JP2011044844A - Multicast packet route-switching system and multicast packet route-switching method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a multicast packet route-switching system and method, capable of shortening the disconnection time of multicast communication caused by a network fault. <P>SOLUTION: The multicast packet route-switching system includes a plurality of transfer devices and a controller for determining a transfer route having at least one transfer device. The transfer device includes a storage part for storing a table where the header information of a multicast packet and output destination information indicating the output destination of the packet are made to correspond on the basis of the transfer route, and a transfer part for referring to the storage part and transferring the multicast packet to the output destination indicated by the output destination information corresponding to the header information of the packet when receiving the packet. The controller includes: a storage part for storing information indicating the connection relation of the transfer devices with each other and the transfer route; and an updating part for, when a fault occurs in the transfer route, referring to the storage part to determine a new transfer route to substitute for the transfer route and updating the table inside the transfer device inside the new transfer route on the basis of the new transfer route. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a multicast packet path switching system and a multicast packet path switching method.

  In the conventional multicast packet transfer system, the first router that receives the multicast JOIN signal from the terminal determines whether the multicast packet requested by the multicast JOIN signal is distributed to its own router. If the multicast packet is not distributed to the own router, the first router transmits a multicast JOIN signal to the next router on the upstream side.

  When the multicast packet is distributed to the own router, the next-stage router builds a multicast tree and transmits the multicast packet to the first router. On the other hand, if the multicast packet is not distributed to the own router, the next-stage router transmits a multicast JOIN signal to the next-stage router further upstream. The next-stage router constructs a multicast tree after the multicast packet is distributed via the next-stage router, and transmits the multicast packet to the first router.

  The first router constructs a multicast tree after the multicast packet is distributed via the next-stage router, and transmits the multicast packet to the terminal.

  In this multicast packet forwarding system, when a failure occurs in a router in the multicast tree during distribution of the multicast packet, a new multicast tree is constructed by connecting the routers in multiple stages using the multicast JOIN signal again, and then The multicast packet delivery is resumed using the newly constructed multicast tree.

  FIG. 10 is a diagram for explaining an example of a rebuilding operation of a multicast tree when a failure occurs in a router belonging to the multicast tree.

  In FIG. 10, it is assumed that a failure occurs in the router 102 in a situation where a multicast stream composed of multicast packets is distributed from the distribution server 100 to the terminal 107 via the router 103, the router 102, and the router 101. For this reason, distribution of the multicast stream to the terminal 107 is disconnected.

  The router 101 detects the occurrence of a failure in the router 102 because the router 102 does not respond to the Hello packet from the router 101.

When the router 101 detects a failure in the router 102, the router 101 recalculates the route, specifies the router 105 that replaces the router 102, and transmits a multicast JOIN signal to the router 105. In FIG. 10, the multicast JOIN signal is indicated as a multicast viewing request (PIM-Join). When the router 105 receives the multicast JOIN signal from the router 101, the multicast packet requested by the multicast JOIN signal is not transmitted. Check if it is distributed to the router.

  If the multicast packet is not distributed to the own router, the router 105 performs route search and transmits a multicast JOIN signal to the router 103 on the upstream side.

  Upon receiving the multicast JOIN signal from the router 105, the router 103 switches the multicast stream distribution route (multicast tree) from the router 102 to the router 105, and transmits the multicast packet to the router 105.

  When the router 105 receives the multicast packet from the router 103, the router 105 constructs a multicast tree and transmits the multicast packet to the router 101.

  Patent Document 1 describes a function distribution type packet transfer system.

  The function-distributed packet transfer system separates the router functions into a data transfer system (FE: Forwarding Element) and a router control system (CE: Control Element), and deploys each of the FE and CE on the network. Architecture.

JP 2009-159417 A

  In a conventional multicast packet forwarding system, when a failure occurs in a router belonging to a multicast tree during distribution of the multicast packet, the multicast tree is reconstructed by connecting the routers in multiple stages using the multicast JOIN signal again, and then Multicast packet delivery is resumed.

  For this reason, there has been a problem that it takes a lot of time from the occurrence of a network failure until the delivery of the multicast stream is resumed.

  Note that Patent Document 1 describes a function-distributed packet transfer system, but does not describe processing performed when a failure occurs. For this reason, the technique described in Patent Document 1 naturally solves the above-described problem, that is, the problem that a long time is required from the occurrence of a failure until the multicast stream delivery is resumed. It has not been.

