESCRIPTION The following detailed description of the exemplary embodiments of the invention refers to the accompanying drawings, which form a part of this specification, in which, by way of example, the invention is embodied. 2 illustrates a specific exemplary embodiment in which the These embodiments are described in sufficient detail to enable those skilled in the art to practice the invention, that other embodiments can be used, and be logical without departing from the spirit or scope of the invention. It should be appreciated that mechanical, electrical and other changes can be made. Therefore, the following detailed description should not be taken in a limiting sense, and the scope of the present invention is defined only by the appended claims.

The present invention provides a method and apparatus for detecting and managing the state of network connections to facilitate operation of redundant networks in the event of network failure. The present invention can compensate for multiple network failures, including failures in both primary and redundant networks. In some embodiments, the invention provides that the network is intact as long as at least one connection for transmitting data is operational and one connection for receiving data between each pair of networks is operational. Select either a primary or redundant network connection for communicating data between each pair of network nodes so that they can continue to operate.

The present invention is implemented in various forms using existing network technologies such as Ethernet (registered trademark). In one such embodiment, the Ethernet (
Two connections between the respective nodes are made via the registered trademark connection, the primary network connection and the redundant network connection. In some such embodiments, off-the-shelf network adapters are used, and the present invention controls the operation of the network adapters and manages communications via software running on computerized nodes. To do. Which connections are primary connections and which are redundant connections is not critical for the purposes of the present invention, since the connections are physically and functionally similar. In the example embodiments discussed herein, the primary and redundant network connections are interchangeable and are assigned names primarily for the purpose of distinguishing the networks from each other.

FIG. 1 shows an exemplary network with four nodes, 101, 102, 103 and 104. Primary network 105 and redundant network 106 connect each node to other nodes in the network, as indicated by the arrows connecting the nodes to each network. In order to understand how the present invention can operate to correct multiple network failures, the connection from node 3 of 103 to the primary network 105 is broken, resulting in node 107 3 cannot send data to the network 105. Similarly,
The connection connecting node 4 of 104 to redundant bus 106 is broken, resulting in 10
As shown in FIG. 8, node 4 is unable to receive data from the redundant bus.
As shown in FIG. 9, data cannot be transmitted to the redundant bus.

In a typical redundant network system, in the event of a single connection failure between the primary network and a node such as 107, all nodes on the network will communicate via redundant bus 106. Switch to. In the network configuration shown in FIG. 1, the connection between the node 4 and the redundant bus is also inoperable, rendering the network using the redundant bus inoperable. Such multiple failures render the network inoperable when using either the primary or redundant buses exclusively.

The present invention provides a solution to this problem and enables the use of network status data and intelligent routing of data to enable communication between all network nodes during multiple failures as shown in FIG. To do. In some embodiments of the invention, network status data is stored in the network status table shown in FIG.

FIG. 2 shows the network status for node 3 of the network of FIG.
An example of a table is shown, including data indicating the ability of node 3 to receive data from other nodes and the ability of other nodes to receive data from node 3. Specifically, the "Received Data OK" column indicates the ability of Node 3 to receive data from each of Nodes 1, 2, 4 on both the primary and redundant networks. This table indicates with "X" that node 3 cannot receive data from node 4 via the redundant network connection, and with "OK", node 3 transmits via both the primary and redundant network connection. Indicates that data can be received from all other nodes. An "X" indicating that node 3'cannot receive data from node 4 is the result of a broken data transmission connection 109 between redundant network 106 and node 4 (104).

The “Other Nodes Reported Data” column represents the data reported by other nodes to Node 3 regarding their ability to receive data from Node 3 by various other nodes. Since the connection of node 3'to the primary network 105 is broken at 107 and as a result node 3 is unable to send data over that connection, nodes 1, 2,
4 cannot receive data from the node 3 on the primary network,
'' Indicates a failure of node 3 for each of these nodes. Similarly, "node 3 redundancy" because the data connection between node 4 and the redundant network is broken at 108 and as a result node 4 is unable to receive data from the redundant network.
The "X" in the node "4" column of the row also indicates that node 4 is unable to receive data from node 3.

