CN108134747A - The realization method and system of Ethernet switching chip, its multicast mirror image flow equalization - Google Patents
The realization method and system of Ethernet switching chip, its multicast mirror image flow equalization Download PDFInfo
- Publication number
- CN108134747A CN108134747A CN201711408283.3A CN201711408283A CN108134747A CN 108134747 A CN108134747 A CN 108134747A CN 201711408283 A CN201711408283 A CN 201711408283A CN 108134747 A CN108134747 A CN 108134747A
- Authority
- CN
- China
- Prior art keywords
- processing engine
- direction processing
- engine unit
- message
- source messages
- 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.)
- Withdrawn
Links
- 238000000034 method Methods 0.000 title claims abstract description 39
- 230000010076 replication Effects 0.000 claims description 11
- 238000010586 diagram Methods 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 238000002224 dissection Methods 0.000 description 2
- 238000007363 ring formation reaction Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000003362 replicative effect Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000013024 troubleshooting Methods 0.000 description 1
- 238000005303 weighing Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/12—Avoiding congestion; Recovering from congestion
- H04L47/125—Avoiding congestion; Recovering from congestion by balancing the load, e.g. traffic engineering
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L49/00—Packet switching elements
- H04L49/20—Support for services
- H04L49/201—Multicast operation; Broadcast operation
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Small-Scale Networks (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The present invention provides a kind of Ethernet switching chip, the realization method and system of its multicast mirror image flow equalization, and Ethernet switching chip includes：Enter direction processing engine modules, outgoing direction processing engine modules and storage forwarding module, it is described enter direction processing engine modules be connected with storage forwarding module, storage forwarding module and the outgoing direction handle engine modules and are connected；Wherein, it is described enter direction processing engine modules include at least two enter direction handle engine unit；The outgoing direction processing engine modules outgoing direction identical with entering direction processing engine unit quantity handles engine unit；The storage forwarding module includes：Loopback unit and message copied cells；The present invention can effectively improve the bandwidth availability ratio of exchange chip and the process bandwidth of message, can imitate avoid causing message be looped back to it is same enter bandwidth processing ability caused by direction processing engine unit is low and the phenomenon that packet loss.
Description
Technical field
The present invention relates to network communication field more particularly to a kind of Ethernet switching chip, its multicast mirror image flow equalizations
Realization method and system.
Background technology
In network communications, it is often necessary to flow in network be analyzed and be monitored, mirror image just refers to replicate flow
It is analyzed to specific destination, to carry out network detection and troubleshooting.In practical applications, it may be necessary to a flow
It copies to multiple destinations and carries out observation analysis or backup, it is therefore desirable to which network exchanging chip supports the mirror image of multi-destination.
The mirror image of message refers to that replicating source messages is sent to designated destination, does not influence the forwarding process of source messages, applies
Number Chinese patent application for being CN201611269913.9, proposes a kind of interchanger for realizing multi-destination message mirror and side
Method, by one group it is mutual corresponding enter direction handle engine modules and outgoing direction and handle engine modules, and pass through storage forwarding
Module is interconnected, to realize multi-destination message mirror；However, in high speed exchange chip, one group enters direction processing engine
The Message processing of high bandwidth can not be met with outgoing direction processing engine, message is looped back to one by the loopback module in above-mentioned motion
It is a enter direction processing engine modules in, can not meet high bandwidth message mirror processing.
Invention content
The purpose of the present invention is to provide a kind of Ethernet switching chip, the implementation method of its multicast mirror image flow equalization and
System.
One of to achieve the above object, the Ethernet switching chip of an embodiment of the present invention includes：Enter direction processing to draw
Hold up module, outgoing direction processing engine modules and storage forwarding module, it is described enter direction processing engine modules and storage forwarding mould
Block is connected, and the storage forwarding module is connected with outgoing direction processing engine modules；
Wherein, it is described enter direction processing engine modules include at least two enter direction handle engine unit；The outgoing direction
The processing engine modules outgoing direction identical with entering direction processing engine unit quantity handles engine unit；The storage forwarding module
Including：Loopback unit and message copied cells；
When message enters direction processing engine modules,
It is described enter direction processing engine modules be used to that one of them to be selected into fashionable to enter direction processing engine list in source messages
Member receives it, and the direction processing engine unit that enters for passing through selection parses source messages；
The storage forwarding module will form ring after the source messages image copying for entering direction processing engine unit for the first time
Return text, while source messages are forwarded to one of outgoing direction processing engine modules processing；
The loopback Unit selection loopback message is looped back to one of them enters direction processing engine unit；
When loopback Unit selection by loopback message loop back to one of them enter direction processing engine unit after,
It is described enter direction processing engine unit be additionally operable to, according to the ID of the purpose multicast group of loopback Receive message multicast mirror image
It is sent to message copied cells；
The message copied cells are used to be believed according to the ID of the purpose multicast group of its reception and the port of purpose multicast group
Breath carries out the multicast replication of multi-destination, and is selectively sent to one of outgoing direction processing engine modules processing.
