CN106341344A - Flow classification method and device of multi-channel process - Google Patents
Flow classification method and device of multi-channel process Download PDFInfo
- Publication number
- CN106341344A CN106341344A CN201610837575.8A CN201610837575A CN106341344A CN 106341344 A CN106341344 A CN 106341344A CN 201610837575 A CN201610837575 A CN 201610837575A CN 106341344 A CN106341344 A CN 106341344A
- Authority
- CN
- China
- Prior art keywords
- message
- list item
- incidence relation
- cube computation
- control
- 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.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
- H04L67/1004—Server selection for load balancing
- H04L67/1014—Server selection for load balancing based on the content of a request
-
- 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
- H04L47/24—Traffic characterised by specific attributes, e.g. priority or QoS
- H04L47/2441—Traffic characterised by specific attributes, e.g. priority or QoS relying on flow classification, e.g. using integrated services [IntServ]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/14—Session management
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Computer And Data Communications (AREA)
Abstract
The invention provides a flow classification method and device of a multi-channel process applied to multi-core network equipment. The multi-core network equipment establishes a session with opposite end equipment based on a multi-channel protocol. The session comprises a control connection and at least one data connection. The method comprises a step of extracting a specified message feature from the data message of a data connection bearing, a step of inquiring a preset global inquiry table, and determining a target CPU corresponding to the specified message feature, wherein the global inquiry table comprises a mapping relation between the specified message feature corresponding to the data connection and a CPU which processes the control message of the control connection extracted from the control message which controls the connection bearing, a step of sending the data message to the target CPU such that the target CPU processes the data message. According to the embodiment of the invention, the problem of flow disorder caused by sending the flow of a same session to multiple CPUs for processing in processing the flow of a multi-channel protocol is solved.
Description
Technical field
The application is related to communication technical field, particularly to a kind of flow point class method and apparatus of multichannel flow process.
Background technology
In multi-core network environment, in order to ensure that flow forwards in an orderly manner, generally same flow process is assigned on same cpu
Forward, to eliminate same flow when multiple cpu forward, lead to the out of order problem of flow because certain cpu is busy.Common
Agreement is to produce a flow, such as ping (packet internet groper, the Internet packets survey meter), telnet agreement,
Http (hypertext transfer protocol, HTML (Hypertext Markup Language)), can record cpu in session, and passes through this
Follow-up data message is delivered to same cpu process by record, reaches the purpose by assignment of traffic cpu.But work as and enable ftp
(file transfer protocol, file transfer protocol (FTP)), rstp (real time streaming protocol, in real time
Flow host-host protocol) etc. multichannel agreement when, the flow of same session can not be delivered to same cpu process, Ke Nengzao by such scheme
Become the out of order problem of flow.
Content of the invention
In view of this, the application provides a kind of flow point class method and apparatus of multichannel flow process, many in process in order to solve
During the flow of channel protocol, the flow of same session is sent to multiple cpu process, the out of order problem of the flow being likely to result in.
Specifically, the application is achieved by the following technical solution:
A kind of stream sorting technique of multichannel flow process, is applied to multi-core network device, and described multi-core network device is based on many
Channel protocol and opposite equip. set up session;Described session includes one and controls connection and at least one data cube computation, comprising:
From the data message that described data cube computation carries, extract and specify message characteristic;
Inquire about default global query table, determine target cpu corresponding with described specified message characteristic;Wherein, described complete
Office's inquiry table includes connecting, from described control, the specified report corresponding to described data cube computation extracting the control message carrying
Civilian feature and the mapping relations processing the described cpu of control message controlling and connecting;
Described data message is delivered to described target cpu, so that described data message is processed by described target cpu.
In the stream sorting technique of described multichannel flow process, also include:
Connect the specified message characteristic extracting the control message carrying corresponding to described data cube computation from described control;
Corresponding incidence relation list item is generated based on described specified message characteristic, described incidence relation list item is included from described
Control and connect the specified message characteristic corresponding to described data cube computation extracting in the control message carrying and process described control
The mapping relations of the cpu of control message that system connects;
Described incidence relation list item is stored in described default global query table;And, by described incidence relation table
Item is saved in and connects corresponding memory space with described control.
