CN101515889B - Replication and processing method for multicast message and device thereof - Google Patents
Replication and processing method for multicast message and device thereof Download PDFInfo
- Publication number
- CN101515889B CN101515889B CN200910081159XA CN200910081159A CN101515889B CN 101515889 B CN101515889 B CN 101515889B CN 200910081159X A CN200910081159X A CN 200910081159XA CN 200910081159 A CN200910081159 A CN 200910081159A CN 101515889 B CN101515889 B CN 101515889B
- Authority
- CN
- China
- Prior art keywords
- multicast message
- multicast
- replication processes
- message
- group address
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/02—Details
- H04L12/16—Arrangements for providing special services to substations
- H04L12/18—Arrangements for providing special services to substations for broadcast or conference, e.g. multicast
- H04L12/1863—Arrangements for providing special services to substations for broadcast or conference, e.g. multicast comprising mechanisms for improved reliability, e.g. status reports
- H04L12/1877—Measures taken prior to transmission
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/02—Details
- H04L12/16—Arrangements for providing special services to substations
- H04L12/18—Arrangements for providing special services to substations for broadcast or conference, e.g. multicast
- H04L12/1881—Arrangements for providing special services to substations for broadcast or conference, e.g. multicast with schedule organisation, e.g. priority, sequence management
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The embodiment of the invention provides a replication and processing method for multicast message and a device thereof; wherein, the replication and processing method for multicast message comprises: receiving multicast message, determining number of times needed by the whole multicast message to be replicated according to the multicast address of the multicast message and obtaining the number of times of the replication and processing of the predetermined multicast message; when the number of times needed by the whole multicast message to be replicated is larger than the number of times of the replication and processing of the predetermined multicast message, the replication of the number of times of the replication and processing of the predetermined multicast message is carried out on the multicast message and the multicast message is stored. When the number of receiving ends of a multicast group is quite a lot and great amount of replications are needed, a user can preset the number of times of the replication and processing of the multicast message, thus reducing the replication delay of the multicast message.
Description
Technical field
The present invention relates to communication technical field, relate in particular to a kind of duplication method and device of multicast message.
Background technology
Employing IP multicasting technology (Multicast) can obviously reduce the bandwidth requirement to network, has reduced offered load, has improved the efficiency of transmission of network.The IP multicast application probably can be divided three classes: point-to-multipoint, and how point-to-point, multi-multipoint.Point-to-multipoint is used and to be meant that a transmitting terminal correspondence a plurality of receiving terminals, such as: Web TV; Many point-to-point application are meant the corresponding receiving terminal of a plurality of transmitting terminals, such as: the network auction; Multi-multipoint application is meant the corresponding a plurality of receiving terminals of a plurality of transmitting terminals, such as: Web conference.
Particularly, in the IP multicasting technology, transmitting terminal sends a plurality of multicast message information, by receiving terminal receiving group message information, several receiving terminals constitute a multicast group, and (Multicast Group MG), can have a plurality of multicast group simultaneously, each multicast group has the multicast group address (MGID, Multicast Group Identity) of unique correspondence.Transmitting terminal sends a plurality of multicast message information, multicast message duplicates next multicast message after being copied to all receiving terminals of multicast group again, just, a multicast message is replicated a receiving terminal several a multicast group, duplicates next multicast message again.
In realizing process of the present invention, the inventor finds that there are the following problems at least in the prior art:
When the receiving terminal number of a multicast group suitable (as more than or equal to 10000) for a long time, when also adopting a multicast message to remove to duplicate next multicast message after having duplicated again, the time delay that sends after will causing next multicast message to duplicate is excessive, bring in the multicast group one that receives this multicast message and to see, the discontinuous phenomenon of data can occur, influence the ability of IP multicast.
Summary of the invention
Embodiments of the invention provide a kind of duplication method and device of multicast message, have reduced the time-delay that multicast message duplicates.
A kind of duplication method of multicast message comprises:
The receiving group message, multicast group address according to described multicast message, confirm the number of times that described multicast message all duplicates to be needed, and obtain default multicast message replication processes number of times that described default multicast message replication processes number of times is the replication processes number of times default according to delay requirement;
When described multicast message all duplicates the number of times that needs greater than described default multicast message replication processes number of times, add the multicast group address of described multicast message to look-up table, duplicate and send the multicast message that duplicates to what described multicast message carried out described default multicast message replication processes number of times, and store described multicast message, described look-up table is used to store the number of times that all the duplicates needs multicast group address greater than the multicast message of described default multicast message replication processes number of times.
A kind of replication processes device of multicast message comprises:
Duplicate ack unit: be used for the receiving group message, according to described multicast message multicast group address, confirm the number of times that described multicast message all duplicates to be needed, and obtaining default multicast message replication processes number of times, described default multicast message replication processes number of times is the replication processes number of times default according to delay requirement;
Replication processes unit: when being used for number of times that described multicast message all duplicates needs greater than described default multicast message replication processes number of times, duplicate and send the multicast message that duplicates to what described multicast message carried out described default multicast message replication processes number of times, and store described multicast message;
Look-up table stores unit: be used to store the number of times that all duplicates needs multicast group address greater than the multicast message of described default multicast message replication processes number of times.
The technical scheme that is provided by the embodiment of the invention described above as can be seen, and is quite a lot of when the receiving terminal number of multicast group, needs under the situation of massive duplication, and the user can preset the replication processes number of times of multicast message, has reduced the time-delay that multicast message duplicates.
Description of drawings
In order to be illustrated more clearly in the technical scheme of the embodiment of the invention, the accompanying drawing of required use is done to introduce simply in will describing embodiment below, apparently, accompanying drawing in describing below only is some embodiments of the present invention, for those of ordinary skills, under the prerequisite of not paying creative work, can also obtain other accompanying drawing according to these accompanying drawings.
Fig. 1 is the flow chart of the duplication method of embodiment of the invention multicast message;
Fig. 2 is the flow chart of the duplication method of embodiment of the invention multicast message;
Fig. 3 is the flow chart of the duplication method of embodiment of the invention multicast message;
Fig. 4 is the formation block diagram of the replication processes device of embodiment of the invention multicast message;
Fig. 5 is the formation block diagram of the replication processes device of embodiment of the invention multicast message;
Fig. 6 is the formation block diagram of the replication processes device of embodiment of the invention multicast message.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the invention, the technical scheme in the embodiment of the invention is clearly and completely described, obviously, described embodiment only is the present invention's part embodiment, rather than whole embodiment.Based on the embodiment among the present invention, those of ordinary skills belong to the scope of protection of the invention not making the every other embodiment that is obtained under the creative work prerequisite.
As shown in Figure 1, the embodiment of the invention provides a kind of duplication method of multicast message, comprising:
Step 1: the receiving group message, according to the multicast group address of described multicast message, confirm the number of times that described multicast message all duplicates to be needed, and obtain default multicast message replication processes number of times;
Step 2: when described multicast message all duplicates the number of times that needs greater than described default multicast message replication processes number of times, duplicate and send the multicast message that duplicates to what described multicast message carried out described default multicast message replication processes number of times, and store described multicast message.