  An object of the present invention is to provide a multicast packet path switching system and a multicast packet path switching method capable of shortening a multicast communication disconnection time caused by a network failure.

  The multicast packet path switching system of the present invention includes a plurality of transfer apparatuses having a function of transferring a multicast packet, and a control apparatus that determines a transfer path including at least one of the plurality of transfer apparatuses, The transfer device stores a table in which header information of the multicast packet and output destination information indicating the output destination of the multicast packet are associated with each other based on the transfer path, and receives the multicast packet, the storage A transfer unit that transfers the multicast packet to an output destination indicated by output destination information associated with header information of the multicast packet, and the control device connects the transfer devices to each other. Stores connection information indicating the relationship and transfer path information indicating the transfer path. When a failure occurs in the storage unit and the transfer path, the storage unit is referred to determine a new transfer path to replace the transfer path and stored in the transfer device in the new transfer path. And an update unit that updates the table based on the new transfer path.

  A multicast packet route switching method according to the present invention includes a plurality of transfer devices having a function of transferring a multicast packet, and a control device that determines a transfer route including at least one of the plurality of transfer devices. A multicast packet path switching method in a path switching system, wherein the transfer device associates header information of a multicast packet with output destination information indicating an output destination of the multicast packet based on the transfer path Storing in the storage unit, and when the transfer device receives the multicast packet, the storage unit is referred to the output destination indicated by the output destination information associated with the header information of the multicast packet. Forwarding step for forwarding multicast packets A storage step in which the control device stores connection information indicating a connection relationship between the transfer devices and transfer route information indicating the transfer route in a storage unit; and When a failure occurs, the storage unit is referred to determine a new transfer path that replaces the transfer path, and the table stored in the transfer device in the new transfer path is changed to the new transfer path. And updating based on the update step.

  According to the present invention, when a failure occurs in the transfer route of the multicast packet specified by the transfer device, the control device determines a new transfer route to replace the failed transfer route, and the new transfer The table included in the transfer device in the path is updated based on the new transfer path. For this reason, it is possible to shorten the switching time of the transfer path by connecting the routers in multiple stages using the multicast JOIN signal as in the conventional case, as compared with the case of reconstructing the multicast tree.

It is the block diagram which showed the multicast packet path | route switching system of 1st Embodiment of this invention. It is the block diagram which showed the component common to FE1-6. It is the block diagram which showed CE7. It is a block diagram for demonstrating operation | movement of the multicast packet path | route switching system of 1st Embodiment. It is the block diagram which showed the multicast packet path | route switching system of 2nd Embodiment of this invention. It is the block diagram which showed the component common to FE1A-6A. It is the block diagram which showed CE7A. It is a block diagram for demonstrating operation | movement of the multicast packet path | route switching system of 2nd Embodiment. It is the block diagram which showed the application example of the multicast packet path | route switching system of 1st, 2nd embodiment. It is a figure for demonstrating an example of the reconstruction operation | movement of the conventional multicast tree.

  Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

(First embodiment)
FIG. 1 is a block diagram illustrating a multicast packet path switching system according to a first embodiment of this invention.

  In FIG. 1, the multicast packet path switching system includes a plurality of transfer devices (hereinafter referred to as “FE”) 1 to 6 and a control device (hereinafter referred to as “CE”) 7.

  Each of the FEs 1 to 6 is directly or indirectly connected to each other, has a function of transferring a multicast packet, and performs a process classified as a U plane (user plane). Each of the FEs 1 to 6 transfers the multicast packet according to the transfer path determined by the CE 7. Note that the number of FEs is not limited to six and may be a plurality.

  In this embodiment, FE3 and FE6 are connected to the distribution server 8, and FE1 is connected to the terminal 9 and the terminal 10. The FE connected to the distribution server 8 is not limited to FE3 and FE6. Further, the FE connected to the terminal 9 and the terminal 10 is not limited to FE1.

  The CE 7 calculates a transfer route, thereby determining a transfer route (multicast packet transfer route) including at least one of the FEs 1 to 6. The CE 7 controls the transfer by the FE included in the transfer path based on the transfer path. The CE 7 performs processing classified as a C plane (control plane).

  FIG. 2 is a block diagram showing components common to FE1 to FE6. In FIG. 2, FE1 is shown as an example.

