CN102035746A - Point-to-multipoint multicast method and device - Google Patents

Abstract

The embodiment of the invention discloses a point-to-multipoint multicast method and device. Taking the implementation of the method as an example: it includes: sending the data sent by the sending end to the receiving end in the first multicast group; the number of the receiving ends is two or more; if the first receiving end in the receiving end end breaks down, then delete the first receiving end from the first multicast group; if the first receiving end returns to normal, then establish a second multicast group and add the first receiving end to the The first multicast group and the second multicast group; send the data sent by the sending end before the first receiving end fails to recover to the first receiving end through the second multicast, and send the data to the first receiving end The end deletes the second multicast group after sending the data.

Description

A point-to-multipoint multicast method and device

technical field

The invention relates to the technical field of communication, in particular to a point-to-multipoint multicast method and device.

Background technique

In the router communication system, a point-to-multipoint communication mechanism is used. When using multicast to realize point-to-multipoint data communication, it is generally necessary to maintain a one-to-many multicast group, which can be saved in the multicast group. The name, address and other information of the receiving member of the data, the sender sends the same data for each receiving member. All members in the multicast group send data that was not received by the faulty member before the faulty member recovers, thus affecting the sending efficiency of the entire multicast. How to ensure that each receiving member receives data as quickly as possible, that a faulty receiving member affects other receiving members as little as possible, and at the same time take into account the rapid recovery of data transmission after the faulty receiving member is removed becomes the key technology of multicast data.

Take the communication between router boards as an example: Assume that the control board performs data communication with four forwarding boards (can be numbered 1~4), and the control board wants to send data to the four forwarding boards, so it is necessary to maintain a In the process of multicast transmission, when forwarding board 1 fails, how to minimize the impact on the other three forwarding boards, and at the same time, how to recover as soon as possible after forwarding board 1 is removed from the fault is a technical problem that needs to be solved . Usually, the following scheme can be adopted: isolate the faulty member, and re-establish a multicast group for the faulty member after the fault is resolved; the specific implementation is: after the forwarding board 1 fails, the forwarding board 1 is isolated, and the control board only maintains the multicast group of three members ;When the fault of forwarding board 1 is removed, the control board reassigns the multicast group to it. At this time, the control board maintains two multicast groups. In the new multicast group, only forwarding board 1 is the member, and the original multicast group The members inside have forwarding boards 2, 3, and 4. With this solution, the number of multicast groups will increase as the forwarding board fails, and the number of multicast groups is equal to the number of members. Too many multicast groups will waste system memory and reduce system performance.

Usually, in order to reduce the number of multicast groups, you can isolate the faulty member, and add the faulty member back to the multicast group after the fault is resolved. After the faulty member recovers, the data generated during the faulty process needs to be sent to the faulty member again, or at some point, for example, when the forwarding board is powered off and restarted, and the memory data disappears, all data of the faulty member needs to be sent to the faulty member again. There are usually two solutions to this technical problem:

Solution 1: The sender records the fault point, and the members who have resolved the fault first synchronize data. All the data that the sender has sent before the fault point is sent to the above-mentioned members that have resolved the fault, and then the normal multicast transmission is performed. Solution 1 When the amount of data is large, there are many faulty members, or frequent faults occur, the non-faulty members need to wait for the data synchronization of the faulty members to complete when the non-faulty members receive new data, which has a great impact on performance. For example: After the failure of forwarding board 1 is resolved, rejoin the multicast group. At this time, the members of the multicast group are forwarding boards 1, 2, 3, and 4, but the control board needs to synchronize data for forwarding board 1 first, and other members need to wait. , after the data synchronization is completed, send the normal multicast.

Solution 2: Resend all data at the sender according to the multicast group regardless of whether it is a faulty member or not. In this solution, a large amount of data is frequently and repeatedly sent, and normal members receive repeated data, which seriously affects system performance. When a forwarding board has faults such as flashing off or frequent power failures, the sending end is busy sending data non-stop, while the forwarding board that has not failed is also frequently receiving duplicate data, making normal data unable to be sent.

To sum up, the adoption of the above technical solution has the problems of a large number of maintenance multicast groups and a large amount of repeated data transmission, so the multicast efficiency is low and the performance of the multicast system is low.

