KR100573741B1 - Packet processing method using error state information in multiple fault tolerant network arrangement - Google Patents

Abstract

The present invention relates to a packet synthesis method using communication failure information in a multiple fault tolerance network structure capable of securing an optimal communication state even when multiple faults occur in a multiple fault tolerance network structure. To this end, according to the present invention, a packet synthesis method using communication failure information in a multi-fault tolerance network structure, in the multi-fault tolerance network structure, if a packet is input to two input terminals of a current node, whether the first input terminal is normal first according to priority A first step of checking, a second step of receiving the packet from the first input end if the first input end is normal, a third step of checking an error code included in the received packet, and the error code And a fifth step of synthesizing a new packet and updating an error code of the synthesized packet according to the state of two input terminals of the current node identified in the third and fourth steps. do.

Description

PACKET PROCESSING METHOD USING ERROR STATE INFORMATION IN MULTIPLE FAULT TOLERANT NETWORK ARRANGEMENT}

Figure 1a is a structural diagram of a conventional multiple fault tolerance network.

Figure 1b is a diagram illustrating the internal configuration of a node in a conventional multiple fault tolerance network.

Figure 1c is an exemplary diagram showing a case where a failure occurs at two input terminals of a specific node in a conventional multiple fault tolerance network.

2 is a block diagram of a packet including a failure status code applied to the present invention.

3 is a flow chart of a packet synthesis method according to the present invention.

** Explanation of symbols on the main parts of the drawing **

100 to 104: module (node) 105 to 114: connection line (link)

The present invention relates to a packet synthesis method using communication failure information in a multiple fault tolerance network structure capable of securing an optimal communication state even when multiple faults occur in a multiple fault tolerance network structure.

Large systems consist of modules (nodes) according to their function, and form a network for connecting each module with each other. In general, a module with a minimum connection line (link) is compared with a star or mesh structure. The ring structure that can be connected is mainly used. In the ring structure, a single ring structure has a disadvantage in that communication is lost between all nodes when one node or a connection line fails. A dual ring structure can overcome the above disadvantages. In the event of a failure of multiple nodes or multiple connections, communication of some or all of the systems may become impossible.

Therefore, in a system requiring high reliability, a multiple fault tolerance network structure capable of communication of a low total system should be used even if multiple faults occur.

FIG. 1A is a structural diagram of a conventional double ring tolerant multiple fault tolerance network, and related technology is disclosed in Korean Patent No. 0385116.

FIG. 1B shows the internal configuration of node 2 101 in the multiple fault tolerant network structure of FIG. 1A, which is the same for all other nodes 100, 102, 103, 104. Node 2 (101) has two input stages (IN [P], IN [S]) 105, 114 having priority and two output stages (OUT [P], OUT [S]) 107, 108 which are duplicated. ). Node 2 (101) receives two packets from two input terminals (105, 114), selects one packet according to priority, removes the other packet, and duplicates the selected packet to two output terminals (107, 108). . The output stages 107, 108 also become input terminals of other nodes and are connected in a ring form. Therefore, in a multiple fault tolerance network structure, even if a failure occurs in multiple links or multiple nodes, communication of the entire network can be maintained.

FIG. 1C illustrates a case where failure occurs at both input terminals 105 and 114 of the node 2 101. Here, assuming that node 3 (102) is the destination node and node 4 (103) is a start mode, node 2 (101) generates its own packet because node 2 (101) has failed at both input terminals (105, 114). The packet is output to 102 and node 4 (103). Node 3 (102) selects the packets output from node 2 (101) that has an input failure according to the priority because both inputs (106, 107) are normal, and node 4 (103), node 5 (104), and node 1 The packet generated at 100 is discarded.

As such, when multiple links or multiple nodes fail in a multiple fault tolerance network structure, communication of the entire network is maintained, but packets transmitted from the failed node or the previous node of the failed node are lost to maintain optimal communication. There is a problem that can not.

The present invention was devised to solve the above problems, and when multiple failures occur in a multiple fault tolerance network, a dual path input packet input to each node is selectively synthesized according to the communication failure to improve the communication quality. An object of the present invention is to provide a packet synthesis method using communication failure information in an optimally maintainable multiple fault tolerance network structure.

To this end, according to the present invention, a packet synthesis method using communication failure information in a multi-fault tolerance network structure, in the multi-fault tolerance network structure, if a packet is input to two input terminals of a current node, whether the first input terminal is normal first according to priority A first step of checking, a second step of receiving the packet from the first input end if the first input end is normal, a third step of checking an error code included in the received packet, and the error code And a fifth step of synthesizing a new packet and updating an error code of the synthesized packet according to the state of two input terminals of the current node identified in the third and fourth steps. do.

Hereinafter, the present invention will be described in detail with reference to the accompanying drawings.

The present invention synthesizes a packet received at a node according to the type of failure when a failure occurs in a multiple fault tolerance network structure. For this purpose, a failure status code (error code) indicating the type of failure is added to the packet.

2 schematically shows the format of a packet according to the invention.

As shown in FIG. 2, a packet includes a data field (data 1 to data N) containing information, a sync (SYNC) code for determining whether a packet is normally input to a node, and a neighbor immediately adjacent to a current node. It consists of an error (ERR) code indicating whether there is an error in the input packet of the previous node, and a time tag (TT) code.

