CN101227306A - Method and apparatus for copying message for multi-broadcast treatment - Google Patents
Method and apparatus for copying message for multi-broadcast treatment Download PDFInfo
- Publication number
- CN101227306A CN101227306A CNA2008100072470A CN200810007247A CN101227306A CN 101227306 A CN101227306 A CN 101227306A CN A2008100072470 A CNA2008100072470 A CN A2008100072470A CN 200810007247 A CN200810007247 A CN 200810007247A CN 101227306 A CN101227306 A CN 101227306A
- Authority
- CN
- China
- Prior art keywords
- message
- multicast
- information
- copy method
- heading
- 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.)
- Pending
Links
Images
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention provides a method and a device for coping messages which are used in multicast functions, the method comprises following steps: obtaining message body addresses of the message which needs the multicast functions, obtaining multicast group information about the message and multicast member information of each member in a linked list which is pointed by the multicast group information, using the multicast member information of each member and the message body addresses of the message to construct corresponding message heads, adding the message heads into the message and sending the message heads to the member. When the invention is carrying out the multicast functions, a network processor can reach the wire speed processing ability.
Description
Technical field
The present invention relates to the communications field, be specifically related to a kind of packet copy method and device that multicast (broadcasting, multicast) is handled that be used for.
Background technology
Broadcasting is that message is sent to All hosts in the network (the normally local network that links to each other), and multicast is the particular host group that message is sent to network.Broadcasting and multicast are applicable to the application that message need be sent to simultaneously a plurality of recipients.
Ip multicast (also claiming multi-address broadcast or multicast) technology is that a kind of one or more main frame of permission (multicast source) sends single message to the multiple host TCP/IP network technology of (once, while).Multicast can be saved the network bandwidth effectively as some communication to multiple spot.Multicast can make one or more multicast sources only message be sent to specific multicast group, and has only the main frame that adds this multicast group just can receive message.At present, the ip multicast technology is widely used in aspects such as network audio/video broadcasting, the network teleconference.
The simplest at present multicast is transmitted way and exactly original message is copied as many parts of messages, carries out the message encapsulation then respectively, sends respectively.Owing to exist a large amount of IO (input and output) operation, thereby the message replication performance is low.
At above-mentioned defective, a kind of method that multicast is transmitted that realizes on Intel IXP 2400 network processing units is provided in the prior art, this method comprises: receiver module (RX) receives from the next message of external port, deliver to package module (EM) after disposing by processing module, package module (EM) is searched multicast in abutting connection with encapsulating sheet (DMAT) according to multicast index (MID), multicast is a chained list in abutting connection with encapsulating sheet, and each preserves an outlet message encapsulation format information; Newly go out buffer of team then, be used to preserve multicast message head buffer, the heading content is saved among this buffer, structure multicast message head META (message descriptor); The META chain before message SOP (Start Of Package, heading) META, is formed a message chained list; Message is delivered to queue management module (QM), send scheduler module (TXS) scheduling packets and give sending module (TX), final sending module sends message.This method is by only revising heading, and the copy packet body does not duplicate to reduce message, improves the message replication performance.
Buffer is the memory field that is used to preserve message content in the processor.Wherein, each buffer correspondence two internal memory: data DATA (being used to preserve the actual content of message) and message descriptor META (preserving information such as message size).Buffer chain be a plurality of buffer can couple together the expression message, to hold than long message.
Yet, need TX resume module buffer chain, promptly need to handle two buffer in a process of transmitting, so increased considerably the burden of sending module (TX), influenced the message replication performance.
Summary of the invention
The present invention aims to provide a kind of packet copy method and device that multicast is handled that be used for, and can solve the TX problem that over-burden in the above-mentioned prior art.
According to one embodiment of present invention, provide a kind of packet copy method that multicast is handled that is used for, may further comprise the steps: the message body address that obtains the message of wanting the multicast processing; Obtain multicast information about firms about each member in the multicast group information of this message and the multicast group information chained list pointed; And the message body address of using each member's multicast information about firms and message, construct its corresponding heading, add in the message, and send to this member.
According to another embodiment of the present invention, provide a kind of message reproducing unit that multicast is handled that is used for, having comprised: first acquisition module, the message body address that is used to obtain the message of wanting the multicast processing; Second acquisition module is used for obtaining the multicast information about firms about the multicast group information of message and multicast group information each member of chained list pointed; And replication module, be used to use the message body address of each member's multicast information about firms and message, construct its corresponding heading, add in the message, and send to this member.
In above-mentioned prior art, by constructing a plurality of heading buffer, then heading buffer is linked on the buffer of message load, in n reproduction process, need n+1 buffer to go out team and n+1 buffer joins the team.And in the packet copy method of the embodiment of the invention and device, only need directly to revise heading, and avoided queue operation of the prior art, so alleviated the burden of TX module, when carrying out the multicast processing, can make network processing unit reach the linear speed disposal ability.