In the stream sorting technique of described multichannel flow process, also include:
When determining target cpu corresponding with described specified message characteristic, association corresponding with described target cpu is closed
It is that list item is saved in memory space corresponding with described data cube computation, so that after receiving data message again, based on local
Described incidence relation list item described data message is sent to described target cpu, so that described data is processed by described target cpu
Message;
Delete the described described incidence relation list item controlling and connecting in corresponding memory space.
In the stream sorting technique of described multichannel flow process, also include:
When determining target cpu corresponding with described specified message characteristic, by the described pass in described global query table
Connection relation list item is labeled as use state;Wherein, the effect being labeled as use state is to prevent described incidence relation list item
Used by the other messages beyond the data message of described data cube computation.
In the stream sorting technique of described multichannel flow process, also include:
When described control connects aging, corresponding incidence relation list item will be connected from described global query table with this control
Middle deletion;
When described data cube computation is aging, will incidence relation list item corresponding with this data cube computation from this data cube computation pair
Delete in the memory space answered.
A kind of stream sorter of multichannel flow process, is applied to multi-core network device, and described multi-core network device is based on many
Channel protocol and opposite equip. set up session;Described session includes one and controls connection and at least one data cube computation, comprising:
Extraction unit, for, from the data message that described data cube computation carries, extracting and specifying message characteristic;
Query unit, for inquiring about default global query table, determines target corresponding with described specified message characteristic
cpu;Wherein, described global query table include from described control connect carry control message extract corresponding to described number
Mapping relations according to the specified message characteristic connecting and the cpu processing the described control message controlling and connecting;
Allocation unit, for described data message is delivered to described target cpu, to process described number by described target cpu
According to message.
In the stream sorter of described multichannel flow process, also include:
Described extraction unit, is further used for extracting corresponding to described from the described control message controlling and connecting carrying
The specified message characteristic of data cube computation;
Signal generating unit, for generating corresponding incidence relation list item, described incidence relation based on described specified message characteristic
List item includes connecting, from described control, the specified message spy corresponding to described data cube computation extracting the control message carrying
Seek peace and process the mapping relations of the described cpu of control message controlling and connecting;
Memory element, for storing described incidence relation list item in described default global query table;And, by institute
State incidence relation list item and be saved in and connect corresponding memory space with described control.
In the stream sorter of described multichannel flow process, also include:
Storage unit, for when determining target cpu corresponding with described specified message characteristic, will be with described target
Cpu corresponding incidence relation list item is saved in memory space corresponding with described data cube computation, so that receiving data again
After message, based on local described incidence relation list item, described data message is sent to described target cpu, with by described target
Cpu processes described data message;
Delete unit, for deleting the described described incidence relation list item controlling and connecting in corresponding memory space.
In the stream sorter of described multichannel flow process, also include:
Indexing unit, for when determining target cpu corresponding with described specified message characteristic, by described global query
Described incidence relation list item in table is labeled as use state;Wherein, be labeled as use state effect be prevent described
Incidence relation list item is used by the other messages beyond the data message of described data cube computation.
In the stream sorter of described multichannel flow process, also include:
Described deletion unit, is further used for, when described control connects aging, to connect corresponding association with this control
Relation list item is deleted from described global query table;When described data cube computation is aging, will association corresponding with this data cube computation
Relation list item is deleted from memory space corresponding with this data cube computation.
In the embodiment of the present invention, from the data message that described data cube computation carries, extract and specify message characteristic, according to institute
State specified message characteristic and inquire about default global query table, determine target cpu corresponding with described specified message characteristic, then general
Described data message delivers to described target cpu, to process described data message by described target cpu.
Due to, in default global query table, carrying including from the control message belonging to same session with described data cube computation
The specified message characteristic corresponding to described data cube computation taking out, and process reflecting of the described cpu of control message controlling and connecting
Penetrate relation, therefore described data message is to be sent to the cpu processing the control message belonging to same session with described data message,
Thus solving when processing the flow of multichannel agreement, the flow of same session is sent to multiple cpu process, is likely to result in
The out of order problem of flow.
Brief description
Fig. 1 is a kind of flow chart of the stream sorting technique of multichannel flow process shown in the application;
Fig. 2 is a kind of logic diagram of the stream sorter of the multichannel flow process shown in the application;
Fig. 3 is a kind of hardware structure diagram of the stream sorter of the multichannel flow process shown in the application.