  In FIG. 2, FE1 includes a storage unit 1a, a transfer unit 1b, and a transmission unit 1c. The transfer unit 1b includes one or more input units 1b1, one or more output units 1b2, and a packet transfer unit 1b3.

  The storage unit 1a stores a routing information table (hereinafter simply referred to as “table”) 1a1. In the table 1a1, the header information of the multicast packet and the output destination information indicating the output destination of the multicast packet are associated with each other based on the transfer path determined by the CE7.

  In the present embodiment, the output destination information indicates one or more output units 1b2 that output the multicast packet as the output destination of the multicast packet. Also, as header information of the multicast packet, information on the source address that is the address of the distribution server 8 that is the distribution source of the multicast packet and information on the group address that is the address of the group to which the multicast packet is distributed are stored.

  When receiving the multicast packet, the transfer unit 1b refers to the storage unit 1a and transfers the multicast packet to the output destination indicated by the output destination information associated with the header information of the multicast packet.

  The input unit 1b1 is connected to an output unit of another FE, the terminal 9, the terminal 10, or the distribution server 8.

  The output unit 1b2 is connected to an input unit of another FE, the terminal 9, the terminal 10, or the distribution server 8.

  When the input unit 1b1 receives a multicast packet, the packet transfer unit 1b3 transmits the multicast packet from one or more output units 1b2 indicated by the output destination information associated with the header information of the multicast packet.

  When the transmission unit 1c receives a predetermined failure detection signal transmitted from the CE 7 to the FE, the transmission unit 1c transmits a response signal to the failure detection signal to the CE 7 (transmission source of the failure detection signal).

  FIG. 3 is a block diagram showing CE7.

  In FIG. 3, the CE 7 includes a storage unit 7a and an update unit 7b. The update unit 7b includes a detection unit 7b1 and an update control unit 7b2.

  The storage unit 7a stores connection information indicating the connection relationship of FE1 to FE6 and transfer path information indicating the transfer path determined by the CE7. In the present embodiment, the storage unit 7a stores a table 1a1 stored in each storage unit 1a of the FE1 to FE6 as transfer path information. The storage unit 7a may store the same table as the table 1a1 in the FE1 to FE6, or may store a table in which the tables 1a1 in the FE1 to 6 are integrated.

  When a failure occurs in the transfer path, the update unit 7b refers to the storage unit 7a (specifically, connection information and transfer path information), and recalculates the transfer path to replace the transfer path. The new transfer route is determined, and the table 1a1 stored in the FE included in the new transfer route is updated based on the new transfer route.

  The detection unit 7b1 detects an FE that does not transmit a predetermined signal among the FE1 to FE6 as a FE in which a failure has occurred.

  For example, the detection unit 7b1 transmits an HB (heartbeat) signal to each of the FE1 to FE6, and a failure has occurred in the FE that does not transmit a response signal to the HB signal during a predetermined time thereafter. Detect as FE. Note that each of the FEs 1 to 6 receives a HB signal from the detection unit 7b1 in a normal state and transmits a response signal to the HB signal to the detection unit 7b1 during a predetermined time thereafter. The HB signal is an example of a failure detection signal, and the response signal to the HB signal is an example of a predetermined signal.

  In addition, when each transmission part 1c of FE1-6 is set so that HB signal may be periodically transmitted to CE7 (detection part 7b1), the detection part 7b1 should be transmitted regularly. An FE that does not transmit an HB signal may be detected as a FE in which a failure has occurred. In this case, the HB signal periodically transmitted is an example of a predetermined signal.

  The update control unit 7b2 refers to the storage unit 7a (specifically, connection information and transfer path information) and recalculates the transfer path when the FE in which the failure has occurred is an FE in the transfer path. Thus, as a new transfer route, a transfer route including an FE different from the failed FE (that is, not including the failed FE and transferring the multicast packet from the distribution server 8 to the terminal 9 and the terminal 10) Transfer path).

  The update control unit 7b2 updates the table 1a1 stored in the FE in the new transfer path based on the new transfer path so that the multicast packet is transferred through the new transfer path.

  Next, the operation will be described.

  FIG. 4 is a block diagram for explaining the operation of the multicast packet path switching system of the first embodiment.