Described default multicast message replication processes number of times is the replication processes number of times default according to delay requirement.
Come as can be seen, under the situation that needs massive duplication, the user can reduce the time-delay that multicast message duplicates according to the replication processes number of times of the default multicast message of delay requirement; And the user can dynamically adjust the replication processes number of times according to delay requirement in the multicast message reproduction process.
Wherein, step 2 can also be:
Described multicast message all duplicates the number of times that needs smaller or equal to described default multicast message replication processes number of times, to described multicast message carry out described all duplicate the number of times that needs duplicate and send the multicast message that duplicates.
Like this, just duplicating of the replication processes number of times that multicast message is only preset sends, and satisfied the delay requirement that multicast message duplicates, and can not be stored and duplicate the multicast message of finishing fully, divides the several times replication processes to realize all duplicating.And, need add the multicast group address that does not duplicate the described multicast message of finishing fully to look-up table, described look-up table is used to store the number of times that all the duplicates needs multicast group address greater than the multicast message of described default multicast message replication processes number of times.
Look-up table can be just to set up well in advance, also can be when first multicast message all duplicates the number of times that needs greater than described default multicast message replication processes number of times, set up look-up table, like this, it is issuable out of order after storage that the employing look-up table can avoid not duplicating fully the multicast message of finishing.
Can know that constantly scheduling obtains new multicast message information, receiving terminal receiving group message information.
So described method also comprises step 3: receive new multicast message, determine to carry out in the multicast message (promptly not duplicating the multicast message of finishing fully) of described new multicast message and described storage the step of the multicast message of replication processes, particularly,
Why with described new multicast message corresponding stored behind described multicast message, be in order to guarantee that message sequence can be not out of order, all duplicate realizing.
Further, after the multicast message of described storage duplicates, when the number of copy times of the multicast message of described storage does not reach its number of times that all duplicates needs, the multicast message of the described storage of write-back, make it wait for follow-up being scheduled, realize that the multicast message of described storage all duplicates; When the number of copy times of the multicast message of described storage had reached its number of times that all duplicates needs, the multicast message of described storage had been finished all and has been duplicated.
Perhaps, step 3 can be, when the multicast group address in the multicast group address of new multicast message and the described look-up table does not match, according to the priority of default new multicast message with the multicast message of storage, dispatches the multicast message that obtains carrying out replication processes.
Particularly, obtain carrying out the difference of the multicast message of replication processes, comprise following 2 kinds of situations according to scheduling:
When the multicast message that obtains carrying out replication processes when scheduling is described new multicast message:
When described new multicast message all duplicated the number of times that needs smaller or equal to described default multicast message replication processes number of times, that described new multicast message all duplicated the number of times that needs duplicated and is sent out the multicast message that duplicates;
Perhaps, when described new multicast message all duplicates the number of times that needs greater than described default multicast message replication processes number of times, duplicate and send the multicast message that duplicates to what described new multicast message carried out described default multicast message replication processes number of times, and store described new multicast message and the multicast group address of described new multicast message is added in the described look-up table.
When the multicast message that obtains carrying out replication processes when scheduling is the multicast message of described storage:
Multicast message to described storage carries out replication processes, when the number of copy times of the multicast message of described storage does not reach its number of times that all duplicates needs, and the multicast message of the described storage of write-back; When the number of copy times of the multicast message of described storage had reached its number of times that all duplicates needs, the multicast message of described storage had been finished all and has been duplicated.
In another embodiment, when the multicast message that obtains carrying out replication processes when scheduling is the multicast message of described storage, need carry out replication processes to the multicast message of described storage, this is to judge earlier, all duplicate the number of times that needs, after deducting number of copy times through default multicast message replication processes number of times, whether the remaining number of times of needs that duplicates is also greater than default multicast message replication processes number of times, if duplicate the multicast message of the described storage of write-back greater than the multicast message replication processes number of times of then described message being preset; If less than, then carry out the remaining number of times that duplicates needs and duplicate, like this, when the number of copy times of the multicast message of described storage had reached its number of times that all duplicates needs, the multicast message of described storage had been finished all and has been duplicated.
And, the priority of the new multicast message of presetting in the step 3 and the multicast message of storage, the user can dynamically adjust in the multicast message reproduction process.
Described scheduling can be adopted and add wooden fork Fair Queue scheduling (WFQ, Weighted FairQueuing), and (Weighted Round Robin, WRR) scheduling or wheel change (Round Robin) scheduling also can to adopt weighted round ring.
And, in the described method, can confirm to receive the order of each described multicast message according to the priority scheduling of a plurality of multicast messages.
It should be noted that, when the multicast group address in the multicast group address of new multicast message and the described look-up table does not match, if the multicast message of described storage or write-back is two when above, can dispatch between the multicast message to described storage or write-back earlier, with the multicast message that obtains dispatching, obtain carrying out at last the multicast message of replication processes again with described new multicast message.
The technical scheme that is provided by the embodiment of the invention described above as can be seen, when the receiving terminal number of multicast group quite a lot of, need under the situation of massive duplication, the user can be according to the replication processes number of times of the default multicast message of different delay requirements, the number of times that duplicates according to the whole needs of the message of reality with the replication processes number of times of default multicast message, multicast message is duplicated, reduced the time-delay that multicast message duplicates; And the user can dispose this time delay arbitrarily, and is convenient, flexible, applied widely.
As shown in Figure 4, corresponding to the duplication method of above-mentioned multicast message, the embodiment of the invention also provides a kind of replication processes device of multicast message, comprising:
Duplicate ack unit 10: be used for the receiving group message,, confirm the number of times that described multicast message all duplicates to be needed, and obtain default multicast message replication processes number of times according to described multicast message multicast group address;
Replication processes unit 20: when being used for number of times that described multicast message all duplicates needs greater than described default multicast message replication processes number of times, duplicate and send the multicast message that duplicates to what described multicast message carried out described default multicast message replication processes number of times, and store described multicast message.
Equally, described default multicast message replication processes number of times is the replication processes number of times default according to delay requirement.
When multicast message all duplicates the number of times that needs greater than described default multicast message replication processes number of times, the duplicating of multicast message replication processes number of times of presetting can not be realized duplicating fully described multicast message, so, can store and not duplicate the multicast message of finishing fully, divide the several times replication processes to realize all duplicating.
The technical scheme that is provided by the embodiment of the invention described above as can be seen, when the receiving terminal number of multicast group quite a lot of, need under the situation of massive duplication, the user can be according to the replication processes number of times of the default multicast message of different delay requirements, the number of times that duplicates according to the whole needs of the message of reality with the replication processes number of times of default multicast message, multicast message is duplicated, reduced the time-delay that multicast message duplicates; And the user can dispose this time delay arbitrarily, and is convenient, flexible, applied widely.And adopted the write-back mechanism of multicast message, made the number of copy times of multicast message reach the number of times that whole needs duplicate; Introduced the look-up table unit, reduced not duplicate fully the multicast message of finishing and after storage, may produce out of order situation.