Specific embodiment
In order that those skilled in the art more fully understand the technical scheme in the embodiment of the present invention, and make the present invention real
Apply the above-mentioned purpose of example, feature and advantage can become apparent from understandable, below in conjunction with the accompanying drawings to prior art and the present invention
Technical scheme in embodiment is described in further detail.
Referring to Fig. 1, it is the flow chart of the stream sorting technique of the application multichannel flow process, the executive agent of this embodiment is many
Nuclear network equipment, this multi-core network device is based on multichannel agreement and sets up session with opposite equip.;Wherein, above-mentioned session includes one
Individual control connects and at least one data cube computation;The method comprising the steps of:
Step 101: from the data message that described data cube computation carries, extract and specify message characteristic.
Step 102: inquire about default global query table, determine target cpu corresponding with described specified message characteristic;Its
In, described global query table include from described control connect carry control message extract corresponding to described data cube computation
Specified message characteristic and process described control connect control message cpu mapping relations.
Step 103: described data message is delivered to described target cpu, so that described datagram is processed by described target cpu
Literary composition.
Above-mentioned multi-core network device, refers to have the network equipment of multiple cpu, each cpu of the network equipment can be independent
Processing business, due to multi-core network device, to compare the monokaryon network equipment more superior in computing capability and performance, by increasingly
It is widely used.Multi-core network environment refers to the environment of multi-core network device inter-process business, in the situation of multiple cpu
Under, processing business can have differences with the situation of single cpu.
Above-mentioned multichannel agreement, refers to need to take the agreement of two or more ports in communication process.Multichannel
Agreement is usually used the fixed port of agreement and connects to set up one and control, then is dynamically selected port and is used for setting up data even
Connect;Control to connect and be responsible for processing the port that all kinds of requests and negotiation data connect, data cube computation is responsible for data transfer.For example, often
Multichannel agreement includes ftp agreement, rtsp agreement, sip (session initiation protocol, assist by conversation initial
View), etc..
Above-mentioned global query table, is built with opposite equip. based on multichannel agreement for storing all above-mentioned multi-core network devices
After vertical session, from the control report controlling the specified message characteristic corresponding to data cube computation extracting connection to be connected with processing controls
The mapping relations of the cpu of literary composition.Wherein, control and connect the access rights that data connection all has this global query's table.
In the related, in multi-core network environment, generally same flow process is fixed on same cpu and processes, with
Eliminate same stream when multiple cpu are processed, the out of order problem of flow that is leading to because some cpu is busy.For only one
The agreement of bar flow, can record cpu in session, by this record, follow-up message is all delivered at same cpu
Reason, reaches the purpose by stream distribution cpu.However, for the agreement having more than one flow, said method is no longer applicable.
In the present embodiment, in order to solve the above problems, can be corresponded to by extracting from the control message controlling connection
In the specified message characteristic of data cube computation, generate and store the control message that above-mentioned specified message characteristic is connected with processing controls
The mapping relations of cpu, after receiving the data message of data cube computation, extract the specified message characteristic of data message, Jing Guocha
Ask target cpu of the control message determining that processing controls connect, then data message is delivered to above-mentioned target cpu, reach with for a moment
The result that the flow of words is all processed by same cpu.
Below taking ftp agreement as a example, the embodiment of the present application is illustrated.Certainly, in actual applications, above-mentioned multichannel
Other types of multichannel agreement beyond agreement or ftp, will not enumerate in this example.
When multi-core network device sets up session based on ftp agreement and opposite equip., first pass through and control the control message connecting
Foundation connects, and connects, to control, source port, the destination interface processing all kinds of requests and negotiation data connection.
In the present embodiment, can parse and control the control message connecting, extract the specified message corresponding to data cube computation
Feature, wherein, above-mentioned specified message characteristic can determine that specifically a certain data connects.
In a kind of embodiment illustrating, ftp agreement can be obtained from the control message controlling connection of ftp agreement
Data cube computation destination interface, and combine and control the source ip connecting, purpose ip and agreement, draw the four-tuple of data cube computation
(source ip, purpose ip, agreement, destination interface).