  In FIG. 4, it is assumed that a failure occurs in FE2 in a situation where multicast packets are transferred from the distribution server 8 in the order of FE3, FE2, and FE1 and distributed to the terminals 9 and 10. Note that the FE in which a failure occurs is not limited to FE2.

  Before the occurrence of a failure in FE2, the update control unit 7b2 in the CE 7 refers to the storage unit 7a in the CE 7 and calculates the transfer route, thereby sending the multicast packet from the distribution server 8 to FE3, FE2, and FE1. The table 1a1 included in the FE3, FE2, and FE1 in the transfer path is set through the C plane so that the multicast packet is transferred in the order of FE3, FE2, and FE1.

  In a situation where a failure has occurred in FE2, when the detection unit 7b1 in the CE 7 transmits an HB signal to each of the FE 1 to 6 through the C plane (step A1), each of the FE 1 and the FE 3 to 6 Although a response signal for the HB signal is transmitted to the CE 7 during the predetermined time, a response signal for the HB signal is not transmitted to the CE 7 from the FE 2 during the subsequent predetermined time. For this reason, the detection unit 7b1 in the CE 7 detects the FE 2 as the FE in which the failure has occurred (step A2).

  Subsequently, the update control unit 7b2 in the CE 7 refers to the storage unit 7a in the CE 7 and confirms whether the failed FE 2 is an FE in the transfer path. When the failed FE2 is an FE in the transfer path, the update control unit 7b2 in the CE 7 refers to the storage unit 7a in the CE 7 to recalculate the transfer path, thereby creating a new transfer path. Then, a new transfer path including FE3, FE5, and FE1 that is different from the FE2 in which the failure has occurred is determined (step A3).

  Subsequently, the update controller 7b2 in the CE 7 allows the FE3, FE5, and FE1 in the new transfer path to pass through the C plane so that the multicast packet from the distribution server 8 is transferred in the order of FE3, FE5, and FE1. The included table 1a1 is updated (steps A4 and A5). In this case, the table 1a1 included in the FE1 may not be updated. As an update method for the table 1a1, the update control unit 7b2 creates the tables 1a1 for FE3, FE5, and FE1, and distributes the created tables 1a1 to FE3, FE5, and FE1, respectively. May update the table 1a1 of FE3, FE5, and FE1.

  According to the present embodiment, when a failure occurs in the transfer path, the update unit 7b refers to the inside of the storage unit 7a, and recalculates the transfer path, thereby replacing the transfer path with a new transfer path. And the table 1a1 included in the FE included in the new transfer path is updated based on the new transfer path.

  For this reason, it is possible to shorten the switching time of the transfer path by connecting the routers in multiple stages using the multicast JOIN signal as in the conventional case, as compared with the case of reconstructing the multicast tree.

  In the present embodiment, the detection unit 7b1 detects an FE that does not transmit a predetermined signal among the FEs 1 to 6 as an FE in which a failure has occurred. For example, the detection unit 7b1 detects an FE that has transmitted an HB signal to each of the FE1 to FE6 and has not transmitted a response signal to the HB signal during a predetermined time thereafter as an FE in which a failure has occurred. Or, an FE that does not transmit a periodically transmitted HB signal is detected as an FE in which a failure has occurred.

  In this case, the CE 7 can detect the failure itself.

(Second Embodiment)
Next, a multicast packet path switching system according to a second embodiment of this invention will be described. In the multicast packet path switching system of the second embodiment, each FE detects a failure in a link with an adjacent FE, notifies the failure to the CE, and the CE transfers based on the notified failure. The point of recalculating the route is different from the multicast packet route switching system of the first embodiment.

  FIG. 5 is a block diagram illustrating a multicast packet path switching system according to the second embodiment of this invention.

  In FIG. 5, the multicast packet path switching system includes a plurality of FEs 1A to 6A and a CE 7A.

  Each of the FEs 1A to 6A is directly or indirectly connected to each other, has a function of transferring a multicast packet, and performs processing classified as a U plane. Each of the FEs 1A to 6A transfers the multicast packet according to the transfer path determined by the CE 7A. Note that the number of FEs is not limited to six and may be a plurality.

  In this embodiment, FE3A and FE6A are connected to the distribution server 8, and FE1A is connected to the terminal 9 and the terminal 10. The FE connected to the distribution server 8 is not limited to FE3A and FE6A. Further, the FE connected between the terminal 9 and the terminal 10 is not limited to the FE1A.

