JP2005130152A - Network fault avoiding system and route switching method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To realize a network fault avoiding method, a route switching method, and a transport layer processing part, which allow communication to be continued without interruption even on the occurrence of communication fault on a network. <P>SOLUTION: In a system for avoiding fault of network, provided with a plurality of routes 1 and 2, data sent with sequence numbers attached from a transmission side are transmitted through a plurality of routes 1 and 2 and stored in a reception buffer 401 and a reception buffer 402 which are provided in a reception part of a transport layer processing part 13. When routes through which data will be provided for an application layer processing part 14 are switched, data having sequence numbers following the sequence number of data before switching is provided for the application layer processing part 14 from the reception buffer of routes after switching. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a method for avoiding a communication network failure, and more particularly, to a network failure avoidance method and a route switching method capable of performing communication without interrupting processing even when a communication failure occurs in a network.

  In a communication network system, when a failure occurs in the network, communication is interrupted or stopped, and important transmission data is lost in some cases, causing a great loss to the user. For this reason, avoiding a network failure provides a good service for the user.

  An example of a method for avoiding an obstacle is described in, for example, Japanese Patent Laid-Open No. 2001-60956 (Patent Document 1).

  In the method of Patent Document 1, communication connection routes are multiplexed, one piece of communication information is divided into a plurality of routes (subflows), and bundled as one communication (total flow) on the receiving side. Provide as information. In this method, information on the transmitting side application layer is divided into a plurality of networks, data with sequence numbers is transmitted in different subflows, each subflow is bundled on the receiving side, and rearranged according to the sequence numbers. And

As another failure avoidance method, a method of switching the path of an ATM (Asynchronous Transfer Mode) network is described in Japanese Patent Laid-Open No. 8-237253 (Patent Document 2). In this method, a shift in delay at the time of switching an active / backup virtual path (hereinafter abbreviated as VP) in an ATM network is detected, and VPs that have established synchronization are switched.
JP 2001-60956 A JP-A-8-237253

  All of the conventional techniques described above have the following problems.

  In the case of the method of Patent Document 1, when the subflow is disconnected, data after a time point when it is not confirmed that the subflow has arrived is retransmitted as a total flow. This method can cope with communication failure due to sudden network failure. However, since retransmission processing is performed, a delay occurs, and communication interruption cannot be avoided.

  In the case of the method of Patent Document 2, it is assumed that the cause of the failure is a delay in switching the ATM network when switching between the active and standby VPs, and does not correspond to a sudden network failure including an accident or the like.

  An object of the present invention is to provide a network failure avoidance system and a path switching method that can solve the above-described drawbacks of the prior art and continue communication without interrupting processing even if a communication failure occurs in the network. There is.

  In order to achieve the above object, the present invention provides a network failure avoidance system in a network system in which a transmission side and a reception side are connected by a plurality of paths, and the received layer data is subjected to protocol processing in a transport layer. When a failure occurs in a route that is accumulated in the transport layer processing unit that performs, a normal route is selected and switched from the plurality of networks, and data received after the failure occurs through the selected route is It is provided to an application layer processing unit that performs layer protocol processing.

  The present invention of claim 2 is a network failure avoidance system in a network system in which a transmission side and a reception side are connected by a plurality of paths, and the data sent with a sequence number assigned from the transmission side is transferred to the transport layer. When a failure occurs in the path, the transport layer processing unit that performs the protocol processing of the protocol stores a failure, and if a failure occurs in the route, the normal route is selected from the plurality of networks and switched. Data after the sequence number received when a failure occurs in the path is provided from the reception buffer of the selected path to an application layer processing unit that performs protocol processing of the application layer.

  The network failure avoidance system according to a third aspect of the present invention is characterized in that selection of a route to be provided to the application layer processing unit is performed by a selector provided in the transport layer processing unit.

  The network failure avoidance system according to claim 4 of the present invention is based on the communication status report information for each route reported by the selector from the TCP processing unit of the transport layer processing unit. It is characterized by performing.

  The network failure avoidance system of the present invention according to claim 5 is characterized in that the selector of the transport layer processing unit is provided with a sequence number giving unit for giving the sequence number to the data to be transmitted.

  The network failure avoidance system according to the present invention of claim 6 is characterized in that the sequence number assignment on the transmission side is performed by providing a sequence number description area of about several bytes in the data head.