In the present embodiment, after obtaining the specified message characteristic of data cube computation of ftp agreement, above-mentioned specified report can be based on
Civilian feature generates corresponding incidence relation list item, and it is above-mentioned with process that this incidence relation list item can include above-mentioned specified message characteristic
Control the mapping relations of the cpu of control message connecting;This mapping relations can make subsequently to be obtained according to above-mentioned specified message characteristic
Take the cpu processing the control message that above-mentioned control connects.
After this incidence relation list item generates, this incidence relation list item can be stored to above-mentioned global query table;
Can be control connection one section of memory space of distribution in advance of ftp agreement, create an incidence relation table in memory space, right
In this incidence relation table, the only corresponding above-mentioned control connection of this incidence relation table has access claim, data cube computation or other
Control to connect and do not access claim;After above-mentioned incidence relation list item generates, can also by this incidence relation list item store with
Control in the above-mentioned incidence relation table connecting in corresponding memory space.
In a kind of embodiment illustrating, because the incidence relation list item of above-mentioned global query table storage may quantity crowd
Many, for improving follow-up inquiry velocity, hash calculating can be carried out for above-mentioned specified message characteristic, obtain hash value, will
To hash value be connected with processing controls control message cpu mapping relations stored in the form of incidence relation list item complete
In office's inquiry table.Wherein, the form of above-mentioned global query table can be chained list.
In the present embodiment, after the data message of the data cube computation receiving ftp agreement, extract from this data message and refer to
Determine message characteristic, according to the specified message characteristic extracting, inquire about above-mentioned global query table, determine and process and this specified message spy
Levy corresponding target cpu.For example: from the data message of the data cube computation of ftp agreement, extract the four-tuple of data cube computation, pin
This four-tuple is carried out with hash calculating, obtains hash value, above-mentioned global query table is inquired about according to this hash value, inquiry obtains and is somebody's turn to do
Hash value corresponding incidence relation list item, determines target cpu from this incidence relation list item.
In the present embodiment, one section of memory space can be distributed in advance, in memory space for the data cube computation of ftp agreement
Create an incidence relation table, for this incidence relation table, the only corresponding above-mentioned data cube computation of this incidence relation table has visit
Ask claim, control connection or other data cube computation not to access claim.When above-mentioned data cube computation is according to from above-mentioned global query table
In the incidence relation list item that inquires, when determining target cpu corresponding with above-mentioned specified message characteristic, can be by above-mentioned association
Relation list item is saved in the above-mentioned incidence relation table in memory space corresponding with data cube computation, so when subsequently receiving again
To data cube computation data message when, can obtain and above-mentioned specified message characteristic directly according to local incidence relation list item
Corresponding target cpu, it is no longer necessary to inquiry global query table, improves operating rate.
Meanwhile, after above-mentioned data cube computation saves above-mentioned incidence relation list item, above-mentioned control can be deleted and connect correspondence
Memory space in above-mentioned incidence relation list item.
In addition, when above-mentioned data cube computation determines target cpu corresponding with above-mentioned specified message characteristic, can also will be upper
The above-mentioned incidence relation list item stated in global query's table is labeled as use state, is marked as the incidence relation of use state
List item, if found it is impossible to be used by other data cube computation;Because each incidence relation list item can only be by a data
Connect use, this measure be prevent from having the message characteristic that identical is specified data cube computation use this incidence relation list item, thus
Avoid there is the exception message of identical specified message characteristic, attack message causes shadow using this incidence relation list item to session
Ring.
In the present embodiment, after determining and processing target cpu corresponding with above-mentioned specified message characteristic, by above-mentioned data
The data message connecting delivers to this target cpu.After this target cpu receives data message, this data message is processed.
In the present embodiment, controlling to connect the control message carrying and extract corresponding to belonging to same from ftp agreement
The specified message characteristic of the data cube computation of session, after generating incidence relation list item, the data message of the data cube computation of ftp agreement can
Can be because some reasons (for example: opposite equip. breaks down) fail to reach the multi-core network device of local terminal.In this case,
Incidence relation list item with control connect in memory space corresponding with data cube computation and above-mentioned complete in corresponding memory space
State in office's inquiry table is different from the situation after the data message receiving data cube computation.For incidence relation list item
Process, can first judge after incidence relation list item creates, if had and specify message with this incidence relation list item identical
The data cube computation of feature uses.