One of to achieve the above object, multicast mirror image flow equalization in the Ethernet switching chip of an embodiment of the present invention
Implementation method include：In S1, source messages repeating process, the source messages be selectively entered one of them enter direction processing draw
Unit is held up, the source messages are parsed and be searched, obtains its corresponding forwarding destination and image banner；
S2, will be entered in direction processing engine unit by storing forwarding module according to the corresponding forwarding destination of source messages
Source messages are selectively sent to out one of outgoing direction processing engine unit, meanwhile, duplication shape is carried out to the source messages
Cyclization return text, and pass through the loopback unit loop back to one of them enter direction processing engine unit；
S3, according to loopback message by it is described enter direction handle engine unit and obtain the purpose group of its corresponding multicast mirror image
The ID of group is broadcast, and is sent to message copied cells, so that the ID of purpose multicast group that the message copied cells are received according to it
And the port information of purpose multicast group carries out the multicast replication of multi-destination, forms multiple message mirrors, and selectively send out
It send to the processing engine unit processing of one of outgoing direction；
S4, each message mirror is turned according to its corresponding port information by outgoing direction processing engine unit
Hair.
As being further improved for an embodiment of the present invention, the step S1 is specifically included：
In source messages repeating process, the magnitude range of inbound port number carried according to source messages is corresponding selectively into its
One of them entered in direction processing engine modules enters direction processing engine unit.
As being further improved for an embodiment of the present invention, the step S1 is specifically included：
It is obtained according to the interchanger inbound port number of the Ethernet switching chip or the VLAN characteristic informations of source messages is used
Take the corresponding image banner of source messages.
As being further improved for an embodiment of the present invention, in the step S2, according to the corresponding forwarding mesh of source messages
Ground by store forwarding module will enter direction handle engine unit in source messages be selectively sent to out one of them
Direction processing engine unit specifically includes：
The magnitude range of the inbound port number carried according to source messages is selectively sent to its corresponding outgoing direction processing and draws
Hold up mould one of outgoing direction processing engine unit in the block.
As being further improved for an embodiment of the present invention, in the step S2, duplication shape is carried out to the source messages
Cyclization return text, and pass through the loopback unit loop back to one of them enter direction processing engine unit specifically include：
The loading condition that loopback message is entered to processing engine unit in direction according to selectively loops back to one of them
Enter direction processing engine unit.
One of to achieve the above object, multicast mirror image flow equalization in the Ethernet switching chip of an embodiment of the present invention
Realization system include：Searching module is parsed, in source messages repeating process, the source messages being made to be selectively entered it
In one enter direction processing engine unit, the source messages are parsed and be searched, obtain its corresponding forwarding destination
And image banner；
Forwarding module, for according to source messages it is corresponding forwarding destination by store forwarding module will enter direction processing draw
It holds up the source messages in unit and is selectively sent to out one of outgoing direction processing engine unit, meanwhile, to the source messages
Carry out replicate form loopback message, and pass through the loopback unit loop back to one of them enter direction processing engine unit；
Multicast processing module, for according to loopback message by it is described enter direction handle engine unit obtain its corresponding group
The ID of the purpose multicast group of mirror image is broadcast, and is sent to message copied cells, so that the message copied cells were received according to it
The ID of purpose multicast group and the port information of purpose multicast group carry out the multicast replication of multi-destination, form multiple message mirrors
Picture, and selectively it is sent to one of outgoing direction processing engine unit processing；
Output module handles engine unit to each message mirror according to its corresponding port for passing through the outgoing direction
Information is forwarded.
As being further improved for an embodiment of the present invention, the parsing searching module is specifically used for：Turn in source messages
During hair, according to source messages carry inbound port number magnitude range selectively enter its it is corresponding enter direction handle engine mould
It is in the block one of them enter direction processing engine unit.
As being further improved for an embodiment of the present invention, the parsing searching module is specifically used for：
It is obtained according to the interchanger inbound port number of the Ethernet switching chip or the VLAN characteristic informations of source messages is used
Take the corresponding image banner of source messages.
As being further improved for an embodiment of the present invention, the forwarding module is specifically used for：It is carried according to source messages
The magnitude range of inbound port number be selectively sent to one of them side of going out in its corresponding outgoing direction processing engine modules
To processing engine unit.
As being further improved for an embodiment of the present invention, the forwarding module is specifically used for：By loopback message according to
It is described enter direction processing engine unit loading condition selectively loop back to one of them enter direction processing engine unit.
