Communication method in multicast network and multicast network

Abstract

To provide a communication method in a multicast network, which can reduce the time during which multicast communication is interrupted when a topology change occurs, and the multicast network.SOLUTION: A communication method includes a step in which when a failure occurs in a path between two L2 switches, a multicast network sets a predetermined bit in a predetermined frame to a predetermined value; a step in which in response to the predetermined bit being set to the predetermined value, the multicast network returns a logically blocked path to a connected state; and a step in which in response to the predetermined bit being set to the predetermined value, a first L2 switch transmits a first packet to the network separately from periodic transmission.SELECTED DRAWING: Figure 14

Description

[Technical field]

[0001] The present disclosure relates to a communication method in a multicast network and a multicast network. [Background technology]

[0002] In a network, a switching hub in the data link layer (Layer 2) is sometimes called an L2 switch. A multicast network is known in which multicast communication is performed in a network in which multiple L2 switches are configured in a ring shape. In a multicast network, if a communication path failure occurs somewhere in the ring network, the communication path may be cut off depending on the positions of the multicast sending and receiving devices, and multicast communication may not be possible for an extended period of time.

[0003] To address such problems, in Patent Document 1, each router has alternative route information for the distribution route of multicast traffic, and when a link failure occurs upstream of the own router on a distribution route that passes through the own router, based on the alternative route information, transmits a distribution route change notification to at least one router on the alternative route, and in response to the transmission or reception of the distribution route change notification, executes processing to change the distribution route within the own router so that the part of the distribution route related to the link failure is replaced with an alternative route without waiting for an update of the multicast routing table based on the multicast routing protocol. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2012-199689 A Summary of the Invention [Problem to be solved by the invention]

[0005] However, the method described in Patent Document 1 involves complex processes such as obtaining an alternative route and switching to the alternative route.

[0006] Therefore, an object of the present disclosure is to provide a communication method in a multicast network and a multicast network that can reduce the time during which multicast communication is interrupted when a topology change occurs. [Means for solving the problem]

[0007] The communication method in a multicast network in which a plurality of L2 switches are arranged in a ring shape according to the present disclosure includes the steps of: the multicast network logically setting a path between two specified L2 switches to a blocked state; the first L2 switch transmitting a first packet to the multicast network at a fixed first period; a receiving device that desires multicast data transmitting a second packet toward the first L2 switch through one or more intermediate L2 switches; the transmitting device transmitting multicast data to the receiving device through the one or more intermediate L2 switches and the first L2 switch; when a failure occurs in the path between two L2 switches, the multicast network setting a specified bit in a specified frame to a specified value; in response to the specified bit being set to the specified value, the multicast network returning the logically blocked path to a connected state; and in response to the specified bit being set to the specified value, the first L2 switch transmitting a first packet to the multicast network separately from the periodic first period transmission. Effect of the Invention

[0008] According to the present disclosure, it is possible to reduce the time during which multicast communication is interrupted when a topology change occurs. [Brief description of the drawings]

[0009] [Figure 1]FIG. 1 is a diagram for explaining locations where communication is cut off in a multicast network under normal conditions. [Diagram 2] FIG. 1 is a diagram for explaining a topology change in a multicast network. [Diagram 3] 1A is a diagram showing a BPDU (Bridge Protocol Data Unit) frame, FIG. 1B is a diagram showing the structure of the data portion of a BPDU, and FIG. 1C is a diagram showing the structure of a 1-byte flag. [Figure 4] FIG. 13 is a diagram showing the transmission of a query packet. [Diagram 5] FIG. 13 is a diagram showing the transmission of a Join packet. [Figure 6] FIG. 13 is a diagram showing the transfer of multicast data. [Figure 7] FIG. 13 illustrates a communication failure during the transfer of multicast data. [Figure 8] A diagram showing reconnection of shared and intentionally cut paths in a network with TC bit=1. [Figure 9] FIG. 13 is a diagram showing the transmission of a query packet. [Figure 10] FIG. 13 is a diagram showing the transmission of a Join packet. [Figure 11] FIG. 13 is a diagram showing the transfer of multicast data. [Figure 12] FIG. 1 is a diagram showing the timing of retransmission of multicast data when a communication failure occurs in a conventional method. [Figure 13] 11 is a diagram showing the timing at which multicast data is retransmitted when a communication failure occurs in an embodiment. FIG. [Figure 14] 1 is a flowchart showing a procedure of a communication method in a multicast network according to an embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] Hereinafter, an embodiment of the invention will be described with reference to the drawings. First, the operation of a multicast network will be described.