Due to after inquiring incidence relation list item corresponding with data cube computation, can by this incidence relation list item be saved in
In the corresponding memory space of this data cube computation, and delete and control this incidence relation list item connecting in corresponding memory space,
This incidence relation list item in global query's table is labeled as use state simultaneously.
So, this incidence relation list item can be whether there is according to controlling to connect in corresponding memory space, judging should
Whether incidence relation list item is used.
In a kind of embodiment illustrating, can periodically check and be connected with controlling according to the time pre-setting
Whether there is certain incidence relation list item in corresponding memory space;For example, for certain incidence relation list item specific, preset
One overtime timer of value, the time-out time of this overtime timer can be 10 minutes, reaches the super of this overtime timer every time
When the time, can check in ftp agreement with control the incidence relation table connecting in corresponding memory space, judge to create before this
Above-mentioned incidence relation list item whether there is:
On the one hand, if above-mentioned incidence relation list item exist, illustrate above-mentioned incidence relation list item establishment after all the time not
Used by corresponding data cube computation.In this case, when above-mentioned control connects aging, close in the above-mentioned association deleting local
It is the above-mentioned incidence relation list item that can delete while list item in above-mentioned global query chained list, to prevent invalid incidence relation
List item committed memory.
On the other hand, if do not existed in above-mentioned incidence relation list item, illustrate above-mentioned incidence relation list item after creating
Used by corresponding data cube computation.In this case, when above-mentioned data cube computation is aging, can delete with above-mentioned data even
Connect the above-mentioned incidence relation list item in corresponding memory space, to prevent invalid incidence relation list item committed memory;Above-mentioned complete
The above-mentioned incidence relation list item of office's inquiry linked list has been labeled as use state, is retained before above-mentioned control connection is aging,
Prevent above-mentioned control from connecting the data cube computation set up again and use above-mentioned incidence relation list item, session is impacted.
In sum, the embodiment of the present invention, from the data message that described data cube computation carries, is extracted and is specified message characteristic,
Default global query table is inquired about according to described specified message characteristic, determines target cpu corresponding with described specified message characteristic,
Then described data message is delivered to described target cpu, so that described data message is processed by described target cpu.Due to default
In global query's table, including from described data cube computation belong to same session control message in extract corresponding to described number
According to the mapping relations of the specified message characteristic connecting and the cpu processing the described control message controlling and connecting, therefore described data
Message is to be sent to the cpu processing the control message belonging to same session with described data message.By these measures so that
Under multi-core network environment, every flow of multichannel agreement can be in same cpu process, it is to avoid the flow of same session exists
When multiple cpu are processed, it is possible to create the out of order problem of flow.
Corresponding with the embodiment of the stream sorting technique of multichannel flow process of the present invention, present invention also offers for execution
State the embodiment of the device of embodiment of the method.
Referring to Fig. 2, it is an embodiment block diagram of the stream sorter of multichannel flow process of the present invention:
As described in Figure 2, the stream sorter 20 of this multichannel flow process includes:
Extraction unit 210, for, from the data message that described data cube computation carries, extracting and specifying message characteristic.
Query unit 220, for inquiring about default global query table, determines target corresponding with described specified message characteristic
cpu;Wherein, described global query table include from described control connect carry control message extract corresponding to described number
Mapping relations according to the specified message characteristic connecting and the cpu processing the described control message controlling and connecting.
Allocation unit 230, for described data message is delivered to described target cpu, described to be processed by described target cpu
Data message.
In this example, described device also includes:
Described extraction unit 210, be further used for from described control connect carry control message extract corresponding to
The specified message characteristic of described data cube computation.
Signal generating unit 240, for generating corresponding incidence relation list item based on described specified message characteristic, described association is closed
It is that list item includes connecting, from described control, the specified message corresponding to described data cube computation extracting the control message carrying
Feature and the mapping relations processing the described cpu of control message controlling and connecting.
Memory element 250, for storing described incidence relation list item in described default global query table;And,
Described incidence relation list item is saved in and connects corresponding memory space with described control.
In this example, described device also includes:
Storage unit 260, for when determining target cpu corresponding with described specified message characteristic, will be with described mesh
Mark cpu corresponding incidence relation list item is saved in memory space corresponding with described data cube computation, so that receiving number again
After message, based on local described incidence relation list item, described data message is sent to described target cpu, with by described mesh
Mark cpu processes described data message.