If the error code of the packet inputted to the current node is 00, it indicates that the first input terminal IN [P] and the second input terminal IN [S] of the immediately adjacent node of the current node are normal, and the error is correct. When the code is 01, the first input terminal IN [P] is normal but the second input terminal IN [S] indicates that a failure has occurred. When the error code is 10, the first input terminal IN [P] is When a failure occurs and the second input terminal IN [S] is normal, when the error code is 11, it indicates that a failure occurs in both the first input terminal IN [P] and the second input terminal IN [S].

Therefore, at the present node, by checking the error code of the packet received from the input terminal, the packet is synthesized according to the error code.

3 shows a flowchart of a packet synthesis method according to the present invention.

As shown in FIG. 3, when a packet is input to two input terminals of a current node, a process for packet synthesis is started (S300), and according to a priority, it is first determined whether the packet of the first input terminal is normal (S301). If the packet of the first input terminal is normal, the packet is received from the first input terminal (S302). If the packet of the first input terminal is not normal, it is determined whether the packet of the second input terminal is normal according to the next priority (S320).

First, in step S301, when the packet of the first input terminal is normal and receives a packet from the first input terminal (S302), an error code of the packet received from the first input terminal may indicate that the packet of the previous node immediately neighbors of the current node. Check the status of the input terminal (S303). If the first input terminal of the previous node is normal (error codes 00, 01), the new packet is synthesized by adding the information obtained from the current node to the packet of the first input terminal of the current node (S304), and the error code is 00. Update to (S305). Here, if the first input terminal of the current node is normal and the first input terminal of the immediate neighboring node is normal, the error code is 00 because the normal packet is output through the output terminal of the current node regardless of the state of the second input terminal of the current node. Is updated.

However, if there is a failure in the first input terminal of the previous node (error codes 10 and 11), it is determined whether the packet of the second input terminal of the current node is normal (S306).

If it is determined in step S306 that the packet of the second input terminal of the current node is normal, the packet is received from the second input terminal (S309), and information of the previous node is extracted from the packet received at the first input terminal of the current node ( S310). The extracted information of the previous node is added to the packet of the second input terminal, synthesized into a new packet (S311), and the error code is updated to 00 (S312).

However, if the packet of the second input terminal of the current node is not normal, the information obtained at the current node is added to the packet of the first input terminal of the current node (S312), and the error code is updated to 01.

Next, in step S301, if the packet of the first input terminal is not normal, it is determined whether the packet of the second input terminal is normal (S320). If the packet of the second input terminal is normal, the packet is received from the second input terminal (S321), information obtained by the current node is added to the packet (S322), and the error code is updated to 10 (S321).

However, if the packets of the second input terminal are not normal, the failure occurs at both input terminals of the current node. Therefore, a new packet including information obtained from the current node is generated (S324), and the error code is updated to 11 (S325). .

As described above, the present invention can synthesize (reconstruct) the packet using two packet informations of the node input terminal according to the type of failure in the multi-failure tolerant network structure, even if a plurality of failures occur and accumulate during system network operation. It is effective to maintain the optimal communication state. Therefore, the present invention can be effectively applied to a field requiring stable operation for several years or decades after network installation, an application field having high repair cost, or an area where network repair is difficult.

Claims (6)

In a multiple fault tolerance network structure, when a packet is input to two input terminals of a current node, a first step of first checking whether the first input terminal is normal according to priority, and Receiving the packet from the first input terminal if the first input terminal is normal; A third step of checking an error code included in the received packet; A fourth step of synthesizing a new packet according to the error code; And a fifth step of updating the error code of the synthesized packet according to the state of the two input terminals of the current node identified in the third and fourth steps. Packet synthesis method using The method of claim 1, If the first input terminal is not normal, checking whether the second input terminal is normal; Receiving the packet from the second input terminal if the second input terminal is normal;  Synthesizing a new packet by adding the information obtained at the current node to the received packet; And updating the error code of the new packet to 10. 10. The method of claim 1, further comprising updating the error code of the new packet to 10. The method of claim 2, Generating a new packet including information obtained at the current node if the second input terminal is not normal; And updating the error code of the new packet to 11, by using communication failure information in a multiple fault tolerance network structure. The method of claim 1, When the error code is 00 or 01, a new packet is synthesized by adding information obtained at the current node to a packet received from the first input terminal, and updating the error code of the new packet to 00. Packet Synthesis Method using Communication Failure Information in Multiple Fault Tolerance Networks. The method of claim 1, When the error code is 10 or 11, check whether the second input terminal is normal, If the second input terminal is normal, the packet is received from the second input terminal, the information obtained from the previous node immediately neighboring the current node is extracted from the packet of the first input terminal, and the extracted information is extracted from the second node. Synthesize a new packet in addition to the packet received at the input, update the error code of the new packet to 00, If the second input terminal is not normal, the multiple fault tolerance network adds the information obtained from the current node to the packet of the first input terminal, synthesizes a new packet, and updates the error code of the new packet to 01. Packet synthesis method using communication failure information in a network structure The method of claim 1, The error code is a packet synthesis method using communication failure information in the multiple fault tolerant network structure, characterized in that indicating the state of the input terminal of the node of the multiple fault tolerant network structure.

Images