Determining whether a node can receive data from other nodes is, in various embodiments, a special purpose diagnostic data signal, a network protocol signal, or any other suitable signal sent between nodes. Is done using different types of data. The data that each node provides to other nodes for inclusion in the "reported data of other nodes" must be data that includes data that is communicated between the nodes and should be reported in one embodiment. A special purpose diagnostic data signal containing node data.

From the data in the network status table of FIG. 2, various network connection states can be determined, and the appropriate connection can be selected for communication between each pair of network nodes. In the embodiments of FIGS. 1 and 2, nodes 1 and 2 are fully operational and can use either connection to communicate, and nodes 3 and 4 are each fully connected to either the primary or redundant network. Have a working connection. Therefore, only nodes 3 and 4 cannot communicate exclusively via either the primary or the redundant network. Node 3 cannot send data to the primary network and node 4 cannot send or receive data to the redundant network.
Can receive data from the node 4 via the primary network. In some embodiments of the invention, node 3 is unable to send data to node 4 because there is no operational direct path through either the primary or redundant network for sending data.

In another embodiment of the invention, the node 1, node 2, etc., via the other node having an “OK” indication for any network in the “report data of other node” row of the table 3 can send data to node 4. In such an embodiment, the "OK" node or intermediate node is known to be able to receive data from node 3 and retransmits the data to node 4 via its fully functional primary network connection. can do. This allows
Communication between two nodes with multiple network failures can prevent direct communication between the two nodes. In a further embodiment, the intermediate nodes to which the data are routed are
Selected by polling an intermediate node to select a node that indicates that it can retransmit the data to node 4 by evaluating the data in the table. In various embodiments of the present invention, the intermediate nodes may include networked computers as in the above examples, may include direct connections between networks, may include routers or bridges, A special purpose intermediate node hardware device may be included or may be implemented in any other way that provides the ability to properly communicate signals between the two networks.

FIG. 3 is a flow chart illustrating a method of practicing an embodiment of the present invention. In 301,
Each node determines the state of the primary network connection that links it to each other node. Similarly, the state of redundant network connections linking each node to each other node is determined at 302. The state of the primary and redundant connections between each pair of nodes can be determined, in various embodiments, by searching the connections for valid data, existing data such as protocol packets, or by using special purpose diagnostic messages. You can judge. This network connectivity status data is used at 303 to build the "received data OK" portion of the network status table for each node, which node reports "other node reports" in the network status table. At 304, the data are exchanged with each other to complete the "data" portion. The network status table is updated periodically and monitored at 305 to determine if the network connection has failed and data rerouting is needed.

At 306, the node determines whether a direct connection can be made to send and receive data between a pair of nodes having a failed connection by examining a network status table. If the connection can be made, such as by sending data over the primary network connection and receiving data over the redundant network connection, then at 307 the data is rerouted through the direct connection and the other Monitoring for a failure restarts at 305. If a direct connection cannot be made, the data is transferred via one or more intermediate nodes at 308 to facilitate communication, as described in accordance with the multiple network failure example shown in FIGS. Rerouted. Once the data path through one or more intermediate nodes has been selected, monitoring resumes at 305 for other network failures.

The present invention enables a network with primary and redundant network connections to manage the routing of data through the network so that multiple network failures can be compensated. In some embodiments, the present invention involves rerouting data that cannot be transferred directly between two nodes to an intermediate node that can facilitate communication between the nodes. The present invention, in some embodiments, also incorporates building and using network status tables for managing data related to network conditions. The present invention, in various embodiments, provides a method for managing network state,
Includes software running on the computer to manage the state of the network, and a hardware network interface operable to manage the state of the network.

While specific embodiments have been shown and described herein, one of ordinary skill in the art will appreciate that any configuration suitable for achieving the same purpose can be substituted for the specific embodiments shown herein. Will. This application covers any adaptations or variations of the present invention. The invention is limited only by the full scope of the claims and their equivalents.

[Brief description of drawings]

FIG. 1 is a diagram of a computer network having multiple nodes with primary and redundant network connections, according to one embodiment of the invention.

FIG. 2 is a diagram showing an example of a network status table according to an embodiment of the present invention.