Contents of the invention

The technical problem to be solved by the embodiments of the present invention is to provide a point-to-multipoint multicast method and device to improve multicast efficiency and multicast system performance.

In order to solve the above-mentioned technical problems, the embodiment of the point-to-multipoint multicast method provided by the present invention can be realized through the following technical solutions:

Sending the data sent by the sending end to the receiving end in the first multicast group; the number of receiving ends is two or more;

If the first receiving end among the receiving ends fails, deleting the first receiving end from the first multicast group;

If the first receiving end returns to normal, then establish a second multicast group and add the first receiving end to the first multicast group and the second multicast group;

Sending the data sent by the sending end before the failure recovery of the first receiving end to the first receiving end through the second multicast, and deleting the second multicast group after sending the data to the first receiving end.

A point-to-multipoint multicast device, comprising:

The sending unit is used to send the data sent by the sending end to the receiving end in the first multicast group; the number of receiving ends is two or more; sending the data to the first receiving end through the second multicast;

A multicast group control unit, configured to delete the first receiving end from the first multicast group if the first receiving end of the receiving ends fails; if the first receiving end returns to normal, Then establish a second multicast group and add the first receiving end to the first multicast group and the second multicast group; delete the second multicast group after sending data to the first receiving end.

The above technical solution has the following beneficial effects: in the process of data synchronization in the embodiment of the present invention, the number of multicast groups to be maintained is reduced by managing the multicast groups; the data received by the normal receiving end is synchronized with the faulty receiving end The data is transmitted at the same time, which eliminates the time for the normal receiving end to wait for the faulty receiving end to recover, and also eliminates the problem of repeatedly sending data to the normal receiving end; thus improving the multicast efficiency and multicast system performance.

Description of drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the accompanying drawings that need to be used in the description of the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For Those of ordinary skill in the art can also obtain other drawings based on these drawings without any creative effort.

Fig. 1 is a schematic flow chart of the method of the embodiment of the present invention;

FIG. 2 is a schematic diagram of a router multicast system according to an embodiment of the present invention;

Fig. 3 is the schematic diagram of the device structure of the embodiment of the present invention;

Fig. 4 is the schematic diagram of the device structure of the embodiment of the present invention;

Fig. 5 is a schematic structural diagram of a device according to an embodiment of the present invention.

Detailed ways

The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

The embodiment of the present invention provides a point-to-multipoint multicast method. The method execution subject of this embodiment can be an entity with a multicast control function, such as a router, a switch, etc., as shown in FIG. 1 , including:

101: Send the data sent by the sending end to the receiving end in the first multicast group; the number of the above receiving ends is two or more;

It should be noted that the expressions "first" and "second" used in the embodiments of the present invention are only for distinguishing multiple identical receiving terminals, time, etc., and should not be construed as limiting the embodiments of the present invention.

102: Delete the first receiving end from the first multicast group if a faulty first receiving end occurs among the receiving ends.

At this time, the sending end still sends data to the receiving end in the first multicast group according to the original sending time sequence.

103: If the above-mentioned first receiving end returns to normal, establish a second multicast group and add the first receiving end to the first multicast group and the second multicast group.

In order to solve the frequent operation of multicast group establishment and deletion caused by the frequent failure and recovery of the receiving end, delay suppression measures can also be set, that is, the above-mentioned first receiving end returning to normal includes: judging and determining after the set first time arrives The above-mentioned first receiving end has returned to normal. From the operations performed above, it can be seen that even if the first receiving end has recovered during the above first period of time, the operation of restoring the normal reception of the first receiving end is not performed immediately, which can be regarded as the first receiving end has performed isolation.

104: Send the data sent by the sending end before the above-mentioned first receiving end fails to recover to the above-mentioned first receiving end through the above-mentioned second multicast;

The data sent by the sending end before the failure recovery of the first receiving end may only include the data sent by the sending end between when the first receiving end fails and when the failure recovers. At some point, for example, when the first receiving end is powered off and restarted, and the memory data disappears, it may also include all data sent by the sending end from the time the sending end sends data to when the first receiving end recovers from failure.

Further, in order to restore the normal sending of the first receiving end and delete the second multicast group more quickly, the sending end may transmit data to the terminals in the first multicast group and the second receiving end in the second multicast group , send data to the second receiving end preferentially.