Delete unit 270, for deleting the described described incidence relation list item controlling and connecting in corresponding memory space.
In this example, described device also includes:
Indexing unit 280, for when determining target cpu corresponding with described specified message characteristic, by the described overall situation
Described incidence relation list item in inquiry table is labeled as use state;Wherein, the effect being labeled as use state is to prevent
Described incidence relation list item is used by the other messages beyond the data message of described data cube computation.
In this example, described deletion unit 270, is further used for:
When described control connects aging, corresponding incidence relation list item will be connected from described global query table with this control
Middle deletion;
When described data cube computation is aging, will incidence relation list item corresponding with this data cube computation from this data cube computation pair
Delete in the memory space answered.
The embodiment of the stream sorter of the application multichannel flow process can be applied on multi-core network device.Device is implemented
Example can be realized by software it is also possible to realize by by way of hardware or software and hardware combining.As a example implemented in software, as
Device on one logical meaning, is will be corresponding in nonvolatile memory by the processor of its place multi-core network device
Computer program instructions read and run formation in internal memory.For hardware view, as shown in figure 3, being the application multichannel
A kind of hardware structure diagram of the stream sorter place multi-core network device of flow process, except the processor shown in Fig. 3, internal memory, net
Outside network interface and nonvolatile memory, the multi-core network device that in embodiment, device is located is generally according to this multichannel
The actual functional capability of the flow point class of flow process, can also include other hardware, this is repeated no more.
In said apparatus, the process of realizing of the function of unit and effect specifically refers to corresponding step in said method
Realize process, will not be described here.
For device embodiment, because it corresponds essentially to embodiment of the method, thus real referring to method in place of correlation
The part applying example illustrates.Device embodiment described above is only schematically, wherein said as separating component
The unit illustrating can be or may not be physically separate, as the part that unit shows can be or can also
It is not physical location, you can with positioned at a place, or can also be distributed on multiple NEs.Can be according to actual
Need to select the purpose to realize application scheme for some or all of module therein.Those of ordinary skill in the art are not paying
In the case of going out creative work, you can to understand and to implement.
The foregoing is only the preferred embodiment of the application, not in order to limit the application, all essences in the application
Within god and principle, any modification, equivalent substitution and improvement done etc., should be included within the scope of the application protection.
Claims (10)
1. a kind of stream sorting technique of multichannel flow process, is applied to multi-core network device, and described multi-core network device is based on and leads to more
Road agreement and opposite equip. set up session;Described session includes one and controls connection and at least one data cube computation, and its feature exists
In, comprising:
From the data message that described data cube computation carries, extract and specify message characteristic;
Inquire about default global query table, determine target cpu corresponding with described specified message characteristic;Wherein, the described overall situation is looked into
Inquiry table includes connecting, from described control, the specified message spy corresponding to described data cube computation extracting the control message carrying
Seek peace and process the mapping relations of the described cpu of control message controlling and connecting;
Described data message is delivered to described target cpu, so that described data message is processed by described target cpu.
2. method according to claim 1 is it is characterised in that methods described also includes:
Connect the specified message characteristic extracting the control message carrying corresponding to described data cube computation from described control;
Corresponding incidence relation list item is generated based on described specified message characteristic, described incidence relation list item is included from described control
Connect the specified message characteristic corresponding to described data cube computation extracting in the control message carrying and process described control even
The mapping relations of the cpu of control message connecing;
Described incidence relation list item is stored in described default global query table;And, described incidence relation list item is protected
It is stored to connect corresponding memory space with described control.
3. method according to claim 2 is it is characterised in that methods described also includes:
When determining target cpu corresponding with described specified message characteristic, will incidence relation table corresponding with described target cpu
Item is saved in memory space corresponding with described data cube computation, so that after receiving data message again, based on local institute
State incidence relation list item to send described data message to described target cpu, to process described datagram by described target cpu
Literary composition;Delete the described described incidence relation list item controlling and connecting in corresponding memory space.
4. method according to claim 3 is it is characterised in that methods described also includes:
When determining target cpu corresponding with described specified message characteristic, the described association in described global query table is closed
It is that list item is labeled as use state;Wherein, the effect being labeled as use state is to prevent described incidence relation list item by institute
State the other messages beyond the data message of data cube computation to use.