So as shown in Figure 5, the replication processes device of described multicast message can also comprise:
Look-up table stores unit 30: be used to store the number of times that all duplicates needs multicast group address greater than the multicast message of described default multicast message replication processes number of times;
Write-back confirmation unit 40: the multicast group address that is used to confirm multicast message whether with described look-up table stores unit in the multicast group address coupling, be used to confirm whether multicast message needs storage or write-back, and be used to manage described look-up table stores unit.
Write-back buffer 50: it can comprise at least one write-back registers 501 and at least one corresponding write-back push-up storage 502, wherein,
At least one write-back registers 501, be used for according to the described number of times that all duplicates needs of multicast group address storage greater than described default multicast message replication processes number of times and with the interior unmatched multicast message of multicast group address in look-up table stores unit 30, and the described multicast message of write-back;
At least one write-back push-up storage 502 is used to store the subsequent packet of multicast messages in the described write-back registers 501, described subsequent packet be multicast group address with described write-back registers 501 in the multicast message that mates of the multicast group address of multicast messages.
Here, in the described write-back buffer 50 number of write-back registers 501 by the number decision of the multicast message of needs storages.The storage depth of write-back push-up storage 502 is according to the number decision of the subsequent packet of the multicast message of described storage in the described write-back buffer 50.
And the follow-up multicast message output of write-back push-up storage 502 enters in the write-back registers 501, avoids multicast message issuable out of order.
Write-back can be interpreted as, and after multicast messages duplicated in the write-back registers 501, its number of copy times did not reach it when all duplicating the number of times that needs, and needs the described multicast message of write-back, waits for that follow-up being scheduled once more carry out replication processes, all duplicates realizing; When the number of copy times of described multicast message had reached its number of times that all duplicates needs, described multicast message had been finished all and has been duplicated.
And the replication processes device of described multicast message can also comprise:
Duplicate scheduling unit 60, be used for confirming that the multicast message in multicast message and the write-back registers 501 carries out the multicast message of replication processes.
Here, the multicast message of dispatching with the multicast message in the write-back registers 501 be with look-up table stores unit 30 in the unmatched multicast message of multicast group address, owing to do not match with the multicast group address in the look-up table stores unit 30, it is not the subsequent packet of multicast message in the write-back registers 501, so, dispatch and draw a multicast message that enters the replication processes unit.
Particularly, the described scheduling unit 60 that duplicates can comprise:
First scheduler module 601 is used for the multicast message in two above write-back registers 501 is dispatched;
Second scheduler module 602 is used for the multicast message that obtains entering described replication processes unit 20 from the scheduling result and the scheduling of described multicast message of first scheduler module 601;
Described scheduling mode is Weighted Fair Queuing, weighted round robin or robin scheduling.
The replication processes device of described multicast message can also comprise:
Inlet first in first out buffer memory 70: be used to receive a plurality of multicast messages and store described a plurality of multicast message according to its priority;
Inlet scheduling unit 80: be used for the priority according to a plurality of multicast messages, scheduling obtain entering the mouth each multicast message in the first in first out buffer memory 70 sends to the order of duplicate ack unit 10.
Inlet first in first out buffer memory 70 can be by the FIFO (first in first out of at least two high low priorities, First-in First-out) memory is formed, according to the priority of a plurality of multicast messages, a plurality of multicast messages are mapped in the FIFO memory of two corresponding high low priorities of difference.Adopt priority mapping, the message that can guarantee high priority is not by the message bandwidth-hogging of low priority.
Inlet scheduling unit 80 can adopt Weighted Fair Queuing, weighted round robin or robin scheduling, is read to confirm which multicast message leaves formation in the inlet first in first out buffer memory 70.
The technical scheme that is provided by the embodiment of the invention described above as can be seen, when the receiving terminal number of multicast group quite a lot of, need under the situation of massive duplication, the user can be according to the replication processes number of times of the default multicast message of different delay requirements, replication processes number of times according to the whole needs of the message of reality number of times that duplicates and the multicast message of presetting, multicast message is duplicated, reduced the time-delay that multicast message duplicates; And the user can dispose this time delay arbitrarily, and is convenient, flexible, applied widely.And adopted the write-back mechanism of multicast message, made the number of copy times of multicast message reach the number of times that whole needs duplicate; Introduced the look-up table unit, reduced not duplicate fully the multicast message of finishing and after storage, may produce out of order situation.
Below in conjunction with concrete multicast group, multicast message, the technical scheme that embodiments of the invention provide is described.
Embodiment one
Referring to shown in Figure 3, with n multicast group, multicast group address is the MGID0-MGIDn-1 example, and the duplication method of multicast message is described, comprising:
Receiving a plurality of multicast messages and also can deposit earlier in the inlet first in first out buffer memory according to priority, can be weight with the priority of multicast message, adopts Weighted Fair Queuing WFQ scheduling to obtain first multicast message, multicast group address MGID0.Certainly also can adopt weighted round ring WRR scheduling or wheel to change the RR scheduling in another embodiment.
The multicast message that step 2, receiving scheduling obtain
Receive first multicast message, can learn the target multicast group by its multicast group address MGID0, and the number of further learning the receiving terminal that the target multicast group is comprised, and confirm that first multicast message is all duplicated the number of times K that needs.And, also obtain the user according to the default replication processes times N of time-delay needs.
The replication processes times N that the number of times K that first multicast message is all duplicated to be needed presets according to the time-delay needs greater than the user, so, first multicast message can only duplicate N time and just send, and it is less to satisfy the time-delay that sends between the multicast message that duplicates after the duplicated multicast message.
Look-up table is used to store the number of times that all the duplicates needs multicast group address greater than the multicast message of described default multicast message replication processes number of times, because first multicast message is the multicast message of first reception, so, look-up table is empty at present, multicast group address does not match in the multicast group address of first multicast message and the look-up table, enters step 4.
If in the subsequent step, when multicast group address mates in the multicast group address of the multicast message that receives and the look-up table, multicast message is the subsequent packet of the multicast message of the multicast group address correspondence of coupling in the look-up table, so, enter step 7, follow-up multicast message is stored, specifically can see below the literary composition narration.
Step 4, scheduling obtain carrying out the replication processes multicast message
Because first multicast message is the multicast message of first reception, carrying out the replication processes multicast message as can be known is first multicast message.
Step 5, replication processes also send the multicast message that duplicates
The replication processes times N that first multicast message is only preset time duplicate the multicast message that i.e. transmission is duplicated.But the number of times K that first multicast message is unrealized and is all duplicated, so, step 6 can be entered.
Step 6, judge whether the storage or the write-back multicast message
Duplicating promptly of the replication processes times N that first multicast message is only preset time sends, and first multicast message does not duplicate fully and finishes, so enter step 7 storage first multicast message, waits until and divides several times to carry out replication processes in the subsequent step to realize all duplicating.