Compared with prior art, the beneficial effects of the invention are as follows：The Ethernet switching chip of the present invention, its multicast mirror image stream
Measure balanced realization method and system, in the multicast mirror image of multinuclear exchange chip, by set it is multiple enter direction handle engine
Unit, multiple outgoing directions processing engine unit；During message is handled, loopback module is according to mirror image multicast group information message
One of them enters direction processing unit to loopback, is assigned to so as to fulfill mirror image flow multiple in multinuclear Ethernet switching chip
Slice, each independent bandwidth allocations of slice realize the load balancing inside exchange chip, can effectively improve exchange chip
The process bandwidth of bandwidth availability ratio and message, it is possible to prevente effectively from cause message be looped back to it is same enter direction processing engine unit
Caused bandwidth processing ability is low and the phenomenon that packet loss.
Description of the drawings
Fig. 1 is the structure diagram of Ethernet switching chip in an embodiment of the present invention；
Fig. 2 is the stream of the implementation method of multicast mirror image flow equalization in Ethernet switching chip in an embodiment of the present invention
Journey schematic diagram；
Fig. 3 is the mould of the realization system of multicast mirror image flow equalization in Ethernet switching chip in an embodiment of the present invention
Block schematic diagram.
Specific embodiment
Below with reference to specific embodiment shown in the drawings, the present invention will be described in detail.But these embodiments are simultaneously
The present invention is not limited, structure that those of ordinary skill in the art are made according to these embodiments, method or functionally
Transformation is all contained in protection scope of the present invention.
Ethernet switching chip in an embodiment of the present invention includes：Enter direction processing engine modules, outgoing direction processing
Engine modules and storage forwarding module, it is described enter direction processing engine modules with store forwarding module be connected, the storage
Forwarding module is connected with outgoing direction processing engine modules；Wherein, it is described enter direction processing engine modules enter including at least two
Direction handles engine unit；The outgoing direction processing engine modules include at least two outgoing directions and handle engine unit；It is described to deposit
Storage forwarding module includes：Loopback unit and message copied cells；When message enters direction processing engine modules, the side of entering
It is used to one of them be selected into fashionable to enter direction processing engine unit in source messages to processing engine modules to receive it,
And the direction processing engine unit that enters for passing through selection parses source messages；The storage forwarding module enters the side of entering by first
Loopback message is formed after the source messages image copying of processing engine unit, while source messages are forwarded to one of outgoing direction
Handle engine modules processing；The loopback Unit selection loopback message is looped back to one of them enters direction processing engine list
Member；When loopback Unit selection by loopback message loop back to one of them enter direction processing engine unit after, it is described enter direction
Processing engine unit is additionally operable to, and being sent to message according to the ID of the purpose multicast group of loopback Receive message multicast mirror image replicates list
Member；The message copied cells are used to be carried out according to the ID of the purpose multicast group of its reception and the port information of purpose multicast group
The multicast replication of multi-destination, and selectively it is sent to one of outgoing direction processing engine modules processing.
As shown in Figure 1, in a specific example of the invention, multinuclear exchange chip sets two groups to enter direction processing engine list respectively
Member, two groups of outgoing direction processing engine units and storage forwarding module, to handle simultaneously message.Multinuclear exchange chip
Two slice are correspondingly arranged, are indicated respectively with slice0 and slice1, each independent bandwidth allocations of slice, so as to fulfill
Load balancing inside exchange chip, and then effectively improve the bandwidth availability ratio of exchange chip and the process bandwidth of message.This shows
In example, in the multinuclear Ethernet switching chip of two slice, including entering direction processing engine unit slice 0, enter direction processing
Engine unit slice1, outgoing direction processing engine unit slice0, outgoing direction processing engine unit slice1 and the above-mentioned side of entering
The storage forwarding module shared to processing engine unit and outgoing direction processing engine unit, wherein, storage forwarding module includes ring
Receipt member and message copied cells.
It is described enter direction processing engine unit slice0 and slice1 be used to message content search, obtain destination and
Edit action, the storage forwarding module is for queue scheduling, duplication of message etc., each outgoing direction processing engine unit
For editing message, and message is sent from corresponding port.
Specifically, please with reference to Fig. 2, above-mentioned Ethernet exchanging in an embodiment of the present invention specific embodiment is introduced
The implementation method of multicast mirror image flow equalization in chip；The method includes：
In S1, source messages repeating process, the source messages are selectively entered one of them and enter direction processing engine unit,
The source messages are parsed and be searched, its corresponding forwarding destination and image banner are obtained.