FIG. 3 is a flow chart of a method for managing the network state of a node having primary and redundant network connections according to an embodiment of the present invention.

─────────────────────────────────────────────────── ─── Continued front page    (81) Designated countries EP (AT, BE, CH, CY, DE, DK, ES, FI, FR, GB, GR, IE, I T, LU, MC, NL, PT, SE, TR), OA (BF , BJ, CF, CG, CI, CM, GA, GN, GW, ML, MR, NE, SN, TD, TG), AP (GH, G M, KE, LS, MW, MZ, SD, SL, SZ, TZ , UG, ZW), EA (AM, AZ, BY, KG, KZ, MD, RU, TJ, TM), AE, AL, AM, AT, AU, AZ, BA, BB, BG, BR, BY, CA, C H, CN, CR, CU, CZ, DE, DK, DM, EE , ES, FI, GB, GD, GE, GH, GM, HR, HU, ID, IL, IN, IS, JP, KE, KG, K P, KR, KZ, LC, LK, LR, LS, LT, LU , LV, MA, MD, MG, MK, MN, MW, MX, NO, NZ, PL, PT, RO, RU, SD, SE, S G, SI, SK, SL, TJ, TM, TR, TT, TZ , UA, UG, UZ, VN, YU, ZA, ZW (72) Inventor Freimark, Ronald Jay             85254, Arizona, United States             Tudale, North Sixty-Seven             De street 13014 (72) Inventor Kozlik, Tony Jay             Fee 85027, Arizona, USA             Nix, West Morrow Drive             2936 (72) Inventor Song, Sejung             Minneata, Minnesota 55414, Minnea             Police, Twenty Seventh Aveni             South East 1026, Apartment             Totoef F-term (reference) 5K032 DA11 EA04 EB06

Claims (30)