Step 7, storage or write-back multicast message and management look-up table
Store first multicast message, and the multicast group address MGID0 of first multicast message is added in the look-up table, avoid multicast message issuable out of order after storage.
Can know constantly have new multicast message to be scheduled out in the step 1, wait for replication processes, below the situation that may exist with 4 example explanations.Here " the new multicast message " that reach hereinafter only is the multicast message that obtains corresponding to scheduling before, is not the qualification to message itself.
Referring to shown in Figure 3, can be to dispatch again in the step 1 and obtain second multicast message, its multicast group address MGID0 again.
The multicast message that step 2, receiving scheduling obtain
Receive second multicast message, confirm that second multicast message is all duplicated the number of times K ' that needs, and known users is according to the default replication processes times N of time-delay needs.
Multicast group address is MGID0 in the look-up table, and multicast group address MGID0 coupling enters step 7 in the multicast group address MGID0 of second multicast message and the look-up table.
Step 7, storage or write-back multicast message and management look-up table
Second multicast message is the subsequent packet of first multicast message, and second multicast message is stored in after first multicast message, and the message that guarantees to belong to same multicast group address is not out of order.
Again referring to shown in Figure 3, can also be, in the step 1 again scheduling obtain second multicast message, its multicast group address MGID1, the second multicast message circulation above-mentioned steps so, as follows:
The multicast message that step 2, receiving scheduling obtain
Receive second multicast message, confirm that second multicast message is all duplicated the number of times K ' that needs, and known users is according to the default replication processes times N of time-delay needs.
Multicast group address is MGID0 in the look-up table, and multicast group address MGID0 does not match in the multicast group address MGID1 of second multicast message and the look-up table, enters step 4.
Step 4, scheduling obtain carrying out the replication processes multicast message
In first multicast message and second multicast message of storage, be weight according to the priority of default new multicast message and the multicast message of storage or write-back, employing adds the wooden fork Fair Queue dispatches first multicast message that obtains storing and carries out replication processes.Certainly also can adopt in another embodiment and add wooden fork repeating query WRR scheduling or wheel commentaries on classics RR scheduling.
Step 5, replication processes also send the multicast message that duplicates
The replication processes times N only preset once more of first multicast message of storage time duplicate the multicast message that i.e. transmission is duplicated.
Step 6, judge whether the storage or the write-back multicast message
If the number of times K that first multicast message of storage is unrealized and is all duplicated enters step 7, if the number of times K that the realization of first multicast message of storage is all duplicated then finishes.
Step 7, storage or write-back multicast message and management look-up table
First multicast message of storage carries out write-back, and waiting until is scheduled in the subsequent step carries out replication processes and realize all duplicating.
Again referring to shown in Figure 3, can also be, in the step 1 again scheduling obtain second multicast message, its multicast group address MGID1, the second multicast message circulation above-mentioned steps so, as follows:
The multicast message that step 2, receiving scheduling obtain
Receive second multicast message, confirm that second multicast message is all duplicated the number of times K ' that needs, and known users is according to the default replication processes times N of time-delay needs.
Multicast group address is MGID0 in the look-up table, and multicast group address MGID0 does not match in the multicast group address MGID1 of second multicast message and the look-up table, enters step 4.
Step 4, scheduling obtain carrying out the replication processes multicast message
In first multicast message and second multicast message of storage, be weight according to the priority of default new multicast message and the multicast message of storage or write-back, employing Weighted Fair Queuing WFQ dispatches second multicast message that obtains storing and carries out replication processes.Certainly also can adopt weighted round ring WRR scheduling or wheel to change the RR scheduling in another embodiment.
Step 5, replication processes also send the multicast message that duplicates
Second multicast message is all duplicated the number of times K ' that needs greater than default replication processes times N, the replication processes times N that second multicast message is only preset time duplicate the multicast message that i.e. transmission is duplicated.
Step 6, judge whether the storage or the write-back multicast message
What the replication processes times N that second multicast message is only preset was inferior duplicates, and the number of times K ' that second multicast message is unrealized and is all duplicated is so enter step 7.
Step 7, storage or write-back multicast message and management look-up table
Store second multicast message, waiting until is scheduled in the subsequent step carries out replication processes and realizes all duplicating.
Again referring to shown in Figure 3, if second multicast message is all duplicated the number of times K ' of needs smaller or equal to default replication processes times N in the above-mentioned steps 5, second multicast message carries out the inferior replication processes of K ', promptly realizes all duplicating and can satisfying default delay requirement.
In sum, each scheduling obtains a new multicast message, the multicast group address that will judge the multicast message that this is new whether with look-up table in the multicast group address coupling.If coupling is then carried out follow-up flow process with reference to above-mentioned steps 7; If do not match, then carry out follow-up flow process, thereby realize duplicating fully all multicast messages with reference to above-mentioned steps 4,5,6.
In the above-mentioned steps 6, there is the write-back processing mode, to realize that multicast message all duplicates.Specifically be, the multicast message of having stored is scheduled once more and carries out replication processes, if number of copy times has satisfied the number of times that all duplicates, then multicast message has been finished all and duplicated; If the number of times that the number of copy times no show is all duplicated so also needs the write-back multicast message, make multicast message be kept by write-back, wait for follow-up being scheduled, realize that multicast message all duplicates.
Here, can obtain in the multicast message in scheduling, just obtain multicast message and whether all duplicate the number of times that needs, confirm whether multicast message needs storage and whether need write-back greater than user's replication processes times N default according to time-delay, further, be multicast message flagged write back mark.
And, in the above-mentioned steps 6, if the multicast message of storage or write-back is more than two, can dispatch between the then plural multicast message, result that scheduling is obtained and new multicast message are dispatched, and then obtain carrying out that multicast message of replication processes step 5, and, needing the priority of the multicast message of the default new multicast message of basis and storage or write-back is weight, and the multicast message of new multicast message and storage or write-back is dispatched.Described scheduling can be adopted Weighted Fair Queuing (WFQ, Weighted Fair Queuing), and (WeightedRound Robin, WRR) scheduling or wheel change (Round Robin) scheduling also can to adopt weighted round ring.
In the above-mentioned steps 7, when all messages that belong to same multicast group all be replicated finish after, promptly all multicast messages of a multicast group address correspondence are all realized all duplicating, and this multicast group address is deleted in look-up table.
Can also know, can when step 4 replication processes also sends the multicast message that duplicates, just carry out step 6.When can and send the multicast message that duplicates in step 4 replication processes, just carry out step 1, scheduling obtains new multicast message.
The technical scheme that is provided by the embodiment of the invention described above as can be seen, when the receiving terminal number of multicast group quite a lot of, need under the situation of massive duplication, the user can be according to the replication processes number of times of the default multicast message of different delay requirements, replication processes number of times according to the whole needs of the message of reality number of times that duplicates and the multicast message of presetting, multicast message is duplicated, reduced the time-delay that multicast message duplicates; And the user can dispose this time delay arbitrarily, and is convenient, flexible, applied widely.And adopted the write-back mechanism of multicast message, made the number of copy times of multicast message reach the number of times that whole needs duplicate; Introduced look-up table, reduced not duplicate fully the multicast message of finishing and after storage, may produce out of order situation.