5. method according to claim 4 is it is characterised in that methods described also includes:
When described control connects aging, corresponding incidence relation list item will be connected with this control and delete from described global query table
Remove;When described data cube computation is aging, will incidence relation list item corresponding with this data cube computation from corresponding with this data cube computation
Delete in memory space.
6. a kind of stream sorter of multichannel flow process, is applied to multi-core network device, and described multi-core network device is based on and leads to more
Road agreement and opposite equip. set up session;Described session includes one and controls connection and at least one data cube computation, and its feature exists
In, comprising:
Extraction unit, for, from the data message that described data cube computation carries, extracting and specifying message characteristic;
Query unit, for inquiring about default global query table, determines target cpu corresponding with described specified message characteristic;Its
In, described global query table include from described control connect carry control message extract corresponding to described data cube computation
Specified message characteristic and process described control connect control message cpu mapping relations;
Allocation unit, for described data message is delivered to described target cpu, to process described datagram by described target cpu
Literary composition.
7. device according to claim 6 is it is characterised in that described device also includes:
Described extraction unit, is further used for extracting corresponding to described data from the described control message controlling and connecting carrying
The specified message characteristic connecting;
Signal generating unit, for generating corresponding incidence relation list item, described incidence relation list item based on described specified message characteristic
Including from described control connect carry control in message the specified message characteristic corresponding to described data cube computation that extracts and
Process the mapping relations of the described cpu of control message controlling and connecting;
Memory element, for storing described incidence relation list item in described default global query table;And, by described pass
Connection relation list item is saved in and connects corresponding memory space with described control.
8. device according to claim 7 is it is characterised in that described device also includes:
Storage unit, for when determining target cpu corresponding with described specified message characteristic, will be with described target cpu pair
The incidence relation list item answered is saved in memory space corresponding with described data cube computation, so that receiving data message again
Afterwards, based on local described incidence relation list item, described data message is sent to described target cpu, with by described target cpu
Process described data message;
Delete unit, for deleting the described described incidence relation list item controlling and connecting in corresponding memory space.
9. device according to claim 8 is it is characterised in that described device also includes:
Indexing unit, for when determining target cpu corresponding with described specified message characteristic, by described global query table
Described incidence relation list item be labeled as use state;Wherein, the effect being labeled as use state is to prevent described association
Relation list item is used by the other messages beyond the data message of described data cube computation.
10. device according to claim 8 is it is characterised in that described device also includes:
Described deletion unit, is further used for, when described control connects aging, to connect corresponding incidence relation with this control
List item is deleted from described global query table;When described data cube computation is aging, will incidence relation corresponding with this data cube computation
List item is deleted from memory space corresponding with this data cube computation.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610837575.8A CN106341344B (en) | 2016-09-21 | 2016-09-21 | A kind of flow point class method and apparatus of multichannel process |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610837575.8A CN106341344B (en) | 2016-09-21 | 2016-09-21 | A kind of flow point class method and apparatus of multichannel process |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106341344A true CN106341344A (en) | 2017-01-18 |
CN106341344B CN106341344B (en) | 2019-10-11 |
Family
ID=57839092
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610837575.8A Active CN106341344B (en) | 2016-09-21 | 2016-09-21 | A kind of flow point class method and apparatus of multichannel process |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106341344B (en) |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2001022666A1 (en) * | 1999-09-20 | 2001-03-29 | Telefonaktiebolaget L M Ericsson (Publ) | Ip flow classifier for distinguishing real-time flows from non-real-time flows |