Description of drawings
Accompanying drawing described herein is used to provide further understanding of the present invention, constitutes the application's a part, and illustrative examples of the present invention and explanation thereof are used to explain the present invention, do not constitute improper qualification of the present invention.In the accompanying drawings:
Fig. 1 shows the flow chart of the packet copy method that is used for the multicast processing according to an embodiment of the invention;
Fig. 2 shows META form according to an embodiment of the invention;
Fig. 3 shows the block diagram of the message reproducing unit that is used for the multicast processing according to an embodiment of the invention.
Embodiment
Below with reference to the accompanying drawings and in conjunction with the embodiments, describe the present invention in detail.
Fig. 1 shows the flow chart of the packet copy method that is used for the multicast processing according to an embodiment of the invention, and it may further comprise the steps:
Step S10, the message body address that obtains the message of wanting the multicast processing;
Step S20 obtains the multicast information about firms about each member in the multicast group information of this message and the multicast group information chained list pointed; And
Step S30, its corresponding heading is constructed in the message body address of using each member's multicast information about firms and message, adds in the message, and sends to this member.
Multicast group information has been preserved member's number of this multicast group, the head pointer of multicast member linked list; The multicast information about firms has been preserved this multicast member's MAC, outgoing interface, Vlan information, and next member's chain list index.
In above-mentioned prior art, by constructing a plurality of heading buffer, then heading buffer is linked on the buffer of message load, in n reproduction process, need n+1 buffer to go out team and n+1 buffer joins the team.And in this packet copy method, only need directly to revise heading, and avoided queue operation of the prior art, so alleviated the burden of TX module, when carrying out the multicast processing, can make network processing unit reach the linear speed disposal ability.
Multicast group information has been preserved member's number of this multicast group, the head pointer of multicast member linked list; The multicast information about firms has been preserved this multicast member's MAC, outgoing interface, Vlan information, and next member's chain list index.
No matter be multicast or broadcasting, all corresponding unique multicast index.Step S20 specifically can comprise: extract the multicast index from multicast group information; (multicast address or broadcast domain ID) searches the multicast membership table according to the multicast index; From the multicast membership table, obtain each member's relevant information.
Preferably, message adopts the META form, and the read_offset field of message body address from the META form of message obtains.After receiving message, IP position can preserving message is read_offset.
Preferably, step S30 specifically comprises: pairing heading of member of every structure, and the total quantity that then will remain the multicast member who does not send subtracts 1; Next member in the pointed chained list; If total quantity is non-vanishing, then next member is constructed heading and add in the message according to pointer.
Message adopts the META form, and total quantity can be the mul_count field in the META form.
Message adopts the META form, and then Gou Zao heading is the L2 head of message.
Fig. 2 shows META form according to an embodiment of the invention.For example message adopts message descriptor META form, and then the heading deviant is the read_offset field in the META form among the figure, and total quantity is the mul_count field in the META form among the figure.
Like this, processing module need be preserved IP place skew of original message in META when handling multicast message; The TX module is revised corresponding L2/L3 heading according to the multicast information about firms; After sending this message, if judging not duplicate finishes, just this message is returned, the relevant information of message is write Scratch Ring, carry out subsequent modification for processing module.
The present invention can use for example above-mentioned packet copy method of (Intel IXP 2400 or Intel IXP 2800) execution of Intel IXP series processors, and the heading of copy packet specifically can comprise: the buffer copy packet head that uses Intel IXP series processors.Intel IXP series processors can be handled the multicast message of META form.Can use the Scratch Ring (hardware loop structure) of Intel IXP series processors to realize the structure of heading.Scratch Ring compares the software ring, and access speed is faster, and bandwidth is bigger.
This scheme is carried out n time an of multicast message and is duplicated, and n unicast message process is similar with handling, thereby can guarantee the handling property of system.Owing to do not had the participation of RX module, so the efficient of duplicating for n time is higher.Through actual test, this replication theme can reach the 10G wire-speed forwarding capability on the IntelIXP2800 platform.
According to one embodiment of present invention, adopt Intel IXP network processing unit to carry out the above-mentioned packet copy method that multicast is handled that is used for, wherein message adopts META form shown in Figure 2, specifically may further comprise the steps:
First part of copy packet:
1) message IP head skew (read_offset in the META form) is kept in the message descriptor (META);
2) this multicast being duplicated total umber (mul_count) that need duplicate is kept among the META;
3) read the multicast information about firms,, revise corresponding L2/L3 heading according to the multicast information about firms;
4) message being sent to the TX module sends.This message if mul_count is non-vanishing, is then write the relevant information of this message the Scratch Ring of processing module correspondence, thereby is carried out subsequent treatment after sending and finishing.