Embodiment two
Referring to shown in Figure 6, a kind of replication processes device of the multicast message corresponding to the duplication method of multicast message among the embodiment one, still with n multicast group, multicast group address is the MGID0-MGIDn-1 example.
The replication processes device of multicast message comprises:
Duplicate ack unit 10: be used for the receiving group message, confirm that according to the multicast group address of described multicast message multicast message all duplicates the number of times of needs, and obtain default multicast message replication processes number of times;
Replication processes unit 20: when being used for number of times that described multicast message all duplicates needs greater than described default multicast message replication processes number of times, duplicate and send the multicast message that duplicates to what described multicast message carried out described default multicast message replication processes number of times, and store described multicast message.
The default multicast message replication processes number of times that duplicate ack unit 10 obtains is user's replication processes number of times default according to delay requirement.And the user can dynamically adjust multicast message replication processes number of times according to delay requirement in the multicast message reproduction process.
When multicast message all duplicated the number of times that needs greater than described default multicast message replication processes number of times, the duplicating of multicast message replication processes number of times of presetting can not be realized duplicating fully described multicast message.
So the replication processes device of multicast message also comprises look-up table stores unit 30, write-back confirmation unit 40, write-back buffer 50, duplicates scheduling unit 60.
Look-up table stores unit 30: look-up table stores unit: be used to store the number of times that all duplicates needs multicast group address greater than the multicast message of described default multicast message replication processes number of times;
Write-back confirmation unit 40: the multicast group address that is used to confirm multicast message whether with look-up table stores unit 30 in the multicast group address coupling, be used to confirm whether multicast message needs storage or write-back, and be used to manage described look-up table stores unit 30.
Particularly, the multicast group address that is used to confirm multicast message whether with look-up table stores unit 30 in the multicast group address coupling be meant, according to the multicast group address in the look-up table stores unit 30, can judge whether the multicast message that receives is the subsequent packet of the multicast message of the multicast group address correspondence of coupling in the look-up table stores unit 30, if, then for fear of packet out-ordering, the multicast message sequential storage that belongs to same multicast group address.
Be used to confirm whether multicast message needs storage to be meant, the number of times that need duplicate when the multicast message that receives is during greater than described default multicast message replication processes number of times, multicast message is stored in the write-back buffer 50, waits follow-up being scheduled, realize that multicast message all duplicates; The number of times that the follow-up new multicast message that receives need duplicate is greater than described default multicast message replication processes number of times, and during with multicast group address coupling in the look-up table stores unit 30, the new report literary composition of group is stored in the write-back buffer 50, be scheduled etc. follow-up, realize that multicast message all duplicates.
Whether the multicast message of confirming described storage needs write-back to be meant, be stored in the multicast message in the write-back buffer 50, follow-up being scheduled once more carried out replication processes, all do not duplicate if still realize, the multicast message write-back that so also needs to store is in write-back buffer 50, make it wait for follow-up being scheduled, realize that multicast message all duplicates.
Management look-up table stores unit 30 is meant, the number of times that need duplicate when the multicast message that receives is during greater than described default multicast message replication processes number of times, and the multicast group address of multicast message is added in the look-up table stores unit 30; Constantly receive new multicast message, multicast group address in new multicast message and look-up table stores unit 30 does not match, and when new multicast message need be stored in the write-back buffer 50, the multicast group address that adds new multicast message arrives in the look-up table stores unit 30 record of the look-up table stores of enriching constantly unit 30; When all multicast messages of look-up table stores unit 30 arbitrary multicast group address all be replicated finish after, with the deletion in 30 of this multicast group address from the look-up table stores unit.Can know, adopt look-up table mechanism can avoid that message may produce after the multicast write-back out of order.
Here " the new multicast message " that reach hereinafter only is the multicast message that obtains corresponding to scheduling before, is not the qualification to message itself.
Write-back buffer 50: it comprises at least one write-back registers 501 and at least one corresponding write-back push-up storage 502, wherein,
At least one write-back registers 501, be used for according to the described number of times that all duplicates needs of multicast group address storage greater than described default multicast message replication processes number of times and with the interior unmatched multicast message of multicast group address in look-up table stores unit 30, and the described multicast message of write-back.Store described multicast message according to multicast group address,, divide the several times replication processes to realize all duplicating so that carry out follow-up scheduling.Illustrate 16 write-back registers 501 (reg_0-reg_15) among Fig. 4.
At least one write-back first in first out (First-in First-out, FIFO) memory 502, be used to store the subsequent packet of multicast message in the described write-back registers 501, described subsequent packet is the multicast message of the multicast group address coupling of multicast message in multicast group address and the described write-back registers 501.The storage mode of first in first out (FIFO) guarantees multicast message and its subsequent packet in reproduction process, and is not out of order.Illustrate 16 write-back push-up storages 502 (fifo_0-fifo_15) among Fig. 4, its storage depth is by the number decision of the subsequent packet of the multicast message of storage.
Because write-back registers 501 is distinguished by different multicast group address MGID, and write-back push-up storage 502 is used to store the back continuation of insurance literary composition of multicast message in the write-back registers 501, so the multicast group address MGID of the multicast message of corresponding write-back push-up storage 502 stored is inevitable different.
And the number of write-back buffer 50 and the burst of the multicast traffic of inlet have relation, and it is more little to happen suddenly, write-back buffer 50 corresponding minimizings, and it is big more to happen suddenly, and write-back buffer 50 is in requisition for increase.
Duplicate scheduling unit 60, be used for confirming that the write-back multicast message in new multicast message and the described write-back registers 501 carries out the multicast message of replication processes.Described new multicast message be with described look-up table stores unit 30 in the unmatched multicast message of multicast group address.
Particularly, duplicating scheduling unit 60 comprises:
First scheduler module 601 is used for the multicast message in two above write-back registers 501 is dispatched, and, adopts robin scheduling here.Can certainly adopt Weighted Fair Queuing (WFQ, Weighted Fair Queuing), perhaps weighted round ring (Weighted Round Robin, WRR) scheduling.
Second scheduler module 602 is used for the multicast message that obtains entering described replication processes unit 20 from the scheduling result and the described new multicast message scheduling of first scheduler module 601.
The second scheduling mould, 602 priority according to default new multicast message and the multicast message of storage or write-back of determining are weight, from the scheduling result of first scheduler module 601 and with have with the unmatched new multicast message of the interior multicast group address in look-up table stores unit 30 obtain entering the multicast message of replication processes unit 20.And the user can dynamically adjust the priority of the multicast message of new multicast message of configuration and write-back in the multicast message reproduction process.
The replication processes device of multicast message also comprises: inlet first in first out buffer memory 70 and inlet scheduling unit 80.
Inlet first in first out (FIFO) buffer memory 70: be used to receive a plurality of multicast messages and store described a plurality of multicast message according to its priority.