[0011] As shown in Figures 1, 2, 3, 4, 6, 7, 8, 9, 10 and 11, a plurality of L2 switches 1 to 8 form a ring-shaped multicast network. An example will be described in which a multicast sending device 14 is connected to the L2 switch 1, and a multicast receiving device 12 is connected to the L2 switch 4. Each of the L2 switches 1 to 8 includes a processor and a memory. The processor executes the communication processing described below.

[0012] In this network, control is performed using STP (Spanning Tree Protocol) to prevent broadcast storms from occurring.

[0013] That is, as shown in Fig. 1, the network logically sets one of the connections between two adjacent L2 switches to a communication cutoff state. In the example of Fig. 1, the communication cutoff state is set between L2 switch 6 and L2 switch 7.

[0014] Then, as shown in Figure 2, if a communication failure such as a cable break occurs at a connection point other than the connection point set to the communication cutoff state, the L2 switch connected to the route where the failure occurred sets the TC bit of the STP BPDU frame to 1, and the TC bit = 1 is shared within the network.

[0015] Fig. 3(a) is a diagram showing a BPDU (Bridge Protocol Data Unit) frame. In Fig. 3(a), the numbers showing the data length indicate the number of bytes.

[0016] Fig. 3(b) shows the structure of the data portion of a BPDU. In Fig. 3(b), the number showing the data length also shows the number of bytes. The data includes a one-byte flag.

[0017] 3C is a diagram showing the configuration of a 1-byte flag, which includes a Topology Change (TC) bit in the first bit.

[0018] After that, the network secures the communication path by returning the connection points that were set to the communication cut-off state to a connected state, as shown in Figure 2. At this point, the topology change is complete.

[0019] Next, a communication procedure in a multicast network will be described. As shown in Fig. 4, one of the multiple L2 switches is selected as the querier. In the example of Fig. 4, L2 switch 6 and L2 switch 7 are set to be logically disconnected, and L2 switch 5 is set as the querier.

[0020] The L2 switch 5, which acts as a querier, periodically (typically at 125-second intervals) sends out packets called queries (hereafter referred to as query packets) into the network and monitors whether there are any devices (hereafter referred to as receiving devices) requesting multicast data.

[0021] 5, receiving device 12 connected to L2 switch 4 replies to the querier (L2 switch 5) with a packet called Join (hereinafter, Join packet) in response to a query packet received from the querier, indicating its intention to receive data. L2 switch 5, which is the querier, and L2 switch 4, L2 switch 3, L2 switch 2, and L2 switch 1 that exist between the querier (L2 switch 5) and receiving device 12 learn and store the route when a Join packet arrives from receiving device 12.

[0022] 6, transmitter 14 connected to L2 switch 1 transmits multicast data to receiver 12 connected to L2 switch 4 using the learned and stored route. That is, the multicast data transmitted from transmitter 14 is sent to receiver 12 via L2 switch 1, L2 switch 2, L2 switch 3, and L2 switch 4. The multicast data is also sent to L2 switch 5, which is the querier.

[0023] As shown in FIG. 7, it is assumed that a fault such as a cable breakage occurs between L2 switch 2 and L2 switch 3 while multicast data is being transferred.

[0024] As shown in Fig. 8, the TC bit of the BPDU (Bridge Protocol Data Unit) is set to "1" and the TC bit = 1 is shared within the network. Through control by STP, the communication connection between L2 switch 6 and L2 switch 7, which was intentionally cut off, is restored and a path is secured.

[0025] Thereafter, conventionally, as shown in FIG. 9, a query packet is periodically transmitted from the L2 switch 5 which is a querier.

[0026] 10, receiving device 12 connected to L2 switch 4 replies to the querier (L2 switch 5) with a Join packet in response to the query packet received from the querier, indicating its intention to receive data. L2 switch 5, which is the querier, and L2 switches 4, 8, 7, and 6 that exist between the querier (L2 switch 5) and receiving device 12 learn and store the route when the Join packet arrives from receiving device 12.

[0027] 11, transmitter 14 connected to L2 switch 1 transmits multicast data to receiver 12 connected to L2 switch 4 using the learned and stored route. That is, the multicast data transmitted from transmitter 14 is sent to L2 switch 5, which is the querier, and then passed through L2 switch 6, L2 switch 7, L2 switch 8, and L2 switch 4 to receiver 12.