  The CE 7A determines a transfer path (multicast packet transfer path) including at least one of the FEs 1A to 6A by calculating a transfer path. The CE 7A controls the transfer in the FE included in the transfer path based on the transfer path. The CE 7A performs processing classified into the C plane.

  FIG. 6 is a block diagram illustrating components common to the FEs 1A to 6A. In FIG. 6, FE1A is shown as an example. In FIG. 6, the same components as those shown in FIG.

  In FIG. 6, FE1A includes a specifying unit 1d and a notification unit 1e instead of the transmission unit 1c of FE1 shown in FIG.

  When the adjacent FE does not transmit a predetermined signal, the specifying unit 1d specifies the link with the FE as a failure location.

  For example, the specifying unit 1d transmits an HB signal to each of adjacent FEs, specifies an FE that does not transmit a response signal for the HB signal during a predetermined time thereafter, and specifies the specified FE. This link is identified as a failure location. Note that each of the specifying units 1d of the FEs 1A to 6A, when receiving the HB signal in a normal state, transmits a response signal to the HB signal to the source of the HB signal during a predetermined time thereafter. . The HB signal is an example of a failure detection signal, and the response signal to the HB signal is an example of a predetermined signal.

  The specifying unit 1d may transmit the HB signal to which the identifier of the FE to which the self belongs (FE1A in this case) and the identifier of the link with the adjacent FE are added to the adjacent FE. In this case, each of the FEs 1A to 6A has a function of identifying the identifier of the FE added to the HB signal and the identifier of the link.

  Further, when each specifying unit 1d of the FEs 1A to 6A is set to periodically transmit the HB signal to the adjacent FE, the specifying unit 1d selects the HB signal to be periodically transmitted. An adjacent FE that has not been transmitted may be detected as a FE in which a failure has occurred. In this case, the HB signal periodically transmitted is an example of a predetermined signal.

  The notification unit 1e notifies the CE 7 of failure information indicating the failure location specified by the specifying unit 1d. Note that the notification unit 1e may use, as the failure information, an identifier that can identify the link that is the failure location and the FE that has not transmitted the predetermined signal.

  FIG. 7 is a block diagram showing the CE 7A. In FIG. 7, the same components as those shown in FIG.

  In FIG. 7, the CE 7A includes a storage unit 7a, and includes an update unit 7c instead of the update unit 7b included in the CE 7 illustrated in FIG. The update unit 7c includes a reception unit 7c1 and an update control unit 7b2.

  When a failure occurs in the transfer path, the update unit 7c refers to the storage unit 7a (specifically, connection information and transfer path information), and recalculates the transfer path, thereby replacing the transfer path. The new transfer route is determined, and the table 1a1 stored in the FE included in the new transfer route is updated based on the new transfer route.

  The receiving unit 7c1 receives the failure information notified from the notification unit 1e.

  The update control unit 7c2 refers to the storage unit 7a and confirms whether the failure location indicated by the failure information received by the reception unit 7c1 is a location in the transfer path. If the failure location is a location in the transfer path, the update control unit 7c2 refers to the storage unit 7a, and the transfer route that bypasses the failure location as a new transfer path (that is, does not include the link where the failure occurred). Then, a transfer route for transferring the multicast packet from the distribution server 8 to the terminal 9 and the terminal 10 is determined.

  The update control unit 7c2 updates the table 1a1 included in the FE in the new transfer path based on the new transfer path so that the multicast packet is transferred through the new transfer path.

  Next, the operation will be described.

  FIG. 8 is a block diagram for explaining the operation of the multicast packet path switching system of the second embodiment.

  In FIG. 8, it is assumed that a failure has occurred in FE2A in a situation where multicast packets are transferred from distribution server 8 in the order of FE3A, FE2A, and FE1A and distributed to terminals 9 and 10. Note that the FE in which a failure occurs is not limited to FE2A.

  Prior to the occurrence of a failure in FE2A, the update control unit 7c2 in the CE 7A refers to the storage unit 7a in the CE 7A to calculate a transfer route, whereby the multicast packet from the distribution server 8 is converted into FE3A, FE2A, and FE1A. The table 1a1 included in the FE3A, FE2A, and FE1A in the transfer path is set through the C plane so that the multicast packet is transferred in the order of FE3A, FE2A, and FE1A.