  The network failure avoidance system of the present invention according to claim 7 is characterized in that the selector of the transport layer processing unit deletes information related to a sequence number in data sent with a sequence number assigned from the transmission side. To do.

  In the network failure avoidance system of the present invention according to claim 8, the selector determines whether data after the sequence number received when a failure occurs in the route in which the failure has occurred exists in the reception buffer of the selected route. If present, the data of the reception buffer is provided to the application layer processing unit. If the data does not exist, the application layer processing unit is received when data after the sequence number is received in the reception buffer of the selected path. It is characterized by providing to.

  The present invention of claim 9 is a path switching method in a network system in which a transmission side and a reception side are connected by a plurality of paths, and transport layer protocol processing of data received for each path is performed. When a failure occurs in a route that is accumulated in the processing unit, a normal route is selected and switched from the plurality of networks, and data received after the failure occurs through the selected route is processed by an application layer protocol process. It is provided to an application layer processing unit that performs the above.

  The present invention of claim 10 is a path switching method in a network system in which a transmission side and a reception side are connected by a plurality of paths, and data sent with a sequence number assigned by the transmission side is transferred to the transport layer. When a failure occurs in a route that is stored in a reception buffer corresponding to a plurality of routes of a transport layer processing unit that performs protocol processing, when a failure occurs in the route, a normal route is selected and switched from the plurality of networks, and a route in which a failure occurs The data after the sequence number received when a failure occurs is provided to the application layer processing unit that performs the protocol processing of the application layer from the reception buffer of the selected route.

  The route switching method of the present invention according to claim 11 is characterized in that a route provided to the application layer processing unit is selected by a selector provided in the transport layer processing unit.

  The route switching method of the present invention according to claim 12 is characterized in that the switching by the selection of the route is performed based on the communication state report information for each route reported from the TCP processing unit of the transport layer processing unit. And

  According to a thirteenth aspect of the present invention, there is provided the path switching method according to the present invention, wherein the sequence number is assigned to the data to be transmitted by the sequence number assigning unit provided in the selector of the transport layer processing unit.

  The route switching method of the present invention according to claim 14 is characterized in that the sequence number assignment on the transmission side is performed by providing a sequence number description area of about several bytes in the data head.

  The route switching method of the present invention according to claim 15 is characterized in that the selector of the transport layer processing unit deletes information relating to a sequence number in data sent with a sequence number assigned from the transmission side. .

  In the route switching method of the present invention according to claim 16, the selector determines whether data after the sequence number received at the time of failure in the route in which the failure has occurred exists in the reception buffer of the selected route, If present, the data of the reception buffer is provided to the application layer processing unit. Otherwise, the data after the sequence number is received by the reception buffer of the selected path to the application layer processing unit. It is characterized by providing.

  According to the network failure avoidance system and the path switching method of the present invention, the following effects are achieved.

  First, since data is sent from the transmission side via a plurality of routes, and information from the normal route is provided to the application layer processing unit by the selector on the reception side, even if a communication failure occurs in the route, Communication can be performed without interrupting processing.

  Secondly, since the process is not interrupted, there is an effect that the communication processing time can be shortened compared with the case of performing the retransmission process, and as a result, a high-quality service can be provided to the user.

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

  FIG. 1 is a block diagram showing the overall configuration of a network failure avoidance system according to a first embodiment of the present invention. Referring to FIG. 1, in the network failure avoidance system according to the present embodiment, a host 10 and a host 20 as network terminals are connected by a route 1 and a route 2 on the network. Each of the host 10 and the host 20 includes a data link layer processing unit 11 that performs data link layer protocol processing, a network layer processing unit 12 that performs network layer protocol processing, and a transport layer processing unit that performs transport layer protocol processing. 13 and an application layer processing unit 14 for performing application layer protocol processing.

  The transport layer processing unit 13 includes a selector 300 that switches networks.

  FIG. 2 is a diagram illustrating a configuration of the receiving unit of the transport layer processing unit 13 according to the present embodiment. The transport layer processing unit 13 includes a TCP processing unit (Transmission Control Protocol) 101, a TCP processing unit 102, and a selector 300. The communication status report interfaces 201 and 202 are provided in the TCP processing unit 101 and the TCP processing unit 102, respectively.

  The selector 300 includes a reception buffer 401, a reception buffer 402, and a sequence number monitoring unit 500. In the reception buffer 401 or 402, data sent from the TCP processing units 101 and 102 is accumulated. The sequence number monitoring unit 500 always monitors the sequence number of the route provided to the application layer processing unit 14.