105: After sending the data to the first receiving end, delete the above-mentioned second multicast group.

The embodiment of the present invention also provides a processing method for continuous and frequent oscillation of the receiving end: if the fault frequency of the first receiving end reaches a set value, the value of the first time in the above 103 can be increased.

The embodiment of the present invention also provides an implementation method in which multiple receivers do not return to normal in the same time period: the above-mentioned process of sending the data sent by the sender before the failure of the first receiver recovers to the first receiver through the second multicast group , if the second receiving end among the above receiving ends returns to normal, one of the following methods can be adopted:

1. Add the above-mentioned second receiving end to the above-mentioned second multicast group, and then send the data sent by the sending end in the third time period to the above-mentioned first receiving end and the second receiving end through the above-mentioned second multicast group; The three time periods are: the earlier failure time point of the failure time point of the first receiving end and the failure time point of the second receiving end to the time point of recovery of the failure of the second receiving end.

2. After the above-mentioned deletion of the above-mentioned second multicast group, re-establish the second multicast group and add the above-mentioned second receiving end to the above-mentioned first multicast group and the above-mentioned second multicast group, and restore the second receiving end The data sent by the sending end is sent to the second receiving end through the second multicast group.

3. Establish a third multicast group and add the above-mentioned second receiving end to the above-mentioned first multicast group and the above-mentioned third multicast group, and pass the data sent by the sending end before the above-mentioned second receiving end fails to recover through the above-mentioned third group broadcast to the above-mentioned second receiving end, and delete the above-mentioned third multicast group after sending data to the second receiving end.

In the process of data synchronization by the method of the embodiment of the present invention, there is a reduction in the number of multicast groups that need to be maintained through the management of the multicast group; the normal receiving end receives data and the faulty receiving end synchronizes data simultaneously, eliminating It reduces the time for the normal receiving end to wait for the faulty receiving end to recover, and also eliminates the problem of repeatedly sending data to the normal receiving end; thereby improving the multicast efficiency and multicast system performance.

The following will take a router communication system as an example for specific illustration. The router communication system is not the only scenario where the multicast method is used, and this example should not be construed as a limitation to the embodiment of the present invention.

As shown in Figure 2, there is one control board 201 and four forwarding boards 202 in the system. The control board 201 needs to send the data in the board to the four forwarding boards 202 through data communication. The communication is sent by multicast, and the control board needs Maintain a multicast group A, including four member forwarding boards 1, 2, 3, and 4.

When the forwarding board 1 fails, the control board will delete the forwarding board 1 from the multicast group A, and can store the relevant information of the forwarding board 1 in the buffer area, waiting to be added to the newly created multicast group B; at this time, the multicast group A There are three members of forwarding boards 2, 3, and 4, and the data of multicast group A is still sent as usual.

At this time, you can start a 10ms timer to delay (10ms is the empirical data in this example, and different scenarios can use different durations). Forwarding board 1 joins multicast group A and multicast group B.

At this time, there are four members of forwarding boards 1, 2, 3, and 4 in multicast group A, and the data sender continues to normally send data to the forwarding boards in multicast group A through multicast group A; multicast group B ( Faulty member group) only includes the member whose fault has been resolved (that is, the forwarding board 1). At this time, the multicast group B can perform data synchronization on the forwarding board 1, that is, the data that the sender has sent before the fault of the forwarding board 1 is restored. Send to forwarding board 1. It should be noted that: after the faulty forwarding board recovers, forwarding board 1 will receive the data of the two multicast groups in a short period of time, but after the data synchronization is completed, the multicast group B will be deleted, and the multicast group B will be deleted. After being deleted, forwarding board 1 will only receive the data of multicast group A. In this example, multicast group A can be regarded as a real-time multicast group, and multicast group B can be regarded as a smooth multicast group. In addition, for forwarding board 1, both multicast group A and multicast group B will send data, and the data sent in multicast group A is the latest data, while the data sent in multicast group B may need to be synchronized The data should be old data. If multicast group B is not deleted in time, some latest data may be sent. In addition, in order to complete the data synchronization operation and delete the multicast group B as soon as possible, the data delivery of the multicast group B can be guaranteed first.