In the specific embodiment of the invention, for the ease of difference, the message just entered in interchanger is known as source messages, in source
When message enters interchanger, according to source messages carry inbound port number magnitude range selectively enter its it is corresponding enter direction at
One of them in reason engine modules enters direction processing engine unit, to be parsed and be searched to source messages, obtains outlet mesh
Ground and edit action.Such as：Interchanger supports 128 ports, can be during interchanger is adjusted, and designated ends slogan is 1-64
Source messages selection into interchanger enters direction processing engine unit slice 0, and designated ends slogan enters friendship for 65-128
The source messages selection changed planes enters direction processing engine unit slice 1.
It, can be according to the interchanger inbound port number or source for using the Ethernet switching chip in an embodiment of the present invention
The VLAN characteristic informations of message obtain the corresponding image banner of source messages.
Need what is illustrated, the corresponding forwarding destination of the source messages that step S1 is obtained is the corresponding common exit port of source messages
Number, i.e., when not doing image copying, normally forward port.
Further, the method includes：S2, forward destination will by storing forwarding module according to source messages are corresponding
Enter the source messages in direction processing engine unit and be selectively sent to out one of outgoing direction processing engine unit, meanwhile,
To the source messages carry out replicate form loopback message, and pass through the loopback unit loop back to one of them enter direction processing draw
Hold up unit.
In the specific embodiment of the invention, the step S2 is specifically included：According to the big of the inbound port number of source messages carrying
Small range is selectively sent to one of outgoing direction processing engine unit in its corresponding outgoing direction processing engine modules.
Such as：Interchanger supports 128 exit ports, can specify the source messages that forwarding port numbers are 1-64 during interchanger is adjusted
When producing exchange chip, outgoing direction processing engine unit slice 0 is selected, the source messages that forwarding port numbers is specified to be 65-128 turn
When going out exchange chip, outgoing direction processing engine unit slice 1 is selected.
In the specific embodiment of the invention, the step S2 is further included：Loopback message is entered direction processing to draw according to
Hold up unit loading condition selectively loop back to one of them enter direction processing engine unit.Such as：Under current state, if ring
Return text needs to loop back to handles engine modules into direction, then judges into direction processing engine unit slice 0 and enter at direction
Manage the flow load situation of engine unit slice 1, select its flow load situation it is small enter direction processing engine unit as
The loopback destination of loopback message, in this way, the processing speed of message can be accelerated, the occupied space of rational deployment exchange chip.
In the embodiment, loopback unit replicates the source messages come from into direction processing engine unit, and with
Mirror image multicast group is combined, and obtains loopback message, and loopback message is looped back to and handles engine unit into direction to do into one
The dissection process of step.
Further, the method further includes：S3, according to loopback message by it is described enter direction handle engine unit obtain it
The ID of the purpose multicast group of corresponding multicast mirror image, and be sent to message copied cells so that the message copied cells according to
The ID of its purpose multicast group received and the port information of purpose multicast group carry out the multicast replication of multi-destination, are formed multiple
Message mirror, and selectively it is sent to one of outgoing direction processing engine unit processing.
In the embodiment, enter direction processing engine unit and purpose multicast group is obtained from the forwarding head of loopback message, and
The ID for extracting its purpose multicast group simultaneously is sent to storage forwarding module；Message copied cells are according to the purpose multicast group of reception
ID and the port information of purpose multicast group carry out the multicast replication of multi-destination, obtain the message mirror of multi-destination, select
Property be sent to one of outgoing direction processing engine unit；When selecting outgoing direction processing engine unit, also according to carrying
The magnitude range of go out port numbers is selected, and is not described in detail herein.
Further, the method further includes：S4, each message mirror is pressed by outgoing direction processing engine unit
It is forwarded according to its corresponding port information.
After outgoing direction processing engine receives message mirror, to it into edlin, and by it according to corresponding end
Message breath is forwarded.
With reference to shown in Fig. 3, in one embodiment of the present invention, multicast mirror image flow is equal in above-mentioned Ethernet switching chip
The realization system of weighing apparatus includes：Parse searching module 100, forwarding module 200, multicast processing module 300, output module 400.
It parses searching module 100 to be used in source messages repeating process, the source messages is made to be selectively entered wherein one
It is a enter direction processing engine unit, the source messages are parsed and be searched, obtain its corresponding forwarding destination and mirror
As mark.
In the specific embodiment of the invention, for the ease of difference, parsing searching module 100 is used to just to enter in interchanger
Message is known as source messages, selective according to the magnitude range of inbound port number that source messages carry when source messages enter interchanger
Into its it is corresponding enter direction processing engine modules in one of them enter direction processing engine unit, to be solved to source messages
Analysis and lookup, obtain outlet destination and edit action.Such as：Interchanger supports 128 ports, can adjust interchanger process
In, the source messages selection that designated ends slogan enters interchanger for 1-64 enters direction processing engine unit slice 0, designated ends
The source messages selection that slogan enters interchanger for 65-128 enters direction processing engine unit slice 1.