  The selector 300 switches a route to be provided to the application layer processing unit 14 based on information of the communication status report interfaces 201 and 202. The selector 300 monitors the information of the status communication status report interfaces 201 and 202 of the route 1 and route 2, and when a communication failure occurs in the route currently provided to the application layer processing unit 14, it switches to another normal route. Switch.

  FIG. 3 is a diagram showing the configuration of the transmission unit of the transport layer processing unit 13 according to this embodiment. The transport layer processing unit 13 includes a TCP processing unit 101, a TCP processing unit 102, and a selector 300. The communication status report interfaces 201 and 202 are provided in the TCP processing unit 101 and the TCP processing unit 102, respectively.

  The selector 300 includes a sequence number assigning unit 600. The sequence number assigning unit 600 assigns a sequence number to the data provided from the application layer processing unit 14.

  The apparatus configuration of the present embodiment has been described above.

  Next, the operation according to this embodiment will be described with reference to FIGS.

  Referring to FIG. 3, the data transmitted from the application layer processing unit on the transmission side is given a sequence number by the sequence number giving unit 600 of the selector 300, and is then transmitted to the TCP processing unit 101 and the TCP processing unit 102. 1 and the route 2 are transmitted to the receiving side. The sequence number is described, for example, by providing a sequence number writing area of about several bytes in the data head.

  Referring to FIG. 2, the data on path 1 and path 2 are provided to the selector 300 via the TCP processing unit 101 and the TCP processing unit 102 on the receiving side. The selector 300 determines which route is to be selected based on the information of the communication status report interfaces 201 and 202, and provides one normal route to the application layer processing unit 14 which is an upper layer. For example, if the TCP processing unit of one route frequently repeats ACK (Acknowledgement) time-over, or if ACK is returned slowly even if it does not reach time-out, it is determined that a failure has occurred. Switch to the route.

  Here, as a specific example, the path switching operation of the selector when a failure occurs in the path 1 will be described. In this example, the route 1 was initially selected, but the route is switched to the route 2 when the failure of the route 1 occurs. If data is lost during this switching, communication is interrupted.

  A flowchart of the route switching method according to this embodiment is shown in FIG. Hereinafter, the path switching will be described with reference to FIG.

  First, it is detected from the communication status report interface 201 provided in the TCP processing unit 101 that a failure has occurred in the path 1 (step 1001).

  When the network failure of the route 1 is detected, the selector 300 switches the route provided to the application layer processing unit from the route 1 to the route 2, and the current buffer 1 becomes the current route 1 in the reception buffer 402 of the route 2 to be newly switched. It is confirmed whether or not there is data after the sequence number of the provided data (step 1002), and if it exists, continuous data after the sequence number is provided from the reception buffer 402 to the application layer processing unit 14 ( Step 1003).

  If there is no data after the sequence number in the reception buffer 402, as soon as the subsequent data after the sequence number arrives from the newly switched path 2 (step 1004), the data is provided to the application layer processing unit 14 to switch the network. Completion (step 1005).

  In the above embodiment, a method of switching two routes is described, but switching of a plurality of three or more routes can be performed in exactly the same manner.

  Further, the amount of data in the reception buffer increases with time. Among the accumulated data, the data before the sequence number of the data currently provided to the application layer processing unit is unnecessary and can be deleted.

  Further, when the selector provides data to the application layer processing unit, sequence number information is added to the transmission source data in order to provide data with a sequence number assigned on the transmission side for the purpose of avoiding a network failure. . In such a case, by removing the sequence number by the selector 300, data without additional information can be provided to the application layer processing unit.

  Although the present invention has been described with reference to the preferred embodiments, the present invention is not necessarily limited to the above embodiments, and various modifications can be made within the scope of the technical idea.

1 is a block diagram showing an overall configuration of a network failure avoidance system according to a first embodiment of the present invention. It is a figure which shows the structure of the receiving part of the transport layer process part by 1st Example of this invention. It is a figure which shows the structure of the transmission part of the transport layer process part by 1st Example of this invention. It is a figure which shows the flowchart of the path | route switching method by 1st Example of this invention.