The processing of subsequent duplicate message:
1) reads the read_offset that preserves among the META, find IP position of message;
2) read the multicast information about firms,, revise corresponding L2/L3 heading according to the multicast information about firms;
3) message being sent to the TX module sends.This message if mul_count is non-vanishing, is then write the relevant information of this message the Scratch Ring of processing module correspondence, thereby is carried out subsequent treatment after sending and finishing.
When this embodiment has made full use of multicast message and has duplicated; have only heading different; and the message body duplicate characteristics that are message content; carrying out multicast message when duplicating; reuse same buffer, only revise the heading part among this buffer, thereby avoided fully in the message reproduction process; the repeatedly buffer application/release that can occur usually, perhaps message content duplicates problem.Simultaneously, owing to adopted the message replication theme of single buffer, and single buffer handling property of TX is the highest, thereby this scheme has farthest improved the treatment effeciency of TX module.
Fig. 3 shows the block diagram of the message reproducing unit that is used for the multicast processing according to an embodiment of the invention, comprising:
Can adopt Intel IXP series network processing unit to realize above-mentioned message reproducing unit.This scheme of this message reproducing unit has guaranteed not reduce TX and RX handling property as can be seen, thereby has guaranteed message replication performance and the same performance of clean culture processing, can realize the surface speed forwarding of respective processor.
As can be seen from the above description, the present invention has realized following technique effect:
1) only needs the copy packet head, avoided queue operation of the prior art;
2) alleviated the burden of TX module, when carrying out multicast and handle, can make network processing unit reach the linear speed disposal ability.
Obviously, those skilled in the art should be understood that, above-mentioned each module of the present invention or each step can realize with the general calculation device, they can concentrate on the single calculation element, perhaps be distributed on the network that a plurality of calculation element forms, alternatively, they can be realized with the executable program code of calculation element, thereby, they can be stored in the storage device and carry out by calculation element, perhaps they are made into each integrated circuit modules respectively, perhaps a plurality of modules in them or step are made into the single integrated circuit module and realize.Like this, the present invention is not restricted to any specific hardware and software combination.
The above is the preferred embodiments of the present invention only, is not limited to the present invention, and for a person skilled in the art, the present invention can have various changes and variation.Within the spirit and principles in the present invention all, any modification of being done, be equal to replacement, improvement etc., all should be included within protection scope of the present invention.
Claims (10)
1. one kind is used for the packet copy method that multicast is handled, and it is characterized in that, may further comprise the steps:
Obtain the message body address of the message of wanting the multicast processing;
Obtain multicast information about firms about each member in the multicast group information of described message and the described multicast group information chained list pointed; And
Use described each member's the multicast information about firms and the message body address of described message, construct its corresponding heading, add in the described message, and send to this member.
2. packet copy method according to claim 1 is characterized in that, the multicast information about firms that obtains each member specifically comprises:
From described multicast group information, extract the multicast index;
According to described multicast index search multicast membership table;
From described multicast membership table, obtain described each member's relevant information.
3. packet copy method according to claim 1 is characterized in that, described message adopts the message descriptor format, and the read_offset field of message body address from described message descriptor format of described message obtains.
4. packet copy method according to claim 3 is characterized in that, also comprises:
IP the position of preserving described message is read_offset.
5. packet copy method according to claim 1 is characterized in that, uses described each member's the multicast information about firms and the message body address of described message, constructs its corresponding heading and add in the described message specifically to comprise:
Pairing heading of described member of every structure, the total quantity that then will remain the described multicast member who does not send subtracts 1;
Next member in the described chained list of pointed;
If described total quantity is non-vanishing, then the described member of the next one is constructed heading and add in the described message according to described pointer.
6. packet copy method according to claim 5 is characterized in that, described message adopts the message descriptor format, and described total quantity is the mul_count field in the described message descriptor format.
7. packet copy method according to claim 1 is characterized in that, described message adopts the message descriptor format, and the described heading of structure is the L2 head of described message.
8. according to each described packet copy method of claim 1 to 7, it is characterized in that, use Intel IXP series processors to carry out described packet copy method.
9. packet copy method according to claim 8 is characterized in that, uses the hardware loop structure of described Intel IXP series processors to realize the structure of described heading.