In an embodiment of the present invention, parsing searching module 100 can be according to the exchange for using the Ethernet switching chip
The VLAN characteristic informations of machine inbound port number or source messages obtain the corresponding image banner of source messages.
Need what is illustrated, the corresponding forwarding destination of the source messages that parsing searching module 100 obtains is corresponding for source messages
Common go out port numbers when not doing image copying, normally forward port.
Forwarding module 200 is used to that direction processing will to be entered by storing forwarding module according to the corresponding forwarding destination of source messages
Source messages in engine unit are selectively sent to out one of outgoing direction processing engine unit, meanwhile, the source is reported
Text carry out replicate forms loopback message, and pass through the loopback unit loop back to one of them enter direction processing engine unit.
In the specific embodiment of the invention, forwarding module 200 is specifically used for：According to the big of the inbound port number of source messages carrying
Small range is selectively sent to one of outgoing direction processing engine unit in its corresponding outgoing direction processing engine modules.
Such as：Interchanger supports 128 exit ports, can specify the source messages that forwarding port numbers are 1-64 during interchanger is adjusted
When producing exchange chip, outgoing direction processing engine unit slice 0 is selected, the source messages that forwarding port numbers is specified to be 65-128 turn
When going out exchange chip, outgoing direction processing engine unit slice 1 is selected.
In the specific embodiment of the invention, forwarding module 200 is additionally operable to：Loopback message is entered direction processing to draw according to
Hold up unit loading condition selectively loop back to one of them enter direction processing engine unit.Such as：Under current state, if ring
Return text needs to loop back to handles engine modules into direction, then judges into direction processing engine unit slice 0 and enter at direction
Manage the flow load situation of engine unit slice 1, select its flow load situation it is small enter direction processing engine unit as
The loopback destination of loopback message, in this way, the processing speed of message can be accelerated, the occupied space of rational deployment exchange chip.
In the embodiment, the source that forwarding module 200 handles into direction coming from engine unit by loopback unit is reported
Text is replicated, and is combined with mirror image multicast group, obtains loopback message, and loopback message is looped back to and is drawn into direction processing
Unit is held up to do further dissection process.
Multicast processing module 300 be used for according to loopback message by it is described enter direction handle engine unit to obtain its corresponding
The ID of the purpose multicast group of multicast mirror image, and message copied cells are sent to, so that the message copied cells are received according to it
Purpose multicast group ID and purpose multicast group port information carry out multi-destination multicast replication, form multiple message mirrors
Picture, and selectively it is sent to one of outgoing direction processing engine unit processing.
In the embodiment, enter direction processing engine unit and purpose multicast group is obtained from the forwarding head of loopback message, and
The ID for extracting its purpose multicast group simultaneously is sent to storage forwarding module；Message copied cells are according to the purpose multicast group of reception
ID and the port information of purpose multicast group carry out the multicast replication of multi-destination, obtain the message mirror of multi-destination, select
Property be sent to one of outgoing direction processing engine unit；When selecting outgoing direction processing engine unit, also according to carrying
The magnitude range of go out port numbers is selected, and is not described in detail herein.
Output module 400 is used to handle engine unit to each message mirror according to its corresponding end by the outgoing direction
Message breath is forwarded.
After outgoing direction processing engine receives message mirror, to it into edlin, and by it according to corresponding end
Message breath is forwarded.
In conclusion the Ethernet switching chip of the present invention, the realization method and system of its multicast mirror image flow equalization,
In the multicast mirror image of multinuclear exchange chip, by set it is multiple enter direction handle engine unit, multiple outgoing directions handle engine list
Member；During message is handled, according to mirror image multicast group information, message loopback, one of them enters direction processing list to loopback module
Member, assigns to multiple slice in multinuclear Ethernet switching chip so as to fulfill mirror image flow, each independent bandwidth allocations of slice,
It realizes the load balancing inside exchange chip, the bandwidth availability ratio of exchange chip and the process bandwidth of message can be effectively improved,
It is possible to prevente effectively from cause message be looped back to it is same enter direction processing engine unit caused by bandwidth processing ability is low and packet loss
The phenomenon that.
For convenience of description, it describes to describe respectively so that various modules can be divided into during apparatus above.Certainly, implementing this hair
The function of each module is realized can in the same or multiple software and or hardware when bright.
Device embodiments described above are only schematical, wherein the module illustrated as separating component
It may or may not be physically separate, the component shown as module may or may not be physics mould
Block, you can be located at a place or can also be distributed on multiple network modules.It can be selected according to the actual needs
In some or all of module realize the purpose of present embodiment scheme.Those of ordinary skill in the art are not paying creation
Property labour in the case of, you can to understand and implement.