  When the specific unit 1d in the FE 1A transmits an HB signal to the adjacent FE 2A in a situation where a failure has occurred in the FE 2A (step B1), a response signal to the HB signal is transmitted from the FE 2A for a predetermined time thereafter. It is not transmitted to the specific unit 1d in the FE 1A. For this reason, the specifying unit 1d in the FE 1A specifies the FE 2A as the FE in which the failure has occurred, and further specifies the link between the FE 1A and the FE 2A as the failure point (step B2).

  Subsequently, the notification unit 1e in the FE 1A notifies the CE 7A of failure information (link failure) indicating the FE 2A where the failure has occurred and the failure location (step B3).

  Subsequently, the reception unit 7c1 in the CE 7A receives the failure information from the notification unit 1e in the FE 1A (step B4).

  Subsequently, the update control unit 7c2 in the CE 7A refers to the storage unit 7a in the CE 7A and confirms whether the location (link) where the failure has occurred is a link in the transfer path. If the location (link) where the failure has occurred is a link in the transfer path, the update control unit 7c2 in the CE 7A refers to the storage unit 7a in the CE 7A, and recalculates the transfer path to create a new one. As a transfer route, a new transfer route (a transfer route for transferring multicast packets in the order of FE3, FE5, and FE1) that bypasses the failed link is determined (step B5).

  Subsequently, the update control unit 7c2 in the CE 7A allows the FE3, FE5, and FE1 in the new transfer path to pass through the C plane so that the multicast packet from the distribution server 8 is transferred in the order of FE3, FE5, and FE1. The included table 1a1 is updated (steps B6 and B7). In this case, the table 1a1 included in the FE 1A may not be updated. As an update method for the table 1a1, the update control unit 7c2 creates the tables 1a1 for FE3A, FE5A, and FE1A, and distributes the created tables 1a1 to the FE3A, FE5A, and FE1A, respectively. FE3A, FE5A, and FE1 may update the table 1a1.

  According to the present embodiment, when a failure occurs in the transfer path, the update unit 7c refers to the inside of the storage unit 7a, and recalculates the transfer path, thereby replacing the transfer path with a new transfer path. And the table 1a1 included in the FE included in the new transfer path is updated based on the new transfer path.

  For this reason, similarly to the first embodiment, it is possible to shorten the transfer path switching time.

  In the present embodiment, the identifying unit 1d identifies an FE that does not transmit a predetermined signal among adjacent FEs as a failed FE, and identifies a link with the identified FE as a failed part. . For example, the identifying unit 1d identifies an FE that has transmitted an HB signal to each of the adjacent FEs and has not transmitted a response signal to the HB signal during a predetermined time after that as an FE in which a failure has occurred. Alternatively, an FE that does not transmit an HB signal that is transmitted periodically is identified as an FE in which a failure has occurred, and a link with the identified FE is identified as a failure location.

  In this case, the CE 7A does not have to detect the failure of the FE itself, and the processing load on the CE 7A can be reduced.

  FIG. 9 is a block diagram showing an application example of the multicast packet path switching system of the first and second embodiments. In the example shown in FIG. 9, in the U plane, there are a plurality of FEs belonging to the backbone subnetwork and a plurality of FEs belonging to the metro subnetwork connected to the backbone subnetwork. In the C plane, there are SCEs provided for each sub-network and NCEs for controlling each SCE.

  In each embodiment described above, the illustrated configuration is merely an example, and the present invention is not limited to the configuration.

1-6, 1A-6A FE
7, 7A CE
8 distribution server 9, 10 terminal 1a storage unit 1a1 routing information table 1b transfer unit 1b1 input unit 1b2 output unit 1b3 packet transfer unit 1c transmission unit 1d identification unit 1e notification unit 7a storage unit 7b, 7c update unit 7b1 detection unit 7c1 reception unit 7b2, 7c2 Update control unit

Claims (10)