Inlet first in first out buffer memory 70 can be made up of first in first out (FIFO) memory of two high low priorities at least, priority according to each headings of a plurality of multicast messages carries is mapped to a plurality of multicast messages in the push-up storage of two corresponding high low priorities of difference.Adopt priority mapping, the message that can guarantee high priority is not by the message bandwidth-hogging of low priority.
Inlet scheduling unit 80: be used for the priority according to a plurality of multicast messages, scheduling obtain entering the mouth each multicast message in the first in first out buffer memory 70 sends to the order of duplicate ack unit 10.Here, inlet scheduling unit 80 is a weight with the priority of a plurality of multicast messages, adopts the Weighted Fair Queuing mode to obtain sending to the multicast message of duplicate ack unit 10.
Referring to shown in Figure 5, during the replication processes device of the concrete multicast message of using present embodiment, with first multicast message, its multicast group address MGID0, its number of times K that duplicates needs is greater than default replication processes times N, and second multicast message, and its multicast group address is MGID1, it duplicates the number of times K ' that needs is example greater than default replication processes times N, comprising:
1: inlet scheduling unit 80 is that weight adopts Weighted Fair Queuing according to the priority of a plurality of multicast messages, obtains multicast message from inlet first in first out buffer memory 70, as first multicast message, and its multicast group address MGID0.
2: duplicate ack unit 10 receives first multicast message, and confirm first multicast message all duplicate the number of times K that needs and obtain user's replication processes times N default according to time-delay, and the whole number of times K that need of duplicating that confirm first multicast message are greater than user's replication processes times N default according to time-delay.
3: write-back confirmation unit 40 adds the multicast group address MGID0 of first multicast message in the look-up table stores unit 30 to, first multicast message is in replication processes unit 20 after replication processes N time, write-back confirmation unit 40 is stored in it in interior unappropriated register of write-back registers 501 (reg_0-reg_15), as reg_0, be scheduled etc. follow-up, realize that multicast message all duplicates.
4: receive the scheduling from inlet first in first out buffer memory 70 of inlet scheduling unit 80 and obtain new multicast message, as second multicast message, multicast group address is MGID1, and the multicast group address MGID0 in the multicast group address MGID1 of second multicast message and the look-up table stores unit 30 is not complementary.
5: the priority according to the multicast message of the new multicast message of user preset and write-back is weight, 602 pairs second multicast messages of second scheduler module of scheduling unit 60 and first multicast message are weighted the Fair Queue scheduling, confirm that entering the multicast message that duplicates and send in replication processes unit 20 is second multicast message.
6: according to the multicast group address MGID1 of second multicast message, duplicate ack unit 10 confirms that the number of times K ' that duplicates needs of second multicast message is greater than default replication processes times N, so, what second multicast message entered the multicast message replication processes times N preset replication processes unit 20 duplicates and sends the multicast message that duplicates, equally, also need to store second multicast message, as second multicast message being stored in the unappropriated register in the write-back registers 501 (reg_0-reg_15), as reg_1, be scheduled etc. follow-up, realize that multicast message all duplicates.
And, write-back confirmation unit 40 also adds the multicast group address MGID1 of second multicast message in the look-up table stores unit 30 to, purpose is, whether the multicast message of judging follow-up new scheduling according to look-up table stores unit 30 is the subsequent packet of second multicast message, if the multicast group address of second multicast message of the multicast group address of the new multicast message of follow-up scheduling and look-up table stores unit 30 stored coupling, then the new multicast message of follow-up scheduling can be stored in the fifo_1 in the write-back push-up storage (fifo_0-fifo_15) corresponding with reg_1, guaranteeing can be not out of order.
Like this, constantly scheduling obtains all multicast messages and realizes duplicating all multicast messages.
In the above-mentioned steps 5, if the multicast message of write-back registers 501 storages is more than two, then 601 pairs of plural multicast messages of first scheduler module carry out robin scheduling, the result that robin scheduling is obtained imports in second scheduler module 602, second scheduler module 602 is weighted the Fair Queue scheduling from robin scheduling result and new multicast message, and then obtains carrying out that multicast message of replication processes.
And after the above-mentioned steps 5, in the above-mentioned steps 6, if duplicate ack unit 10 is confirmed second multicast message and is all duplicated the number of times K ' of needs smaller or equal to default multicast message replication processes times N, then second multicast message enters replication processes unit 20 and all duplicates duplicating and being sent out of the number of times K ' that needs, realizes that message all duplicates.
In the above-mentioned steps 5, if second scheduler module 602 confirms that entering the multicast message that duplicates replication processes unit 20 is first multicast message, what then first multicast message entered the multicast message replication processes times N preset replication processes unit 20 duplicates and sends the multicast message that duplicates, further, at this moment, if number of copy times has satisfied K time, then first multicast message has been finished all and has been duplicated; If number of copy times no show K time, write-back confirmation unit 40 also needs the first multicast message write-back so, and promptly write-back is kept in the reg_0 of write-back registers 501, waits follow-up being scheduled, and realizes that multicast message all duplicates.
Again with first multicast message, its multicast group address MGID0, its number of times K that duplicates needs is greater than default replication processes times N, and second multicast message, its multicast group address is MGID0, it duplicates the number of times K ' that needs is example greater than default replication processes times N, when specifically using the replication processes device of multicast message of present embodiment, comprising:
1: inlet scheduling unit 80 is that weight adopts Weighted Fair Queuing according to the priority of a plurality of multicast messages, obtains multicast message from inlet first in first out buffer memory 70, as first multicast message, and its multicast group address MGID0.
2: duplicate ack unit 10 receives first multicast message, and confirm first multicast message all duplicate the number of times K that needs and obtain user's replication processes times N default according to time-delay, and the whole number of times K that need of duplicating that confirm first multicast message are greater than user's replication processes times N default according to time-delay.
3: write-back confirmation unit 40 is set up look-up table stores unit 30, and the multicast group address MGID0 of first multicast message added in the look-up table stores unit 30, and first multicast message in replication processes unit 20 after replication processes N time, write-back confirmation unit 40 is stored in it in interior unappropriated register of write-back registers 501 (reg_0-reg_15), as reg_0, be scheduled etc. follow-up, realize that multicast message all duplicates.
4: receive the scheduling from inlet first in first out buffer memory 70 of inlet scheduling unit 80 and obtain new multicast message, as second multicast message, multicast group address is MGID0, the multicast group address of second multicast message is that the multicast group address MGID0 in MGID0 and the look-up table stores unit 30 is complementary, and illustrates that then second multicast message is the subsequent packet that has the same multicast group address with first multicast message.
5: write-back confirmation unit 40 stores second multicast message in the fifo_0 in the write-back push-up storage 502 (fifo_0-fifo_15) corresponding with reg_0 into, the subsequent packet that realization belongs to same multicast group is stored together, and second multicast message be stored in after first multicast message, still can guarantee that second multicast message is the subsequent packet of first multicast message, can be not out of order, all duplicate realizing.