It should be appreciated that although this specification is described in terms of embodiments, but not each embodiment only includes one
A independent technical solution, this description of the specification is merely for the sake of clarity, and those skilled in the art should will say
For bright book as an entirety, the technical solution in each embodiment may also be suitably combined to form those skilled in the art can
With the other embodiment of understanding.
Those listed above is a series of to be described in detail only for feasibility embodiment of the invention specifically
Bright, they are not to limit the scope of the invention, all equivalent implementations made without departing from skill spirit of the present invention
Or change should all be included in the protection scope of the present invention.
Claims (11)
1. a kind of Ethernet switching chip, which is characterized in that including：Enter direction processing engine modules, outgoing direction processing engine mould
Block and storage forwarding module, it is described enter direction processing engine modules with storage forwarding module be connected, it is described storage forward mould
Block is connected with outgoing direction processing engine modules；
Wherein, it is described enter direction processing engine modules include at least two enter direction handle engine unit；The outgoing direction processing
The engine modules outgoing direction identical with entering direction processing engine unit quantity handles engine unit；The storage forwarding module packet
It includes：Loopback unit and message copied cells；
When message enters direction processing engine modules,
It is described enter direction processing engine modules be used to that one of them to be selected into fashionable to enter direction processing engine unit pair in source messages
It is received, and the direction processing engine unit that enters for passing through selection parses source messages；
The storage forwarding module will form loopback report after the source messages image copying for entering direction processing engine unit for the first time
Text, while source messages are forwarded to one of outgoing direction processing engine modules processing；
The loopback Unit selection loopback message is looped back to one of them enters direction processing engine unit；
When loopback Unit selection by loopback message loop back to one of them enter direction processing engine unit after,
It is described enter direction processing engine unit be additionally operable to, according to the ID of the purpose multicast group of loopback Receive message multicast mirror image send
To message copied cells；
The message copied cells be used for according to its reception the ID of purpose multicast group and the port information of purpose multicast group into
The multicast replication of row multi-destination, and selectively it is sent to one of outgoing direction processing engine modules processing.
2. the implementation method of multicast mirror image flow equalization in a kind of Ethernet switching chip, which is characterized in that the Ethernet is handed over
Chip is changed to include：Enter direction processing engine modules, outgoing direction processing engine modules and storage forwarding module, it is described enter direction
Processing engine modules are connected with storage forwarding module, and the storage forwarding module is connected with outgoing direction processing engine modules；
Wherein, it is described enter direction processing engine modules include at least two enter direction handle engine unit；The outgoing direction processing
The engine modules outgoing direction identical with entering direction processing engine unit quantity handles engine unit；The storage forwarding module packet
It includes：Loopback unit and message copied cells；
The method includes：
In S1, source messages repeating process, the source messages are selectively entered one of them and enter direction processing engine unit, with right
The source messages are parsed and are searched, and obtain its corresponding forwarding destination and image banner；
S2, according to source messages it is corresponding forwarding destination by store forwarding module will enter direction handle engine unit in source report
Text is selectively sent to out one of outgoing direction processing engine unit, meanwhile, the source messages replicate with forming ring
Return text, and pass through the loopback unit loop back to one of them enter direction processing engine unit；
S3, according to loopback message by it is described enter direction handle engine unit and obtain the purpose multicast group of its corresponding multicast mirror image
ID, and be sent to message copied cells so that the message copied cells according to the ID of its purpose multicast group received and
The port information of purpose multicast group carries out the multicast replication of multi-destination, forms multiple message mirrors, and be selectively sent to
One of outgoing direction processing engine unit processing；
S4, each message mirror is forwarded according to its corresponding port information by outgoing direction processing engine unit.
3. the implementation method of multicast mirror image flow equalization, feature exist in Ethernet switching chip according to claim 2
In the step S1 is specifically included：
In source messages repeating process, its corresponding side of entering selectively is entered according to the magnitude range of inbound port number that source messages carry
One of them into processing engine modules enters direction processing engine unit.
4. the implementation method of multicast mirror image flow equalization, feature exist in Ethernet switching chip according to claim 2
In the step S1 is specifically included：
Source is obtained according to the VLAN characteristic informations for the interchanger inbound port number or source messages for using the Ethernet switching chip
The corresponding image banner of message.
5. the implementation method of multicast mirror image flow equalization, feature exist in Ethernet switching chip according to claim 2
In, in the step S2, according to source messages it is corresponding forwarding destination by store forwarding module will enter direction handle engine list
Source messages in member are selectively sent to out one of outgoing direction processing engine unit and specifically include：
The magnitude range of the inbound port number carried according to source messages is selectively sent to its corresponding outgoing direction processing engine mould
One of outgoing direction processing engine unit in the block.
