CN109918017A - Data dispatching method and device - Google Patents
Data dispatching method and device Download PDFInfo
- Publication number
- CN109918017A CN109918017A CN201711317045.1A CN201711317045A CN109918017A CN 109918017 A CN109918017 A CN 109918017A CN 201711317045 A CN201711317045 A CN 201711317045A CN 109918017 A CN109918017 A CN 109918017A
- Authority
- CN
- China
- Prior art keywords
- buffer
- double buffer
- state
- data flow
- double
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The present invention relates to field of communication technology, a kind of data dispatching method and device are disclosed.Wherein, this method comprises: receiving video and/or image data stream;Received data stream is stored according to the state of buffer area each in the first Double buffer based on writing commands;The data flow stored based on reading order according to the state output of buffer area each in the first Double buffer;Pretreatment is formatted to the data flow of output;Based on writing commands according to the pretreated data flow of state storage formatting of buffer area each in the second Double buffer;The pretreated data flow of formatting stored based on reading order according to the state output of buffer area each in the second Double buffer;Data distribution is carried out to the pretreated data flow of the formatting of output.Thereby, it is possible to handle multiple data stream access, to realize the Real-Time Scheduling of big throughput data.
Description
Technical field
The present invention relates to field of communication technology more particularly to a kind of data dispatching methods and device.
Background technique
In data-link ground application system design field, its software of basic engineering clothes are generally calculated as with multitask at present
Business system.Multi job mode meets that type of service in data-link ground application system is more, task schedule frequency to a certain extent
Numerous characteristics of demand, but the multi-tasks Software service system routinely constructed dependent on operating system do not carry out it is special more
Core calculation optimization, it is generally relatively low to the processing core utilization rate of multi-core computer, therefore its traffic handing capacity and task schedule
Ability is restricted, and is unable to satisfy the demand of complex data chain ground application system.
On the other hand, with the development of datalink, data-link channel capacity also obtains very big expansion, data-link business
Demand is also enriched therewith and diversification.Currently, often except target position information, electronic reconnaissance information, battlefield situation information etc.
It advises outside business datum, battlefield live video stream, burst storage transmitting data flow, unmanned plane wide cut high-resolution reconnaissance image data
Deng gradually as the main service data of data link system.The type business is big with handling capacity, processing real-time requirement is high
Feature, by Conventional multi-tasldng calculate based on data-link ground application system be difficult meet demand.In addition, data-link ground is answered
Growing trend is also presented with the value volume and range of product of the data application terminal of system, considerably increase data-link ground application system into
The pressure that row business datum is distributed in real time proposes very big challenge to the design of data-link ground application system.
Therefore, it is necessary to it is a kind of can satisfy big handling capacity and have can Real-Time Scheduling characteristic data dispatching method and dress
It sets.
Summary of the invention
It is an object of the invention to overcome the shortage of prior art, a kind of data dispatching method and device are provided, can be solved
The certainly above-mentioned Real-Time Scheduling problem that cannot achieve big throughput data in the prior art.
Technical solution of the invention: a kind of data dispatching method, this method comprises: receiving video and/or picture number
According to stream;Received data stream is stored according to the state of buffer area each in the first Double buffer based on writing commands, is ordered based on reading
Enable the data flow stored according to the state output of buffer area each in the first Double buffer;Pre- place is formatted to the data flow of output
Reason;Based on writing commands according to the pretreated data flow of state storage formatting of buffer area each in the second Double buffer, it is based on
The pretreated data flow of formatting that reading order is stored according to the state output of buffer area each in the second Double buffer;To output
The pretreated data flow of formatting carry out data distribution.
It is preferably based on writing commands and received data stream packet is stored according to the state of buffer area each in the first Double buffer
Include: in first Double buffer state be idle state buffer area in store received data flow;Based on reading order root
Data flow according to the state output storage of each buffer area in the first Double buffer includes: that the state from first Double buffer is empty
The data flow of storage is exported in the buffer area of not busy state.