Just should use-case and last application examples difference only be that the multicast group address of second multicast message is MGID0, second multicast message is the subsequent packet that has the same multicast group address with first multicast message, second multicast message need be stored in after first multicast message, guaranteeing can be not out of order.
Should be known in the replication processes device of the multicast message of present embodiment by top narration, when previous multicast message carried out replication processes, inlet scheduling unit 80 just can scheduling obtain next multicast message from inlet first in first out buffer memory 70.
It should be noted that, the scheduling from inlet first in first out buffer memory 70 of inlet scheduling unit 80 obtains multicast message, with the multicast group address MGID in the look-up table stores unit 30 coupling is arranged, but when corresponding write-back push-up storage 502 is expired, then this multicast message is abandoned, the subsequent packet that prevents to enter the mouth in the first in first out buffer memory 70 does not send earlier than duplicating the multicast message of finishing that the same multicast group address is arranged, and causes out of order.
And, when from inlet first in first out buffer memory 70, reading multicast message, need to satisfy in the inlet first in first out buffer memory 70 any non-NULL at least two push-up storages, write-back registers 501 is not taken fully simultaneously, can be described as look-up table stores unit 30 yet and is not taken fully.
The technical scheme that is provided by the embodiment of the invention described above as can be seen, when the receiving terminal number of multicast group quite a lot of, need under the situation of massive duplication, the user can be according to the replication processes number of times of the default multicast message of different delay requirements, the number of times that duplicates according to the whole needs of the message of reality with the replication processes number of times of default multicast message, multicast message is duplicated, reduced the time-delay that multicast message duplicates; And the user can dispose this time delay arbitrarily, and is convenient, flexible, applied widely.And adopted the write-back mechanism of multicast message, made the number of copy times of multicast message reach the number of times that whole needs duplicate; Introduced the look-up table unit, reduced not duplicate fully the multicast message of finishing and after storage, may produce out of order situation.
Through the above description of the embodiments, those skilled in the art can be well understood to the present invention and can realize that can certainly pass through hardware, perhaps the combination of the two is implemented by the mode that software adds essential general hardware platform.Based on such understanding, the part that technical scheme of the present invention contributes to prior art in essence in other words can embody with the form of software product, this software module or computer software product can be stored in the storage medium, comprise that some instructions are with so that a computer equipment (can be a personal computer, server, the perhaps network equipment etc.) carry out the described method of each embodiment of the present invention.Storage medium can be the storage medium of any other form known in random asccess memory (RAM), internal memory, read-only memory (ROM), electrically programmable ROM, electrically erasable ROM, register, hard disk, moveable magnetic disc, CD-ROM or the technical field.
The above; only for the preferable embodiment of the present invention, but protection scope of the present invention is not limited thereto, and anyly is familiar with those skilled in the art in the technical scope that the present invention discloses; the variation that can expect easily or replacement all should be encompassed within protection scope of the present invention.Therefore, protection scope of the present invention should be as the criterion with the protection range of claim.
Claims (11)
1. the duplication method of a multicast message is characterized in that, comprising:
The receiving group message, multicast group address according to described multicast message, confirm the number of times that described multicast message all duplicates to be needed, and obtain default multicast message replication processes number of times that described default multicast message replication processes number of times is the replication processes number of times default according to delay requirement;
When described multicast message all duplicates the number of times that needs greater than described default multicast message replication processes number of times, add the multicast group address of described multicast message to look-up table, duplicate and send the multicast message that duplicates to what described multicast message carried out described default multicast message replication processes number of times, and store described multicast message, described look-up table is used to store the number of times that all the duplicates needs multicast group address greater than the multicast message of described default multicast message replication processes number of times.
2. the duplication method of multicast message according to claim 1, it is characterized in that, described method also comprises: receive new multicast message, determine to carry out in the multicast message of described new multicast message and described storage the step of the multicast message of replication processes:
When the multicast group address in the multicast group address of new multicast message and the described look-up table mates, the subsequent packet of the multicast message of the multicast group address correspondence that described new multicast message is described coupling, described new multicast message by corresponding stored behind the multicast message of the multicast group address correspondence of described coupling;
Perhaps, when the multicast group address in the multicast group address of new multicast message and the described look-up table does not match,, dispatch the multicast message that obtains carrying out replication processes according to the priority of default new multicast message with the multicast message of storage.
3. the duplication method of multicast message according to claim 2 is characterized in that, the multicast group address in the multicast group address of new multicast message and the described look-up table does not match:
Scheduling obtains described new multicast message and carries out replication processes, when described new multicast message all duplicates the number of times that needs smaller or equal to described default multicast message replication processes number of times, described new multicast message all duplicated duplicating of the number of times that needs, and send the multicast message duplicate
Perhaps, when described new multicast message all duplicates the number of times that needs greater than described default multicast message replication processes number of times, described new multicast message is carried out duplicating of described default multicast message replication processes number of times, and send the multicast message duplicate, store described new multicast message and the multicast group address of described new multicast message is added in the described look-up table;
Perhaps, the multicast message that scheduling obtains described storage carries out replication processes, and the multicast message of described storage is carried out replication processes, and the number of copy times of the multicast message of described storage does not reach it when all duplicating the number of times that needs, the multicast message of the described storage of write-back.
4. the duplication method of multicast message according to claim 1 is characterized in that, obtains receiving the order of each described multicast message according to the priority scheduling of a plurality of multicast messages.
5. according to the duplication method of claim 2 or 3 or 4 described multicast messages, it is characterized in that Weighted Fair Queuing, weighted round robin or robin scheduling are adopted in described scheduling.
6. the replication processes device of a multicast message is characterized in that, comprising:
Duplicate ack unit: be used for the receiving group message, according to described multicast message multicast group address, confirm the number of times that described multicast message all duplicates to be needed, and obtaining default multicast message replication processes number of times, described default multicast message replication processes number of times is the replication processes number of times default according to delay requirement;
Replication processes unit: when being used for number of times that described multicast message all duplicates needs greater than described default multicast message replication processes number of times, duplicate and send the multicast message that duplicates to what described multicast message carried out described default multicast message replication processes number of times, and store described multicast message;
Look-up table stores unit: be used to store the number of times that all duplicates needs multicast group address greater than the multicast message of described default multicast message replication processes number of times.
7. the replication processes device of multicast message according to claim 6 is characterized in that, described device also comprises:
The write-back confirmation unit: the multicast group address that is used to confirm multicast message whether with described look-up table stores unit in the multicast group address coupling, be used to confirm whether multicast message needs storage or write-back, and be used to manage described look-up table stores unit.
8. the replication processes device of multicast message according to claim 7 is characterized in that, described device also comprises:
Write-back buffer: it comprises at least one write-back registers and at least one corresponding write-back push-up storage, wherein,
Described at least one write-back registers, be used for according to the described number of times of needs that all duplicates of multicast group address storage greater than described default multicast message replication processes number of times, and with the unmatched multicast message of multicast group address in the described look-up table stores unit, and the described multicast message of write-back;
Described at least one write-back push-up storage is used to store the subsequent packet of multicast message in the described write-back registers, described subsequent packet be multicast group address with described write-back registers in the multicast message that mates of the multicast group address of multicast message.