6. the implementation method of multicast mirror image flow equalization, feature exist in Ethernet switching chip according to claim 2
In, in the step S2, to the source messages carry out replicate forming loopback message, and pass through the loopback unit and loop back to wherein
One enters direction processing engine unit and specifically includes：
The loading condition that loopback message is entered to processing engine unit in direction according to selectively loops back to one of them side of entering
To processing engine unit.
7. a kind of realization system of multicast mirror image flow equalization in Ethernet switching chip, which is characterized in that
The Ethernet switching chip includes：Enter direction processing engine modules, outgoing direction processing engine modules and storage forwarding
Module, it is described enter direction processing engine modules be connected with storage forwarding module, storage forwarding module and the outgoing direction processing
Engine modules are connected；
Wherein, it is described enter direction processing engine modules include at least two enter direction handle engine unit；The outgoing direction processing
The engine modules outgoing direction identical with entering direction processing engine unit quantity handles engine unit；The storage forwarding module packet
It includes：Loopback unit and message copied cells；
The system comprises：
Searching module is parsed, in source messages repeating process, the source messages being made to be selectively entered one of them side of entering
To processing engine unit, the source messages are parsed and be searched, its corresponding forwarding destination and image banner are obtained；
Forwarding module, for direction processing engine list will to be entered by storing forwarding module according to the corresponding forwarding destination of source messages
Source messages in member are selectively sent to out one of outgoing direction processing engine unit, meanwhile, the source messages are carried out
Duplication forms loopback message, and pass through the loopback unit loop back to one of them enter direction processing engine unit；
Multicast processing module, for according to loopback message by it is described enter direction handle engine unit obtain its corresponding multicast mirror
The ID of the purpose multicast group of picture, and message copied cells are sent to, so that the purpose that the message copied cells are received according to it
The ID of multicast group and the port information of purpose multicast group carry out the multicast replication of multi-destination, form multiple message mirrors, and
Selectively it is sent to one of outgoing direction processing engine unit processing；
Output module handles engine unit to each message mirror according to its corresponding port information for passing through the outgoing direction
It is forwarded.
8. the realization system of multicast mirror image flow equalization, feature exist in Ethernet switching chip according to claim 7
In,
The parsing searching module is specifically used for：In source messages repeating process, according to the big of the inbound port number of source messages carrying
Small range selectively enter its it is corresponding enter direction processing engine modules in one of them enter direction processing engine unit.
9. the realization system of multicast mirror image flow equalization, feature exist in Ethernet switching chip according to claim 7
In,
The parsing searching module is specifically used for：
Source is obtained according to the VLAN characteristic informations for the interchanger inbound port number or source messages for using the Ethernet switching chip
The corresponding image banner of message.
10. the realization system of multicast mirror image flow equalization, feature exist in Ethernet switching chip according to claim 7
In,
The forwarding module is specifically used for：It is right that the magnitude range of the inbound port number carried according to source messages is selectively sent to its
One of outgoing direction processing engine unit in the outgoing direction processing engine modules answered.
11. the realization system of multicast mirror image flow equalization, feature exist in Ethernet switching chip according to claim 7
In,
The forwarding module is specifically used for：Loopback message is entered to the loading condition selectivity of direction processing engine unit according to
Ground loops back to one of them and enters direction processing engine unit.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711408283.3A CN108134747A (en) | 2017-12-22 | 2017-12-22 | The realization method and system of Ethernet switching chip, its multicast mirror image flow equalization |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711408283.3A CN108134747A (en) | 2017-12-22 | 2017-12-22 | The realization method and system of Ethernet switching chip, its multicast mirror image flow equalization |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108134747A true CN108134747A (en) | 2018-06-08 |
Family
ID=62392309
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711408283.3A Withdrawn CN108134747A (en) | 2017-12-22 | 2017-12-22 | The realization method and system of Ethernet switching chip, its multicast mirror image flow equalization |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108134747A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109120554A (en) * | 2018-09-25 | 2019-01-01 | 杭州迪普科技股份有限公司 | A kind of flow mirror method and switching equipment based on true mirror image |