CN1787479A (en) * | 2004-12-09 | 2006-06-14 | 上海贝尔阿尔卡特股份有限公司 | Method for processing of double layer channel protocol message and apparatus thereof |
CN101426004A (en) * | 2007-10-29 | 2009-05-06 | 华为技术有限公司 | Three layer conversation access method, system and equipment |
CN101764751A (en) * | 2009-12-25 | 2010-06-30 | 杭州华三通信技术有限公司 | Method, system and equipment for forwarding roaming messages for wireless user terminal travelling across VLAN |
KR101145389B1 (en) * | 2010-09-28 | 2012-05-15 | 한국전자통신연구원 | Scalable centralized network architecture with de-centralization of network control and network switching apparatus therefor |
CN102932842A (en) * | 2012-09-14 | 2013-02-13 | 北京中创信测科技股份有限公司 | User conversation level shunting method applied to Gn interface |
CN103561130A (en) * | 2013-11-06 | 2014-02-05 | 北京神州绿盟信息安全科技股份有限公司 | Network address translation device and method suitable for multiple application layer protocols |
-
2016
- 2016-09-21 CN CN201610837575.8A patent/CN106341344B/en active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2001022666A1 (en) * | 1999-09-20 | 2001-03-29 | Telefonaktiebolaget L M Ericsson (Publ) | Ip flow classifier for distinguishing real-time flows from non-real-time flows |
CN1787479A (en) * | 2004-12-09 | 2006-06-14 | 上海贝尔阿尔卡特股份有限公司 | Method for processing of double layer channel protocol message and apparatus thereof |
CN101426004A (en) * | 2007-10-29 | 2009-05-06 | 华为技术有限公司 | Three layer conversation access method, system and equipment |
CN101764751A (en) * | 2009-12-25 | 2010-06-30 | 杭州华三通信技术有限公司 | Method, system and equipment for forwarding roaming messages for wireless user terminal travelling across VLAN |
CN101764751B (en) * | 2009-12-25 | 2013-06-05 | 杭州华三通信技术有限公司 | Method, system and equipment for forwarding roaming messages for wireless user terminal travelling across VLAN |
KR101145389B1 (en) * | 2010-09-28 | 2012-05-15 | 한국전자통신연구원 | Scalable centralized network architecture with de-centralization of network control and network switching apparatus therefor |
CN102932842A (en) * | 2012-09-14 | 2013-02-13 | 北京中创信测科技股份有限公司 | User conversation level shunting method applied to Gn interface |
CN103561130A (en) * | 2013-11-06 | 2014-02-05 | 北京神州绿盟信息安全科技股份有限公司 | Network address translation device and method suitable for multiple application layer protocols |
Also Published As
Publication number | Publication date |
---|---|
CN106341344B (en) | 2019-10-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10681009B2 (en) | Rule swapping in a packet network | |
US20210083857A1 (en) | Monitoring encrypted network traffic flows in a virtual environment using dynamic session key acquisition techniques | |
CN104580192B (en) | The treating method and apparatus of the network access request of application program | |
US20170295264A1 (en) | Network extended tcp splicing | |
US9794282B1 (en) | Server with queuing layer mechanism for changing treatment of client connections | |
US20180332081A1 (en) | Policy based on a requested behavior | |
US20160371178A1 (en) | Method and system for testing cloud based applications in a production environment using fabricated user data | |
CN108632110B (en) | Device performance testing method, system, computer device and storage medium | |
US9356844B2 (en) | Efficient application recognition in network traffic | |
WO2015149486A1 (en) | Page push method, device and server, and centralized network management controller | |
CN114024758B (en) | Flow characteristic extraction method, system, storage medium and electronic equipment | |
CN106341344A (en) | Flow classification method and device of multi-channel process | |
CN108011801A (en) | Method, unit and the system of data transfer | |
CN108063713B (en) | Network request processing method and device | |
CN106060155B (en) | The method and device of P2P resource-sharing | |
CN106664223A (en) | Detection method and detection device for the number of shared access hosts | |
KR101511474B1 (en) | Method for blocking internet access using agent program | |
CN114944996B (en) | Data acquisition method and device and computer readable medium | |
CN106230924A (en) | The compression method of message and device | |
CN117354047A (en) | Data packet control method, device, apparatus, storage medium and program product | |
CN115190077A (en) | Control method and device and computing equipment | |
CN116647598A (en) | Cross-network data exchange method, device, system, server and storage medium | |
CN115904691A (en) | Coordinating packet processing between kernel space and user space | |
CN116707984A (en) | Network access control method and device | |
KR20170097983A (en) | Network data processing apparatus and method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
CB02 | Change of applicant information | ||
CB02 | Change of applicant information |
Address after: Binjiang District and Hangzhou city in Zhejiang Province Road 310051 No. 68 in the 6 storey building Applicant after: Hangzhou Dipu Polytron Technologies Inc Address before: Binjiang District and Hangzhou city in Zhejiang Province Road 310051 No. 68 in the 6 storey building Applicant before: Hangzhou Dipu Technology Co., Ltd. |
|
GR01 | Patent grant | ||
GR01 | Patent grant |