[0028] FIG. 12 is a diagram showing the timing of retransmission of multicast data when a communication failure occurs in the conventional method.

[0029] As shown in FIG. 12, the transmission interval of a query packet is ΔT (=125 seconds). Assume that a communication failure occurs at time t1. Then, in response to TC bit=1, the blocked route is reconnected at time t2, and the topology change is completed. Then, at the regular transmission time t3 of the query packet, the query packet is retransmitted. Then, at time t4, the receiving device 12 connected to the L2 switch 4 retransmits the Join packet. Then, at time t5, the multicast data is retransmitted. If the time interval dt between time t1 and time t3 is short, the time CT (=t5-t1) during which communication is interrupted is short, but if the time interval dt between time t1 and time t3 is long, the time CT (=t5-t1) during which communication is interrupted is long. The length of the time interval dt cannot be controlled.

[0030] In this embodiment, in addition to reconnecting the blocked route, the querier also sends a query packet in response to the TC bit being set to 1. This makes it possible to shorten the time that communication is interrupted compared to the conventional method of only periodically sending query packets.

[0031] FIG. 13 is a diagram showing the timing at which multicast data is retransmitted when a communication failure occurs in this embodiment.

[0032] As shown in FIG. 13, the transmission interval of a query packet is ΔT (=125 seconds). Assume that a communication failure occurs at time t1. Thereafter, in response to TC bit=1, the blocked route is reconnected at time t2, completing the topology change. Without waiting for the regular transmission time t3 of the query packet, in response to TC bit=1, at time t31, the querier (e.g., L2 switch 5) retransmits the query packet. The query packet is transmitted multiple times at time intervals of ΔS (=1 second), for example, 10 times from time t31 to t40. The period ΔS is shorter than the period ΔT. Preferably, the period from the start of retransmission of the query packet to the end of retransmission ((number of retransmissions-1)×period ΔS) is shorter than the period ΔT.

[0033] Thereafter, when a query packet is received by the L2 switch 4 after a failure occurs, the receiving device 12 connected to the L2 switch 4 retransmits a Join packet at time t4'. Then, at time t5', multicast data is retransmitted. Since the query packet is transmitted at time t31 without waiting for the regular transmission time t3, the time CT' (= t5' - t1) during which communication is interrupted can be made shorter than the conventional time CT (= t5 - t1) during which communication is interrupted. The receiving device 12 may be configured to immediately retransmit a Join packet when a query packet is received by the connected L2 switch, for example, the L2 switch 4, after a failure occurs.

[0034] 13, time t31 is set to be later than time t2 (topology change completion time), so the query packet is transmitted after the topology change is completed. In this case, the query packet is transmitted normally via the new route.

[0035] If time t31 is earlier than t2, the query packet sent at time t31 will not be sent over the new route. However, among the times t32 to t40 when one or more query packets are sent thereafter, the query packet will be sent normally over the new route at a time after t2. Therefore, there is no problem if the querier starts sending query packets in response to TC bit=1 before the topology change is completed and the blocked route is reconnected (before time t2).

[0036] FIG. 14 is a flowchart showing the procedure of a communication method in a multicast network according to the embodiment.

[0037] In step S1, the multicast network sets the first L2 switch (L2 switch 5) as a querier.

[0038] In step S2, the multicast network logically sets one of the connections between two adjacent L2 switches to a communication cutoff state. That is, two predetermined L2 switches (L2 switches 6 and 7) logically set the path between the two L2 switches to a communication cutoff state.

[0039] In step S3, the first L2 switch (L2 switch 5) set as a querier transmits a first packet (query packet) to the network at a fixed period ΔT. The other L2 switches (L2 switches 1 to 4, L2 switches 6 to 8) receive the query packet, learn the route along which the query packet was forwarded, and store it. The period ΔT is an example of the first period.

[0040] In step S4, a receiving device 12 connected to an L2 switch of a multicast network that desires multicast data transmits a second packet (Join packet) to a first L2 switch (L2 switch 5) through one or more relay L2 switches (L2 switches 4, 3, 2, 1). The one or more relay L2 switches (L2 switches 4, 3, 2, 1) receive the second packet, and learn and store the route along which the second packet (Join packet) has been forwarded.