| CN109194777A (en) * | 2018-09-10 | 2019-01-11 | 盛科网络（苏州）有限公司 | The method and device of MAC address aging in stacking system |
| CN111092829A (en) * | 2019-12-09 | 2020-05-01 | 昆高新芯微电子(江苏)有限公司 | Multi-core switching chip based on switching architecture and data transmission method thereof |
| CN114205682A (en) * | 2021-12-18 | 2022-03-18 | 网络通信与安全紫金山实验室 | Programmable switch and its realization method |
| CN115086253A (en) * | 2022-06-16 | 2022-09-20 | 苏州盛科通信股份有限公司 | Ethernet switching chip and high-bandwidth message forwarding method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101394362A (en) * | 2008-11-12 | 2009-03-25 | 清华大学 | Method of Load Balancing Multi-core Network Processors Based on Flow Fragmentation |
| CN102377646A (en) * | 2010-08-12 | 2012-03-14 | 盛科网络（苏州）有限公司 | Forwarding chip, network switching system and multicast implementation method |
| US20150381710A1 (en) * | 2014-06-30 | 2015-12-31 | Fortinet, Inc. | Socket application program interface (api) for efficient data transactions |
| CN106603440A (en) * | 2016-12-30 | 2017-04-26 | 盛科网络（苏州）有限公司 | Switch unit and method for realizing multi-destination message mirroring |
-
2017
- 2017-12-22 CN CN201711408283.3A patent/CN108134747A/en not_active Withdrawn
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101394362A (en) * | 2008-11-12 | 2009-03-25 | 清华大学 | Method of Load Balancing Multi-core Network Processors Based on Flow Fragmentation |
| CN102377646A (en) * | 2010-08-12 | 2012-03-14 | 盛科网络（苏州）有限公司 | Forwarding chip, network switching system and multicast implementation method |
| US20150381710A1 (en) * | 2014-06-30 | 2015-12-31 | Fortinet, Inc. | Socket application program interface (api) for efficient data transactions |
| CN106603440A (en) * | 2016-12-30 | 2017-04-26 | 盛科网络（苏州）有限公司 | Switch unit and method for realizing multi-destination message mirroring |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109194777A (en) * | 2018-09-10 | 2019-01-11 | 盛科网络（苏州）有限公司 | The method and device of MAC address aging in stacking system |
| CN109194777B (en) * | 2018-09-10 | 2021-08-31 | 苏州盛科通信股份有限公司 | Method and device for aging MAC address in stacking system |
| CN109120554A (en) * | 2018-09-25 | 2019-01-01 | 杭州迪普科技股份有限公司 | A kind of flow mirror method and switching equipment based on true mirror image |
| CN111092829A (en) * | 2019-12-09 | 2020-05-01 | 昆高新芯微电子(江苏)有限公司 | Multi-core switching chip based on switching architecture and data transmission method thereof |
| CN114205682A (en) * | 2021-12-18 | 2022-03-18 | 网络通信与安全紫金山实验室 | Programmable switch and its realization method |
| CN114205682B (en) * | 2021-12-18 | 2024-03-12 | 网络通信与安全紫金山实验室 | Programmable exchanger and implementation method thereof |
| CN115086253A (en) * | 2022-06-16 | 2022-09-20 | 苏州盛科通信股份有限公司 | Ethernet switching chip and high-bandwidth message forwarding method |
| CN115086253B (en) * | 2022-06-16 | 2024-03-29 | 苏州盛科通信股份有限公司 | Ethernet exchange chip and high-bandwidth message forwarding method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108134747A (en) | The realization method and system of Ethernet switching chip, its multicast mirror image flow equalization | |
| Tobagi | Fast packet switch architectures for broadband integrated services digital networks | |
| US5274642A (en) | Output buffered packet switch with a flexible buffer management scheme | |
| EP1388235B1 (en) | Apparatus and methods for efficient multicassting of data packets | |
| US8873554B2 (en) | Data structure-less distributed fabric multicast | |
| JP3846880B2 (en) | System and method for controlling multicast traffic in a data packet switch | |
| CN106603440A (en) | Switch unit and method for realizing multi-destination message mirroring | |
| US8902887B2 (en) | Load-balancing structure for packet switches and its constructing method | |
| CN106603441B (en) | Multicast message processing method and exchange chip in distributed aggregated link network | |
| US5768270A (en) | ATM switch using synchronous switching by groups of lines | |
| JPS60500936A (en) | broadband digital switching network | |
| CN107786456B (en) | Flow control methods and system, packet switching equipment and user equipment | |
| JPH0637800A (en) | Exchange provided with deblocking self- route designation exchange network | |
| US11228524B1 (en) | Methods and apparatus for efficient use of link aggregation groups | |
| CN101262436B (en) | Multicast duplication method, device and line card | |
| WO1998020647A1 (en) | Method and apparatus to translate data streams among multiple parties | |
| US8737412B2 (en) | Network relay system and method of automatically setting a network relay system | |
| CN101986601B (en) | Multicast data transmission method and equipment | |
| US20130063441A1 (en) | Measuring and Displaying Bandwidth Contention | |
| CN109379764A (en) | File transmitting method and device | |
| WO1993006675A1 (en) | Nonblocking multicast fast packet/circuit switching networks | |
| CN109218220A (en) | A load balancing method for hybrid switching of single multicast services | |
| CN113472656A (en) | Two-layer switching equipment and method for forwarding multicast data message | |
| US8594130B2 (en) | Method and system for shortening the length of time gaps between data units in a data switching network | |
| Newman | Fast packet switching for integrated services |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20180608 |