Preferably, this method further include: receive the first response signal, first response signal includes the number of instruction storage
The mark completed according to the output of stream;Two buffer zones in first Double buffer are controlled based on first response signal
Read-write state switching.
The state storage formatting that writing commands are preferably based on according to buffer area each in the second Double buffer is pretreated
Data flow includes: the pretreated number of storage formatting in the buffer area that the state in second Double buffer is idle state
According to stream;The pretreated data of formatting stored based on reading order according to the state output of buffer area each in the second Double buffer
Stream includes: pretreated for the formatting for exporting storage in the buffer area of idle state from the state in second Double buffer
Data flow.
Preferably, this method further include: receive the second response signal, second response signal includes the lattice of instruction storage
The mark that the output of the pretreated data flow of formulaization is completed;It is controlled in second Double buffer based on second response signal
Two buffer zones read-write state switching.
The present invention also provides a kind of data scheduling device, which includes: data flow access queue management module, is used for
Receive video and/or image data stream;First Double buffer scheduler module, for being based on writing commands according to each in the first Double buffer
The state of buffer area stores received data stream;The first Double buffer scheduler module is also used to based on reading order according to the
The data flow of the state output storage of each buffer area in one Double buffer;Preprocessing module is formatted, for the data flow to output
It is formatted pretreatment;Second Double buffer scheduler module, for being based on writing commands according to buffer area each in the second Double buffer
The pretreated data flow of state storage formatting;The second Double buffer scheduler module be also used to based on reading order according to
The pretreated data flow of formatting of the state output storage of each buffer area in second Double buffer;Data stream distribution queue management
Module, for carrying out data distribution to the pretreated data flow of the formatting of output.
Preferably, the first Double buffer scheduler module is based on writing commands according to the shape of buffer area each in the first Double buffer
State storage received data stream include: in first Double buffer state be idle state buffer area in store it is received
Data flow;The first Double buffer scheduler module is deposited based on reading order according to the state output of buffer area each in the first Double buffer
The data flow of storage includes: that the data flow of storage is exported from the buffer area that the state in first Double buffer is idle state.
Preferably, the first Double buffer scheduler module is also used to receive the first response signal, first response signal
The mark that the output of data flow including instruction storage is completed;The first Double buffer scheduler module is also used to based on described first
Response signal controls the read-write state switching of two buffer zones in first Double buffer.
Preferably, the second Double buffer scheduler module is based on writing commands according to the shape of buffer area each in the second Double buffer
The pretreated data flow of state storage formatting includes: in the buffer area that the state in second Double buffer is idle state
The pretreated data flow of storage formatting;The second Double buffer scheduler module is based on reading order according in the second Double buffer
The pretreated data flow of formatting of the state output storage of each buffer area includes: that the state from second Double buffer is
The pretreated data flow of formatting of storage is exported in the buffer area of idle state.
Preferably, this method further include: the second Double buffer scheduler module is also used to receive the second response signal, described
Second response signal includes the mark that the output of the pretreated data flow of formatting of instruction storage is completed;Described second pair slow
Deposit the read-write that scheduler module is also used to control two buffer zones in second Double buffer based on second response signal
State switching.
Through the above technical solutions, received number can be carried out in the first Double buffer based on writing commands and reading order
According to the storage and output of stream, the data flow of output can be formatted pretreatment, based on writing commands and can read life later
The storage and output that pretreated data flow is formatted in the second Double buffer are enabled, and can be pre- to the formatting of output
Data flow that treated carries out data distribution.Thereby, it is possible to handle multiple data stream access (for example, the reliable of high amount of traffic " gulps down
Enter "), to realize the Real-Time Scheduling of big throughput data.
Detailed description of the invention
Included attached drawing is used to provide to be further understood from the embodiment of the present invention, and which constitute one of specification
Point, for illustrating the embodiment of the present invention, and come together to illustrate the principle of the present invention with verbal description.It should be evident that below
Attached drawing in description is only some embodiments of the present invention, for those of ordinary skill in the art, is not paying creation
Property labour under the premise of, be also possible to obtain other drawings based on these drawings.