9. the replication processes device of multicast message according to claim 8 is characterized in that, described device also comprises:
Duplicate scheduling unit, be used for confirming that the multicast message in multicast message and the described write-back registers carries out the multicast message of replication processes.
10. the replication processes device of multicast message according to claim 9 is characterized in that, the described scheduling unit that duplicates comprises:
First scheduler module is used for the multicast message in two the above write-back registers is dispatched;
Second scheduler module is used for the multicast message that obtains entering described replication processes unit from the scheduling result and the scheduling of described multicast message of described first scheduler module;
Described scheduling mode is Weighted Fair Queuing, weighted round robin or robin scheduling.
11. the replication processes device of multicast message according to claim 6 is characterized in that, described device also comprises:
Inlet first in first out buffer memory: be used to receive a plurality of multicast messages and store described a plurality of multicast message according to its priority;
Inlet scheduling unit: be used for priority, adopt Weighted Fair Queuing, weighted round robin or robin scheduling obtain entering the mouth each multicast message in the first in first out buffer memory to send to the order of duplicate ack unit according to a plurality of multicast messages.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200910081159XA CN101515889B (en) | 2009-04-03 | 2009-04-03 | Replication and processing method for multicast message and device thereof |
| PCT/CN2010/071345 WO2010111928A1 (en) | 2009-04-03 | 2010-03-26 | Multicast message replication processing method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200910081159XA CN101515889B (en) | 2009-04-03 | 2009-04-03 | Replication and processing method for multicast message and device thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101515889A CN101515889A (en) | 2009-08-26 |
| CN101515889B true CN101515889B (en) | 2011-07-06 |
Family
ID=41040190
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN200910081159XA Expired - Fee Related CN101515889B (en) | 2009-04-03 | 2009-04-03 | Replication and processing method for multicast message and device thereof |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN101515889B (en) |
| WO (1) | WO2010111928A1 (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101515889B (en) * | 2009-04-03 | 2011-07-06 | 华为技术有限公司 | Replication and processing method for multicast message and device thereof |
| CN101854307B (en) * | 2010-06-09 | 2013-01-23 | 中国人民解放军国防科学技术大学 | Processing method of network node memory congestion in delay-tolerant network |
| CN102347890B (en) * | 2010-08-04 | 2014-03-26 | 杭州华三通信技术有限公司 | Processing method and device of multicast message |
| CN104270316B (en) * | 2014-09-23 | 2017-05-17 | 烽火通信科技股份有限公司 | Efficient multicast realization system and method in transmission equipment |
| CN110995603A (en) * | 2019-12-09 | 2020-04-10 | 广州信天翁信息科技有限公司 | Many-to-many transmission method of data and heterogeneous data transmission layer |
| CN111884942A (en) * | 2020-08-03 | 2020-11-03 | 航天宏图信息技术股份有限公司 | Multicast data transmission method, device, receiving host and multicast system |
| CN115460178A (en) * | 2022-08-31 | 2022-12-09 | 新华三半导体技术有限公司 | Multicast copying method and device and chip functional module |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1503507A (en) * | 2002-11-26 | 2004-06-09 | ��Ϊ��������˾ | Method for realizing registration from multicast source to collection point |
| CN101106515A (en) * | 2006-07-10 | 2008-01-16 | 华为技术有限公司 | Service quality guarantee method and system in multicast network |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1323531C (en) * | 2003-07-02 | 2007-06-27 | 华为技术有限公司 | A method for implementing multicast on transmission network of asynchronous transfer mode |
| KR20070061066A (en) * | 2005-12-08 | 2007-06-13 | 한국전자통신연구원 | Apparatus and method for forwarding multi-cast packet of router |
| JP2007228227A (en) * | 2006-02-23 | 2007-09-06 | Fujitsu Ltd | Communication device |
| US7978698B2 (en) * | 2006-03-16 | 2011-07-12 | Panasonic Corporation | Terminal for performing multiple access transmission suitable to a transmission path having varied characteristics |
| CN101030876A (en) * | 2007-04-06 | 2007-09-05 | 中兴通讯股份有限公司 | Method for realizing PPPoE packet-broadcasting service |
| CN101515889B (en) * | 2009-04-03 | 2011-07-06 | 华为技术有限公司 | Replication and processing method for multicast message and device thereof |
-
2009
- 2009-04-03 CN CN200910081159XA patent/CN101515889B/en not_active Expired - Fee Related
-
2010
- 2010-03-26 WO PCT/CN2010/071345 patent/WO2010111928A1/en active Application Filing
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1503507A (en) * | 2002-11-26 | 2004-06-09 | ��Ϊ��������˾ | Method for realizing registration from multicast source to collection point |
| CN101106515A (en) * | 2006-07-10 | 2008-01-16 | 华为技术有限公司 | Service quality guarantee method and system in multicast network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101515889A (en) | 2009-08-26 |
| WO2010111928A1 (en) | 2010-10-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101515889B (en) | Replication and processing method for multicast message and device thereof | |
| US5684797A (en) | ATM cell multicasting method and apparatus | |
| US6618752B1 (en) | Software and method for multicasting on a network | |
| CN1622645A (en) | System and method for synchronizing of information without data duplication | |
| CN101729407B (en) | Low delay jitter exchanging method and equipment based on unicast and multicast differentiated treatment | |
| US6744741B1 (en) | System and method for maintaining a plurality of media conferences | |
| CN101267331A (en) | A search method and device for multicast forward table | |
| CN101420390A (en) | Internet instant communication data transmission method, apparatus and system | |
| CN109067578A (en) | A kind of method and apparatus of rapidly channel switching | |
| US20030189942A1 (en) | Crosspoint switch having multicast functionality | |
| CN103220258A (en) | Conference sound mixing method, terminal and media resource server (MRS) | |
| CN101286866B (en) | Multicast implementing method and system based on switching network of high-speed peripheral extended interface | |
| CN101924910B (en) | Data sending method, receiving method and device during channel switching process | |
| CN109547310B (en) | Modular bus data receiving method | |
| CN101459615A (en) | Method, terminal port and system for packet transmission | |
| US9166804B2 (en) | Packet transmission control device, packet transmission control method, and program | |
| CN102684983B (en) | A kind of cell scheduling method and apparatus | |
| CN103701721A (en) | Message transmission method and device | |
| CN101459585A (en) | P2P stream media data request method | |
| CN102170364A (en) | Multicast communication method and device for broadband remote access server | |
| CN104022965A (en) | Message dequeue scheduling method and equipment thereof | |
| TW201540076A (en) | Method and system for providing a video service | |
| CN1300992C (en) | Method of realizing multitransmission | |
| CN100512421C (en) | Resource scheduling method for implementing image processing in conference telephone system | |
| CN101867528B (en) | Multicast service processing method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20110706 Termination date: 20180403 |