[0041] In step S5, sending device 14 connected to an L2 switch in the multicast network transmits multicast data to receiving device 12 via one or more relay L2 switches (L2 switches 1, 2, 3, 4) and a querier (L2 switch 5).

[0042] If it is detected in step S6 that a communication failure has occurred on the path between two L2 switches (L2 switches 6 and 7), the process proceeds to step S7.

[0043] In step S7, the multicast network (specifically, the L2 switch (L2 switch 6 or 7) that detected the communication failure) sets a predetermined bit (TC bit) of a predetermined frame (BPDU) to a predetermined value (1).

[0044] In step S8, in response to a predetermined bit (TC bit) being set to a predetermined value (1), the multicast network (specifically, two predetermined L2 switches (L2 switches 6 and 7)) logically restores a blocked route to a connected state.

[0045] In step S9, in response to the predetermined bit (TC bit) being set to a predetermined value (1), the first L2 switch (L2 switch 5) set as the querier transmits a first packet (query packet) multiple times (e.g., 10 times) at a period ΔS (=1 second) in addition to the periodic transmission. The other L2 switches (L2 switches 1 to 4, L2 switches 6 to 8) receive the query packet, and learn and store the route through which the query packet has been forwarded. The period ΔS is an example of the second period.

[0046] In step S10, a receiving device 12 that desires multicast data retransmits a second packet (Join packet) to the first L2 switch (L2 switch 5) through one or more intermediate L2 switches (L2 switches 4, 8, 7, 6). The one or more intermediate L2 switches (L2 switches 4, 8, 7, 6) receive the second packet, and learn and store the route along which the second packet (Join packet) has been forwarded.

[0047] In step S11, sending device 14 retransmits the multicast data toward receiving device 12 via one or more relay L2 switches (L2 switches 6, 7, 8, 4) and the querier (L2 switch 5).

[0048] According to this embodiment, the L2 switch selected as the querier immediately transmits a query packet into the network when the TC bit of the BPDU frame becomes 1. Once the query packet is transmitted, the join packet from the receiving device arrives via a new route, and the L2 switch can re-learn the route for forwarding multicast data to the receiving device, and can then forward the multicast data.

[0049] This allows routes to be immediately re-learned regardless of when or where a topology change occurs, thereby shortening the time that multicast communication is interrupted when a topology change occurs compared to the conventional method and improving system stability.

[0050] The embodiments disclosed herein should be considered to be illustrative and not restrictive in all respects. The scope of the present invention is defined by the claims, not the above description, and is intended to include all modifications within the meaning and scope of the claims. [Explanation of symbols]

[0051] 1~8 L2 switches, 12 receiving equipment, 14 transmitting equipment.

Claims

1. A communication method in a multicast network in which a plurality of L2 switches are configured in a ring shape, comprising: A step in which the multicast network sets a path between two predetermined L2 switches to a logically blocked state; A first L2 switch transmits a first packet to the multicast network at a first fixed period; A receiving device that desires multicast data transmits a second packet to the first L2 switch through one or more intermediate L2 switches; A sending device transmits multicast data to the receiving device through one or more intermediate L2 switches and the first L2 switch; setting a predetermined bit of a predetermined frame to a predetermined value by the multicast network when a failure occurs in a path between two L2 switches; in response to the predetermined bit being set to the predetermined value, the multicast network restores a logically disconnected route to a connected state; A communication method in a multicast network comprising: a step in which, in response to the specified bit being set to the specified value, the first L2 switch transmits the first packet to the multicast network separately from periodic transmissions of the first cycle.

2. 2. The communication method in a multicast network according to claim 1, wherein the step of transmitting the first packet to the multicast network separately from the transmission of the first period includes a step of the first L2 switch transmitting the first packet to the multicast network a plurality of times.

3. 3. The communication method in a multicast network according to claim 2, wherein the step of transmitting the first packet to the multicast network separately from the transmission at the first period includes a step of the first L2 switch transmitting the first packet to the multicast network multiple times at a second period shorter than the first period.

4. 2. The communication method in a multicast network according to claim 1, wherein the multicast network is controlled by STP, the first L2 switch is a querier, the first packet is a query packet, the second packet is a join packet, the specified frame is a BPDU frame, the specified bit is a TC bit, and the value of the specified bit is 1.

5. A multicast network that performs the communication method according to any one of claims 1 to 4.

Images