Fig. 1 is the flow chart of data dispatching method provided in an embodiment of the present invention;And
Fig. 2 is the block diagram of data scheduling device provided in an embodiment of the present invention.
Specific embodiment
Specific embodiments of the present invention are described in detail below in conjunction with attached drawing.In the following description, for solution
Purpose and not restrictive is released, elaborates detail, to help to be apparent from the present invention.However, to those skilled in the art
It is readily apparent that the present invention can also be practiced in the other embodiments departing from these details for member.
It should be noted that only showing in the accompanying drawings in order to avoid having obscured the present invention because of unnecessary details
Gone out with closely related device structure and/or processing step according to the solution of the present invention, and be omitted with relationship of the present invention not
Big other details.
Fig. 1 is the flow chart of data dispatching method provided in an embodiment of the present invention.
As shown in Figure 1, the data dispatching method that an embodiment of the present invention provides includes:
S100 receives video and/or image data stream;
S102 stores received data stream, base according to the state of buffer area each in the first Double buffer based on writing commands
In the data flow that reading order is stored according to the state output of buffer area each in the first Double buffer;
S104 is formatted pretreatment to the data flow of output;
S106, based on writing commands according to the pretreated number of state storage formatting of buffer area each in the second Double buffer
According to stream, the pretreated data of formatting stored based on reading order according to the state output of buffer area each in the second Double buffer
Stream;
S108 carries out data distribution to the pretreated data flow of the formatting of output.
Through the above technical solutions, received number can be carried out in the first Double buffer based on writing commands and reading order
According to the storage and output of stream, the data flow of output can be formatted pretreatment, based on writing commands and can read life later
The storage and output that pretreated data flow is formatted in the second Double buffer are enabled, and can be pre- to the formatting of output
Data flow that treated carries out data distribution.Thereby, it is possible to handle multiple data stream access (for example, the reliable of high amount of traffic " gulps down
Enter "), to realize the Real-Time Scheduling of big throughput data.
In other words, the present invention provides it is a kind of meet big handling capacity and have can Real-Time Scheduling characteristic collection data access,
Format the data dispatching method of pretreatment with distribution.This method can not only handle multiple data stream access, realize high amount of traffic
Reliable " swallowing ", guarantee the integrality of data flow, while being directed to diversification of the ground data chain application system to business datum
Application demand, this method provides the scheduling method with certain real-time, which can provide the real-time of business datum
Distribution performance realizes reliable " discharge " of high amount of traffic.
It, (can be due to using Double buffer in the method although showing each step in Fig. 1 in a sequential manner
It is read while executing deposit operation), so each step may be performed simultaneously.For example, while receiving data,
The formatting pretreatment that can carry out data flow and the distribution for formatting pretreated data.
In the present invention, above-mentioned method uses OpenMP multicore Computational frame, plays the calculating core of multi-core computer
While heart predominance, the general character of multi-core computer can be played by the OpenMP proprietary calculating core missions coordination ability
Advantage.
Also, when carrying out task cutting decomposition using OpenMP multicore computing engines, this method propose data flows to connect
Enqueue management formats three pretreatment, data stream distribution queue management key step parallel processings, passes through between each step
Double buffer carries out business bonding between any two, fully ensure that the independence and concurrency of task.
It is, can be realized the uninterrupted operation of three step real meanings, Limited computational resources are played to greatest extent
Computing capability, effectively promote big handling capacity multiple data stream access capability.Meanwhile " data flow connects using dual-cache mechanism progress
Enter-format pretreatment ", the task bonding of " format pretreatment-data stream distribution " is to form complete data throughput
Stream.
According to an embodiment of the present invention, institute is stored according to the state of buffer area each in the first Double buffer based on writing commands
Received data flow include: in first Double buffer state be idle state buffer area in store received data flow;
It according to the data flow that the state output of buffer area each in the first Double buffer stores include: slow from described first pair based on reading order
The state deposited be idle state buffer area in export the data flow of storage.