Delete multicast group B after data synchronization of multicast group B is completed. So far, the entire data synchronization process is completed.

In the above example corresponding to Figure 1, the timer can be started when the forwarding board fails. Before the timer expires (that is, before the first time arrives), the forwarding board fault recovery only records the recovery status and does not perform data synchronization processing. Perform data synchronization. If the timer expires and the forwarding board has not recovered, it means that there is no need for delay suppression, so the fault recovery of the forwarding board exceeding 10ms can be processed immediately; the assumption based on the above example is that forwarding board 1 can return to normal within 10ms, 10ms If internal forwarding board 1 fails to return to normal, multicast group B can be established immediately when forwarding board 1 returns to normal, and then forwarding board 1 can be added to multicast group B and multicast group A, and multicast group B can Forwarding board 1 synchronizes all data, and deletes multicast group B after the synchronization is completed.

In addition, it is also possible that the forwarding board fails repeatedly. In the above example, when the forwarding board 1 fails for the second time (this is considered to be multiple failures in a short period of time, for example, the forwarding board 1 has The second failure occurs within 30 seconds after the first failure; in the implementation process, it is necessary to record the state change of the forwarding board within 30 seconds, and use this as the basis for fault isolation), basically the same operation as above, with two differences , one is: delete forwarding board 1 from multicast group A and delete it from multicast group B at the same time, another point is that the timer length becomes longer, set it to 100ms. The third failure is set to 1000ms (the empirical data in this example, the fault isolation time is 10ms^the number of failures)...and so on. In the above implementation method, a fault recovery attenuation is set, and each time a fault occurs on the forwarding board, the isolation time is extended to make the faulty board recover slowly, so as to avoid the system being busy due to frequent faults. From the above operations, it can be seen that even if the forwarding board 1 has recovered within a certain period of time, the data transmission recovery operation for the forwarding board 1 is not immediately performed. This period of time can be regarded as an isolation of the forwarding board 1 .

It should be noted that the above-mentioned attempt to increase the duration of the timer does not require an infinite number of attempts, that is, the number of isolation times does not need to be infinite, and the number of isolation times can be set to a maximum value. If the above-mentioned specified time (30 seconds) exceeds The maximum number of isolation times (the empirical data in this example is 4 times, that is, 10ms/100ms/1000ms/10000ms has been tried four times in total, and it took more than 11 seconds to try, plus system loss, within 30 seconds), then it is considered that the Forwarding board failures occur too frequently, and there is no need to continue trying. The forwarding board can be completely isolated (you can no longer consider restoring the data transmission of forwarding board 1, or you can set a relatively long timer, such as 1 minute). The relevant information of the isolated member can be stored in the buffer; for the isolated faulty member, an isolation time can be set (in this example, it can be 1 minute, which also belongs to empirical data), and during this time, the forwarding You don't need to deal with whether the board is faulty or not, because it is in the isolated area and will no longer involve the communication of the multicast group. After the isolation time, if the isolated faulty member has recovered, you can add forwarding board 1 of the faulty member to multicast group A and multicast group B again, and perform the previous normal data delivery and data synchronization delivery.

In addition, it may happen that the faulty forwarding board recovers in multiple time periods. In the above example, if the member forwarding board 1 fails and then the forwarding board 2 fails, and after the forwarding board 1 returns to normal, during the data synchronization process, the forwarding board 2 Return to normal, you can add forwarding board 2 to multicast group B, and then re-synchronize data (starting from the failure of forwarding board 1 to when the fault of forwarding board 2 recovers), at this time, the members of multicast group B include Forwarding boards 1 and 2 (forwarding board 1 will repeatedly receive some previously synchronized data), when multicast group B completes data synchronization, delete multicast group B.

In the above example, there is another alternative solution, which can further reduce the mutual influence between the two faulty forwarding boards 1 and 2. This solution adopts the permanent isolation method, that is, as long as there is a multicast group B, the isolation area will not be processed. As a member, the isolated forwarding board 2 is added to the multicast group B until the multicast group B is deleted, and then the forwarding board 2 data is delivered synchronously.