[Claims] 1. A method of determining a state of a primary network connection between each pair of networked nodes, a step of determining a redundant network connection state between each pair of networked nodes, and for communicating. Data between each pair of networked nodes so that the selected network path is selected independently based on the network conditions determined for each pair of networked nodes. Selecting either a primary network connection or a redundant network connection for sending and receiving, and a method of managing a computer network state having a redundant network connection comprising :. 2. The method of claim 1, further comprising: building a network status table showing the results of determining the primary and redundant network connection states between each pair of networked nodes. 3. The method of claim 2, wherein the network status table includes data representing network status based on data received at the node from other network nodes. 4. The method of claim 3, wherein the data received at the node from another network node comprises a diagnostic message. 5. The method of claim 4, wherein data received at another node from another network node comprises data representative of said other node's ability to receive data from another different network node. . 6. The method of claim 2, wherein the network status table includes data representing network status based on a node's ability to send data to other nodes. 7. The method of claim 3, wherein the network status table further comprises data representing network status based on a node's ability to send data to other nodes. 8. The step of selecting a primary or redundant network connection for communication between each pair of networked nodes comprises the primary network connection if the state of the primary network connection is determined to be operational. 2. The method of claim 1, comprising: selecting a redundant network connection if the state of the primary network connection is determined to be inoperable. 9. The step of selecting a primary or redundant network connection for communication between each pair of networked nodes comprises the step of: if the primary network connection state is determined to be operable for transmitting data. Selecting a primary network connection to send data, and selecting a primary network connection to receive data if the state of the primary network connection is determined to be operational to receive data Selecting a redundant network connection for transmitting data when the state of the primary network connection is determined to be inoperable for transmitting data, and the state of the primary network connection for transmitting data Redundant network to receive data when it is determined to be inoperable to receive. Comprises selecting a click connection, the method according to claim 1. 10. The step of selecting a connection for transmitting and receiving data between each pair of network nodes comprises transmitting and receiving data from the first node to one or more connected intermediate nodes. 2. The method of claim 1, comprising selecting a connection for sending data to and receiving data from an intermediate node to a second node. 11. A step of determining a primary network connection state between a network interface and a network interface of another network node, the network interface and the network of another network node.
Determining the state of redundant network connections between the interfaces, and other selected network connections to be independently selected based on the network state determined for each of the other network nodes. A computer network interface operable for selecting either a primary network connection or a redundant network connection for communication with each network node. 12. The method of claim 11, further comprising a network status table showing results of determining primary and redundant network connection states between the computer network interface and the network interfaces of other network nodes. Computer network interface. 13. The computer network of claim 12, wherein the network status table includes data representing network status based on data received at the node from other network nodes.
interface. 14. The computer network interface of claim 13, wherein the data received at the node from another network node comprises a diagnostic message. 15. The data from another network node received at a node further comprises data representative of said other node's ability to receive data from another different network node. Computer network interface. 16. The computer network interface of claim 12, wherein the network status table includes data representing network status based on a node's ability to send data to other nodes. 17. The computer network interface of claim 13, wherein the network status table further includes data representing network status based on a node's ability to send data to other nodes. 18. The step of selecting either a primary network connection or a redundant network connection for communicating with each of the other network nodes includes selecting the primary network connection if the primary network connection state is determined to be operational. 12. The computer network interface of claim 11, comprising: selecting, and selecting a redundant network connection if the state of the primary network connection is determined to be inoperable. 19. The step of selecting either a primary network connection or a redundant network connection to communicate with each of the other network nodes, wherein the primary network connection state is determined to be operable for transmitting data. Selecting a primary network connection for sending data, if the primary network connection state is determined to be operational for receiving data, then selecting the primary network connection for receiving data. Selecting a redundant network connection for transmitting data when the primary network connection state is determined to be inoperable for transmitting data,
And selecting a redundant network connection to receive data if the primary network connection state is determined to be inoperable to receive data. interface. 20. The step of selecting a connection for transmitting and receiving data between each pair of network nodes comprises transmitting and receiving data from the first node to one or more connected intermediate nodes. 12. The computer network interface of claim 11, comprising selecting the connection for selecting the connection for sending and receiving data from the intermediate node to the second node. 21. A machine-readable medium having instructions on the medium, the network interface and a network of other network nodes when executed on a computer.
Determining the primary network connection state between the interface, the network interface and the network interfaces of other network nodes.
Determining redundant network connection states to and from the interfaces, and so that selected network connections are independently selected based on the network states determined for each of the other network nodes. Selecting either a primary network connection or a redundant network connection for communicating with each of the network nodes of the machine readable medium having instructions operable to cause a computer to perform the. 22. A network status table showing the results of determining the status of primary and redundant network connections between a computer network interface and the network interfaces of other network nodes,
22. The machine-readable medium of claim 21, wherein the instructions are further operable to cause a computer to create and maintain. 23. The machine-readable medium of claim 22, wherein the created network status table includes data representative of network status based on data received at the node from other network nodes. 24. The machine-readable medium of claim 23, wherein the data received at the node from another network node comprises a diagnostic message. 25. The data received from another network node at a node further comprises data representative of said other node's ability to receive data from another different network node. Machine-readable medium. 26. The machine-readable medium of claim 22, wherein the created network status table includes data representing network status based on a node's ability to send data to other nodes. 27. The machine-readable medium of claim 23, wherein the network status table further comprises data representing network status based on a node's ability to send data to other nodes. 28. Choosing either a primary network connection or a redundant network connection to communicate with each of the other network nodes includes determining whether the state of the primary network connection is operational. 22. The machine-readable medium of claim 21, comprising: selecting a redundant network connection if the state of the primary network connection determines to be inoperable. 29. Choosing either a primary network connection or a redundant network connection to communicate with each of the other network nodes determines that the primary network connection state is operable for transmitting data. Select a primary network connection to send data, if the status of the primary network connection is determined to be operational to receive the data, then select the primary network connection to receive the data. Selecting a redundant network connection for transmitting data when the state of the primary network connection is determined to be inoperable for transmitting the data, and selecting a redundant network connection for transmitting the data Redundant to receive data when it is determined to be inoperable to receive Selecting a Ttowaku connection, including, machine-readable medium of claim 21. 30. Choosing a connection for sending and receiving data between each pair of network nodes comprises sending and receiving data from a first node to one or more connected intermediate nodes. 22. The machine-readable medium of claim 21, comprising selecting a connection for transmitting and receiving data from an intermediate node to a second node.