First Double buffer can have independent two panels buffer area, and a piece of buffer area in the first Double buffer can be used for depositing
Enter (write-in) operation, another buffer area in the first Double buffer can be used for taking out (readings) and operate, two panels buffer area it is each
From deposit and take out operation can run parallel.For example, a piece of buffer area is used to store received data flow wherein
While, the data flow (received data flow) of storage before another buffer area can be used for exporting.
According to an embodiment of the present invention, this method can also include:
The first response signal is received, first response signal includes the mark that the output of the data flow of instruction storage is completed
Know;
The read-write state switching of two buffer zones in first Double buffer is controlled based on first response signal.
Thus, it is possible to solve following problems: two buffer areas in the first Double buffer run (deposit operation and taking-up parallel
Operation carries out simultaneously) and in the case where having no the coordination system therebetween, it is competing that the two is directed to the same module resource not only having read but also write
It strives, to ensure that the safety and reliability of two buffer area data parallel distribution schedulings in the first Double buffer.
Wherein, the mark for including in the first response signal for example can complete mark for the first access.
According to an embodiment of the present invention, lattice are stored according to the state of buffer area each in the second Double buffer based on writing commands
The pretreated data flow of formulaization includes: the storage format in the buffer area that the state in second Double buffer is idle state
Change pretreated data flow;The formatting stored based on reading order according to the state output of buffer area each in the second Double buffer
Pretreated data flow includes: the lattice that storage is exported from the buffer area that the state in second Double buffer is idle state
The pretreated data flow of formulaization.
According to an embodiment of the present invention, this method can also include:
The second response signal is received, second response signal includes the pretreated data flow of formatting of instruction storage
Output complete mark;
The read-write state switching of two buffer zones in second Double buffer is controlled based on second response signal.
Thus, it is possible to solve following problems: two buffer areas in the second Double buffer run parallel and have no therebetween
In the case where the coordination system, the two is directed to the resource contention that same module had not only been read but also write, to ensure that in the second Double buffer
The safety and reliability of two buffer area data parallel distribution schedulings.
Wherein, the mark for including in the second response signal for example can complete mark for the second access.
In addition, in scheduling aspect, devise data flow access, format pretreatment, data stream distribution priority step by step
The scheduling mechanism (initiative that the height of priority determines the step of initiating Double buffer scheduling request) of promotion, to guarantee number
According to the real-time of distribution.
Fig. 2 is the block diagram of data scheduling device provided in an embodiment of the present invention.
As shown in Fig. 2, the data scheduling device that an embodiment of the present invention provides includes: the first Double buffer 30 and second pair
Caching 32;Data flow accesses queue management module 20, for receiving video and/or image data stream;First Double buffer dispatches mould
Block 22, for storing received data stream according to the state of each buffer area in the first Double buffer 30 based on writing commands;It is described
First Double buffer scheduler module 22 is also used to be deposited based on reading order according to the state output of each buffer area in the first Double buffer 30
The data flow of storage;Preprocessing module 24 is formatted, for being formatted pretreatment to the data flow of output;Second Double buffer tune
Module 26 is spent, for pretreated according to the state storage formatting of each buffer area in the second Double buffer 32 based on writing commands
Data flow;The second Double buffer scheduler module 26 is also used to based on reading order according to each buffer area in the second Double buffer 32
The pretreated data flow of formatting of state output storage;Data stream distribution queue management module 28, for the lattice to output
The pretreated data flow of formulaization carries out data distribution.
Through the above technical solutions, the first Double buffer scheduler module can be based on writing commands and reading order at first pair
The storage and output of the data flow access received data flow of queue management module are carried out in caching, the data flow of output can be by lattice
Formula preprocessing module is formatted pretreatment, and the second Double buffer scheduler module can be based on writing commands and reading life later
Enable the storage and output that pretreated data flow is formatted in the second Double buffer, data stream distribution queue management module
Data distribution can be carried out to the pretreated data flow of the formatting of output.Thereby, it is possible to handle multiple data stream access (example
Such as, reliable " swallowing " of high amount of traffic), to realize the Real-Time Scheduling of big throughput data.