The embodiment of the present invention also provides a point-to-multipoint multicast device. The point-to-multipoint multicast device may specifically be an entity with a multicast control function, such as a router, a switch, and the like. As shown in Figure 3, including:

The sending unit 301 is configured to send the data sent by the sending end to the receiving end in the first multicast group; the number of the receiving ends is two or more; sending the data to the first receiving end through the second multicast;

The multicast group control unit 302 is configured to delete the first receiving end from the first multicast group if the first receiving end of the receiving ends fails; if the first receiving end returns to normal, establish the first receiving end Two multicast groups and adding the first receiving end to the first multicast group and the second multicast group; deleting the second multicast group after sending data to the first receiving end.

Optionally, the multicast group control unit 302 is configured to establish a second multicast group and add the first receiver to the first multicast group and the second multicast group if the first receiver returns to normal The group includes: used to determine that the above-mentioned first receiving end has returned to normal after the set first time arrives, establish a second multicast group and add the above-mentioned first receiving end to the above-mentioned first multicast group and the above-mentioned second multicast group Group.

Further, as shown in FIG. 4, the sending unit 301 includes:

The sending subunit 401 is used to send the data sent by the sending end to the receiving end in the first multicast group; the number of the receiving end is two or more; sending the data to the first receiving end through the second multicast;

The sending control unit 402 is configured to control the sending subunit to preferentially send data to the second receiving end when sending data to the terminals in the first multicast group and the second receiving end in the second multicast group.

Further, as shown in Figure 5, the above-mentioned device also includes:

The time control unit 501 is configured to increase the value of the first time if the fault frequency of the first receiving end reaches a set value.

Optionally, during the process of sending the data sent by the sending end to the first receiving end through the second multicast group before the first receiving end fails to recover, the second receiving end among the above receiving ends returns to normal, then the above group The broadcast group control unit 302 can also realize the following functions:

The multicast group control unit 302 is further configured to add the second receiving end to the second multicast group, and the sending unit 301 is configured to pass the data sent by the sending end in the third time period through the second multicast group sent to the above-mentioned first receiving end and the second receiving end; the above-mentioned third time period is: the earlier failure time point of the failure time point of the first receiving end and the failure time point of the second receiving end to the time when the failure recovery of the second receiving end point in time.

The multicast group control unit 302 is further configured to re-establish the second multicast group and add the second receiver to the first multicast group and the second multicast group after the second multicast group is deleted. The sending unit 301 is further configured to send the data sent by the sending end before the second receiving end fails to recover to the second receiving end through the second multicast group.

The above-mentioned multicast group control unit 302 is also used to establish a third multicast group and add the above-mentioned second receiving end to the above-mentioned first multicast group and the above-mentioned third multicast group; after sending data to the second receiving end, delete the above-mentioned The third multicast group: the sending unit 301 is further configured to send the data sent by the sending end before the second receiving end fails to recover to the second receiving end through the third multicast.

In the process of data synchronization through the scheme of the embodiment of the present invention, there is a reduction in the number of multicast groups that need to be maintained through the management of multicast groups; the normal receiving end receives data and the faulty receiving end synchronizes data at the same time, eliminating The time for the normal receiving end to wait for the recovery of the faulty receiving end also eliminates the problem of repeatedly sending data to the normal receiving end; thereby improving the multicast efficiency and multicast system performance.

Those of ordinary skill in the art can understand that all or part of the steps in the method of the above-mentioned embodiments can be completed by instructing related hardware through a program, and the above-mentioned program can be stored in a computer-readable storage medium, the above-mentioned storage medium It can be read-only memory, disk or CD-ROM, etc.

A point-to-multipoint multicast method and device provided by the embodiments of the present invention have been described above in detail. In this paper, specific examples are used to illustrate the principle and implementation of the present invention. The description of the above embodiments is only used To help understand the method of the present invention and its core idea; at the same time, for those of ordinary skill in the art, according to the idea of the present invention, there will be changes in the specific implementation and scope of application. In summary, the content of this specification does not It should be understood as a limitation of the present invention.