Explanation of symbols

1, 2: Path 10, 20: Host 11: Data link layer processing unit 12: Network layer processing unit 13: Transport layer processing unit 14: Application layer processing unit 101, 102: TCP processing unit 201, 202: Communication status report Interface 300: Selector 401, 402: Reception buffer 500: Sequence number monitoring unit 600: Sequence number giving unit

Claims (16)

A network failure avoidance system in a network system in which a transmission side and a reception side are connected by a plurality of paths,
The data received for each path is accumulated in a transport layer processing unit that performs protocol processing of the transport layer,
When a failure occurs in the route, an application layer processing unit that selects and switches a normal route from the plurality of networks and performs data processing received after the failure occurs on the selected route, in an application layer. A network failure avoidance system, characterized by being provided for. A network failure avoidance system in a network system in which a transmission side and a reception side are connected by a plurality of paths,
The data sent with the sequence number assigned from the transmission side is accumulated in the reception buffer corresponding to the plurality of paths of the transport layer processing unit that performs the protocol processing of the transport layer,
When a failure occurs in a route, a normal route is selected and switched from the plurality of networks, and data after the sequence number received at the time of failure in the route in which the failure has occurred is received from the reception buffer of the selected route. A network failure avoidance system, characterized in that the system is provided to an application layer processing unit that performs application layer protocol processing.   The network failure avoidance system according to claim 1 or 2, wherein a route to be provided to the application layer processing unit is selected by a selector provided in the transport layer processing unit.   The said selector performs switching by the selection of the said path | route based on the report information of the communication state for every said path | route reported from the TCP process part of the said transport layer process part. Network failure avoidance system.   5. The network failure avoidance system according to claim 3, wherein a sequence number assigning unit that assigns the sequence number to the data to be transmitted is provided in the selector of the transport layer processing unit.   6. The network failure avoidance system according to claim 3, wherein the sequence number is assigned on the transmission side by providing a sequence number description area of about several bytes in the data head. .   7. The information on the sequence number in the data sent with the sequence number assigned by the transmission side is deleted by the selector of the transport layer processing unit. Network failure avoidance system described in 1.   The selector determines whether there is data after the sequence number received at the time of the failure in the route where the failure has occurred in the reception buffer of the selected route, and if so, the data of the reception buffer is transferred to the application The data is provided to a layer processing unit, and when it does not exist, the data after the sequence number is provided to the application layer processing unit at the time when the data is received by the reception buffer of the selected path. Item 8. The network failure avoidance system according to any one of Item 7. A route switching method in a network system in which a transmission side and a reception side are connected by a plurality of routes,
The data received for each path is accumulated in the transport layer processing unit,
When a failure occurs in the route, an application layer processing unit that selects and switches a normal route from the plurality of networks and performs data processing received after the failure occurs on the selected route, in an application layer. A route switching method characterized by being provided for. A route switching method in a network system in which a transmission side and a reception side are connected by a plurality of routes,
The data sent with the sequence number assigned from the transmission side is accumulated in the reception buffer corresponding to the plurality of paths of the transport layer processing unit that performs the protocol processing of the transport layer,
When a failure occurs in a route, a normal route is selected and switched from the plurality of networks, and data after the sequence number received at the time of failure in the route in which the failure has occurred is received from the reception buffer of the selected route. A route switching method comprising: providing an application layer processing unit that performs application layer protocol processing.   The route switching method according to claim 9 or 10, wherein selection of a route to be provided to the application layer processing unit is performed by a selector provided in the transport layer processing unit.   12. The path switching method according to claim 11, wherein switching by selecting the path is performed based on report information on a communication state for each path reported from a TCP processing unit of the transport layer processing unit.   13. The path switching method according to claim 11 or 12, wherein the sequence number is assigned to the data to be transmitted by a sequence number assigning unit provided in the selector of a transport layer processing unit.   The path switching method according to any one of claims 11 to 13, wherein the sequence number assignment on the transmission side is performed by providing a data head with a sequence number description area of about several bytes.   15. The information on the sequence number in the data sent with the sequence number assigned by the transmission side is deleted by the selector of the transport layer processing unit. The route switching method described in 1.   The selector determines whether there is data after the sequence number received at the time of the failure in the route where the failure has occurred in the reception buffer of the selected route, and if so, the data of the reception buffer is transferred to the application 11. The data is provided to a layer processing unit, and when it does not exist, the data after the sequence number is provided to the application layer processing unit at the time when the data is received by the reception buffer of the selected path. Item 16. The route switching method according to any one of Item 15.

Images