Wherein, the data flow access received video of queue management module 20 and/or image data stream for example may include war
Field real time video data stream and/or unmanned plane wide cut high-resolution investigate image source data stream.Format preprocessing module 24
It such as may include video stream format device and/or Image formatter.Data stream distribution queue management module 28 for example can wrap
Include video distribution channel and/or image distribution channel.
For example, data flow access queue management module 20 is the container of all data flow accesses, and it is most to be responsible for management
It receives and according to stream is reliable by the buffer area in the first Double buffer scheduler module 22 the first Double buffer 30 of deposit;Format pre- place
Reason unit 24 is gone forward side by side row format by the data in the buffer area in the first Double buffer scheduler module 22 the first Double buffer 30 of taking-up
Pass through the buffer area in the second Double buffer scheduler module 26 the second Double buffer 32 of deposit after change processing;Data stream distribution queue management
The data that module 28 is taken out by the second Double buffer scheduler module 26 in the buffer area in second Double buffer 32 are distributed.
According to an embodiment of the present invention, it is slow according to first pair to be based on writing commands for the first Double buffer scheduler module 22
Deposit each buffer area in 30 state storage received data stream include: in first Double buffer 30 state be idle state
Buffer area in store received data flow;The first Double buffer scheduler module 22 is based on reading order according to the first Double buffer
It is idle state that the data flow of the state output storage of each buffer area, which includes: the state from first Double buffer 30, in 30
The data flow of storage is exported in buffer area.
According to an embodiment of the present invention, the first Double buffer scheduler module 22 is also used to receive the first response signal,
First response signal includes the mark that the output of the data flow of instruction storage is completed;The first Double buffer scheduler module 22
It is also used to control the read-write state switching of two buffer zones in first Double buffer 30 based on first response signal.
Thus, it is possible to guarantee the safety of two buffer area data parallel distribution schedulings in the first Double buffer and reliable
Property.
Wherein, the mark for including in the first response signal for example can complete mark for the first access.
For example, data flow access queue management module 20, formatting preprocessing module 24 are run parallel, and first pair slow
It deposits scheduler module 22 to be responsible for carrying out the two modules the access control of the buffer area in the first Double buffer 30, avoid because of parallel fortune
It goes and causes to guarantee that the accessing operation of two modules is complete to the competitive access with a piece of buffering to destroy data integrity
Isolation is coordinated to carry out.
According to an embodiment of the present invention, it is slow according to second pair to be based on writing commands for the second Double buffer scheduler module 26
The pretreated data flow of state storage formatting for depositing each buffer area in 32 includes: the state in second Double buffer 32
For the pretreated data flow of storage formatting in the buffer area of idle state;The second Double buffer scheduler module 26 is based on reading
Take order according in the second Double buffer 32 each buffer area state output storage the pretreated data flow of formatting include: from
State in second Double buffer 32 be idle state buffer area in export the pretreated data flow of formatting of storage.
According to an embodiment of the present invention, the second Double buffer scheduler module 26 is also used to receive the second response signal,
Second response signal includes the mark that the output of the pretreated data flow of formatting of instruction storage is completed;Described second
Double buffer scheduler module 26 is also used to control two buffer areas in second Double buffer 32 based on second response signal
The read-write state in domain switches.
Thus, it is possible to guarantee the safety of two buffer area data parallel distribution schedulings in the second Double buffer 32 and reliable
Property.
Wherein, the mark for including in the second response signal for example can complete mark for the second access.
For example, preprocessing module 24 and the same parallel operation of data stream distribution queue management module 28 are formatted, the
Two Double buffer scheduler modules 26 are responsible for carrying out the two modules the access control of the buffer area in the second Double buffer 32, to guarantee
Access visit isolation safe.