Claims (10)

1. A point-to-multipoint multicast method, characterized in that, comprising: Sending the data sent by the sending end to the receiving end in the first multicast group; the number of receiving ends is two or more; If the first receiving end among the receiving ends fails, deleting the first receiving end from the first multicast group; If the first receiving end returns to normal, then establish a second multicast group and add the first receiving end to the first multicast group and the second multicast group; Sending the data sent by the sending end before the failure recovery of the first receiving end to the first receiving end through the second multicast, and deleting the second multicast group after sending the data to the first receiving end. 2. The method according to claim 1, characterized in that, restoring the first receiving end to normal comprises: After the set first time arrives, it is determined that the first receiving end has returned to normal. 3. The method according to claim 1, wherein when the sending end sends data to a terminal in the first multicast group and a second receiving end in the second multicast group, it preferentially sends data to the second receiving end . 4. The method according to claim 2, wherein, if the fault frequency of the first receiving end reaches a set value, the value of the first time is increased. 5. The method according to any one of claims 1 to 4, wherein, in the process of sending the data sent by the sending end to the first receiving end through the second multicast group before the first receiving end fails to recover, The second receiving end among the receiving ends returns to normal, then, adding the second receiving end to the second multicast group, and then sending the data sent by the sending end during the third time period to the first receiving end and the second receiving end through the second multicast group; The third time period is: the earlier failure time point of the failure time point of the first receiving end and the failure time point of the second receiving end to the time point of recovery of the failure of the second receiving end; or, After the deletion of the second multicast group, re-establish the second multicast group and add the second receiving end to the first multicast group and the second multicast group, and add the second receiving end The data sent by the sending end before the fault recovery is sent to the second receiving end through the second multicast group; or, establishing a third multicast group and adding the second receiving end to the first multicast group and the third multicast group; Sending the data sent by the sending end before the second receiving end fails to recover to the second receiving end through the third multicast, and deleting the third multicast group after sending the data to the second receiving end. 6. A point-to-multipoint multicast device, characterized in that it comprises: The sending unit is used to send the data sent by the sending end to the receiving end in the first multicast group; the number of receiving ends is two or more; sending the data to the first receiving end through the second multicast; A multicast group control unit, configured to delete the first receiving end from the first multicast group if the first receiving end of the receiving ends fails; if the first receiving end returns to normal, Then establish a second multicast group and add the first receiving end to the first multicast group and the second multicast group; delete the second multicast group after sending data to the first receiving end. 7. The device according to claim 6, wherein the multicast group control unit is configured to establish a second multicast group and join the first receiving end if the first receiving end returns to normal The first multicast group and the second multicast group include: used for determining that the first receiving end has returned to normal after the set first time arrives, establishing a second multicast group and sending the first multicast group A receiving end joins the first multicast group and the second multicast group. 8. The device according to claim 6, wherein the sending unit comprises: The sending subunit is used to send the data sent by the sending end to the receiving end in the first multicast group; the number of receiving ends is two or more; The sent data is sent to the first receiving end through the second multicast; The sending control unit is configured to control the sending subunit to preferentially send data to the second receiving end when sending data to the terminals in the first multicast group and the second receiving end in the second multicast group. 9. The device according to claim 6, further comprising: A time control unit, configured to increase the value of the first time if the fault frequency of the first receiving end reaches a set value. 10. The device according to any one of claims 6 to 9, characterized in that, In the process of sending the data sent by the sending end to the first receiving end through the second multicast group before the first receiving end fails to recover, the second receiving end in the receiving end returns to normal, then, The multicast group control unit is further configured to add the second receiving end to the second multicast group, and the sending unit is configured to pass the data sent by the sending end in the third time period through the second The multicast group is sent to the first receiving end and the second receiving end; the third time period is: the earlier failure time point of the failure time point of the first receiving end and the failure time point of the second receiving end to the second the point in time at which the receiving end fails to recover; or, The multicast group control unit is further configured to re-establish a second multicast group and add the second receiver to the first multicast group and the first multicast group after the deletion of the second multicast group The second multicast group; the sending unit is further configured to send the data sent by the sending end before the second receiving end fails to recover to the second receiving end through the second multicast group; or The multicast group control unit is also used to establish a third multicast group and add the second receiving end to the first multicast group and the third multicast group; sending data to the second receiving end is completed Then delete the third multicast group; the sending unit is further configured to send the data sent by the sending end before the second receiving end fails to recover to the second receiving end through the third multicast.

Images