A plurality of transfer devices having a function of transferring a multicast packet; and a control device that determines a transfer path including at least one of the plurality of transfer devices;
The transfer device is
A storage unit for storing a table in which header information of a multicast packet and output destination information indicating an output destination of the multicast packet are associated with each other based on the transfer path;
When the multicast packet is received, the storage unit is referred to, and a transfer unit that transfers the multicast packet to the output destination indicated by the output destination information associated with the header information of the multicast packet includes:
The controller is
A storage unit that stores connection information indicating a connection relationship between the transfer devices and transfer path information indicating the transfer path;
When a failure occurs in the transfer path, the storage unit is referred to determine a new transfer path that replaces the transfer path, and the table stored in the transfer device in the new transfer path is stored in the new transfer path. A multicast packet path switching system, comprising: an updating unit configured to update based on a simple transfer path. The multicast packet path switching system according to claim 1, wherein
The update unit
A detecting unit that detects a transfer device that does not transmit a predetermined signal among the plurality of transfer devices as a transfer device in which a failure has occurred;
If the transfer device in which the failure has occurred is a transfer device in the transfer route, the transfer path refers to the storage unit and includes a transfer device that is different from the transfer device in which the failure has occurred as the new transfer route And an update control unit that updates a table stored in the transfer device in the new transfer path based on the new transfer path. The multicast packet path switching system according to claim 1, wherein
The transfer device is
When an adjacent transfer device does not transmit a predetermined signal, a specifying unit that specifies a link with the adjacent transfer device as a failure location,
A notification unit for notifying the control device of failure information indicating the failure location, and
The update unit
A receiver for receiving the failure information;
If the failure location indicated by the failure information is a location in the transfer path, the transfer section is referred to as the new transfer route, and a transfer route that bypasses the failure location is determined as the new transfer route. A multicast packet path switching system, comprising: an update control unit that updates a table stored in a transfer apparatus in a path based on the new transfer path. In the multicast packet path switching system according to claim 2,
The detection unit transmits a failure detection signal to the plurality of transfer devices,
The multicast packet path switching system, wherein the transfer device includes a transmission unit that transmits, as the predetermined signal, a response signal to the failure detection signal to a transmission source of the failure detection signal. In the multicast packet path switching system according to claim 2,
The multicast packet path switching system, wherein the transfer device includes a transmission unit that periodically transmits to the control device a signal to be transmitted to the control device as the predetermined signal. In the multicast packet path switching system according to claim 3,
When the specific unit transmits a failure detection signal to an adjacent transfer device and receives the failure detection signal from the adjacent transfer device, the specific unit responds to the failure detection signal as the predetermined signal. A multicast packet path switching system, which transmits a response signal to a transmission source of the failure detection signal. In the multicast packet path switching system according to claim 3,
The multicast packet path switching system, wherein the specifying unit transmits a signal to be periodically transmitted to an adjacent transfer device as the predetermined signal to the adjacent transfer device. Multicast packet path switching method in a multicast packet path switching system, comprising: a plurality of transfer apparatuses having a function of transferring multicast packets; and a control apparatus for determining a transfer path including at least one of the plurality of transfer apparatuses Because
A storage step of storing, in the storage unit, a table in which the transfer device associates the header information of the multicast packet and the output destination information indicating the output destination of the multicast packet based on the transfer path;
When the forwarding device receives the multicast packet, the forwarding step refers to the storage unit and forwards the multicast packet to the output destination indicated by the output destination information associated with the header information of the multicast packet;
A storage step in which the control device stores connection information indicating a connection relationship between the transfer devices and transfer route information indicating the transfer route in a storage unit;
When a failure occurs in the transfer path, the control device refers to the storage unit to determine a new transfer path to replace the transfer path, and stores the transfer path in the new transfer path. An update step of updating the table based on the new transfer route. In the multicast packet path switching method according to claim 8,
The updating step includes
A detection step of detecting a transfer device that does not transmit a predetermined signal among the plurality of transfer devices as a transfer device in which a failure has occurred;
If the transfer device in which the failure has occurred is a transfer device in the transfer route, the transfer path refers to the storage unit and includes a transfer device that is different from the transfer device in which the failure has occurred as the new transfer route And an update control step of updating a table stored in the transfer device in the new transfer path based on the new transfer path. In the multicast packet path switching method according to claim 8,
When the transfer device does not transmit a predetermined signal by an adjacent transfer device, a specifying step for specifying a link with the adjacent transfer device as a failure point;
The transfer device further includes a notification step of notifying the control device of failure information indicating the failure location,
The updating step includes
Receiving the failure information; and
If the failure location indicated by the failure information is a location in the transfer path, the transfer section is referred to as the new transfer route, and a transfer route that bypasses the failure location is determined as the new transfer route. An update control step of updating a table stored in a transfer device in a path based on the new transfer path, and a multicast packet path switching method.

Images