10. one kind is used for the message reproducing unit that multicast is handled, and it is characterized in that, comprising:
First acquisition module, the message body address that is used to obtain the message of wanting the multicast processing;
Second acquisition module is used for obtaining the multicast information about firms about the multicast group information of described message and described multicast group information each member of chained list pointed; And
Replication module is used to use the message body address of described each member's multicast information about firms and described message, constructs its corresponding heading, adds in the described message, and sends to this member.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2008100072470A CN101227306A (en) | 2008-02-19 | 2008-02-19 | Method and apparatus for copying message for multi-broadcast treatment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2008100072470A CN101227306A (en) | 2008-02-19 | 2008-02-19 | Method and apparatus for copying message for multi-broadcast treatment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101227306A true CN101227306A (en) | 2008-07-23 |
Family
ID=39859075
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2008100072470A Pending CN101227306A (en) | 2008-02-19 | 2008-02-19 | Method and apparatus for copying message for multi-broadcast treatment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101227306A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107276905A (en) * | 2016-04-08 | 2017-10-20 | 思科技术公司 | Asymmetric multi-destination flow in overlay network is replicated |
| CN107707546A (en) * | 2017-09-29 | 2018-02-16 | 江苏农林职业技术学院 | A kind of packet copy method and the new network adapter for supporting message copy function |
| CN109302349A (en) * | 2018-09-29 | 2019-02-01 | 迈普通信技术股份有限公司 | A kind of message forwarding method and routing device |
| CN110134086A (en) * | 2019-05-17 | 2019-08-16 | 温州大学 | Realize the data collection and control device of middleware function in a kind of digitlization workshop |
| CN115460134A (en) * | 2022-09-05 | 2022-12-09 | 国网智能电网研究院有限公司 | MEC data multicast forwarding method for power 5G service |
-
2008
- 2008-02-19 CN CNA2008100072470A patent/CN101227306A/en active Pending
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107276905A (en) * | 2016-04-08 | 2017-10-20 | 思科技术公司 | Asymmetric multi-destination flow in overlay network is replicated |
| CN107276905B (en) * | 2016-04-08 | 2021-03-12 | 思科技术公司 | Asymmetric multi-destination traffic replication in overlay networks |
| CN107707546A (en) * | 2017-09-29 | 2018-02-16 | 江苏农林职业技术学院 | A kind of packet copy method and the new network adapter for supporting message copy function |
| CN107707546B (en) * | 2017-09-29 | 2020-02-18 | 江苏农林职业技术学院 | Message copying method and novel network adapter supporting message copying function |
| CN109302349A (en) * | 2018-09-29 | 2019-02-01 | 迈普通信技术股份有限公司 | A kind of message forwarding method and routing device |
| CN110134086A (en) * | 2019-05-17 | 2019-08-16 | 温州大学 | Realize the data collection and control device of middleware function in a kind of digitlization workshop |
| CN110134086B (en) * | 2019-05-17 | 2021-10-08 | 温州大学 | Data acquisition and control device for realizing middleware function in digital workshop |
| CN115460134A (en) * | 2022-09-05 | 2022-12-09 | 国网智能电网研究院有限公司 | MEC data multicast forwarding method for power 5G service |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101304389B (en) | Method, apparatus and system for processing packet | |
| US9299111B2 (en) | Efficient presence distribution mechanism for a large enterprise | |
| CN101605103B (en) | Method and apparatus for static forwarding multicast data | |
| CN101267331A (en) | A search method and device for multicast forward table | |
| US9094219B2 (en) | Network processor having multicasting protocol | |
| CN101894092B (en) | Multi-core CPU and inter-core communication method thereof | |
| CN101227306A (en) | Method and apparatus for copying message for multi-broadcast treatment | |
| US9942054B2 (en) | Systems and methods for an extranet multicast virtual private network in a virtual routing and forwarding based customer edge device | |
| CN105656808A (en) | Message processing method and system thereof | |
| CN102164084A (en) | Multicast message forwarding method and equipment thereof | |
| CN106685827A (en) | Downlink message forwarding method and AP device | |
| CN102891809A (en) | Interface order-preserving method and system of messages of multi-core network device | |
| CN103986585A (en) | Message preprocessing method and device | |
| CN101771611A (en) | Method for precisely forwarding IP multicast data in VLAN and forward device | |
| CN108259348B (en) | Message transmission method and device | |
| CN101312400B (en) | Method for realizing IGMP interception function on modem device with 4 ethernet interfaces | |
| CN107483341A (en) | A kind of across fire wall packet fast forwarding method and device | |
| CN102868636A (en) | Method and system for stream-based order preservation of multi-core network equipment packet | |
| CN105429881A (en) | Multicast message forwarding method and device | |
| CN103067304A (en) | Method and device of message order-preserving | |
| Baltas et al. | Ultra low delay switching for networked music performance | |
| CN103368966A (en) | Server system and transmission method for data flow | |
| CN110061922B (en) | Message forwarding method and device | |
| CN110660205B (en) | Alarm information processing method and device, electronic equipment and readable storage medium | |
| CN106209666A (en) | A kind of link multiplexing method based on load equalizer and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Open date: 20080723 |