It is illustrated below with reference to data scheduling device and method of the example to the embodiment of the present invention.
Data stream distribution queue management module 28 has than formatting the higher priority of preprocessing module 24, data flow point
Hair queue management module 28 then can actively initiate the scheduling to the second Double buffer 32, and formatting preprocessing module 24 can only be passive
Respond the scheduling of the second Double buffer scheduler module 26.Therefore, when formatting preprocessing module 24 and double-buffer area being not used,
As long as the distribution that data stream distribution queue management module 28 completes the data of the second Double buffer 32 output can be initiated double slow immediately
Deposit scheduling switching cache object continue data distribution (for example, switching the second Double buffer two buffer areas read-write state,
The buffer area of output data is switched to write-in data, the buffer area for being stored into data is switched to read data), from
And guarantee the continuity of data distribution.
Formatting preprocessing module 24 has than the data flow access higher priority of queue management module 20, formats pre-
Processing module 24 can actively initiate the scheduling to the first Double buffer 30, and data flow accesses queue management module 20 can only be by the sound of something astir
Answer the scheduling of the first Double buffer scheduler module 22.Therefore, when data flow accesses queue management module 20 and Double buffer has been not used
Qu Shi can be sent out immediately as long as formatting the formatting pretreatment for the data that preprocessing module 24 completes the output of the first Double buffer 30
It plays Double buffer scheduling switching cache object and continues to format pretreatment (for example, two buffer areas of the first Double buffer of switching
Read-write state, the buffer area of output data is switched to write-in data, the buffer area for being stored into data is switched to
Read data), to guarantee the continuity of data prediction.
Therefore, methods and apparatus of the present invention by priority scheduling strategy avoid prime caching completely occupy without
The waiting time being scheduled ensure that scheduling real-time, to ensure that the real-time of distribution.
From above-described embodiment it can be seen that
Firstly, the present invention using special multicore computing technique by data flow access, format pretreatment, data stream distribution into
Row full parellel operating energy loss has higher computational efficiency under the conditions of same computing resource, to computing resource using more
Sufficiently, it is suitable for being disposed under the effective environment of computing resource.
Secondly, priority scheduling strategy proposed by the present invention emphasizes rear class function preferential principle, avoid between functional level
Cause to dispatch the shortcomings that waiting because caching occupancy is less than, improve the real-time of scheduling, ensure that the real-time of data distribution,
Occasion for transmitting live video stream has good application value.
Again, the invention proposes the deposit of the data of safety and taking-up access control, the race problem that two panels will be cached
It is reduced to the race problem to unitary variant (that is, mark is completed in first access and mark is completed in the second access), such scheduling
Mode ensure that the isolation to unitary variant access, to guarantee to realize the safety of scheduling to the harmony of cache access.
In short, the present invention realizes real-time access, formatting pretreatment and distribution in real time to big handling capacity multiple data stream,
Meet the software design demand that the deployment of data-link ground application system is carried out under computing resource limited conditions.Meanwhile the present invention
Above-mentioned data dispatching method and device has versatility, can be applied to the access of multipoint video stream and transmission, high code rate telemetering number
It is received and visualization, the access of Rich Media's tactical data and distribution according to real-time reception and processing, unmanned plane wide cut high-definition picture
Etc. technical fields.
As above it describes for a kind of embodiment and/or the feature that shows can be in a manner of same or similar at one or more
It is used in a number of other embodiments, and/or combines or substitute the feature in other embodiments with the feature in other embodiments
It uses.
It should be emphasized that term "comprises/comprising" refers to the presence of feature, one integral piece, step or component when using herein, but simultaneously
It is not excluded for the presence or additional of one or more other features, one integral piece, step, component or combinations thereof.
The device and method more than present invention can be by hardware realization, can also be by combination of hardware software realization.The present invention
It is related to such computer-readable program, when the program is performed by logical block, the logical block can be made to realize above
The device or component parts, or the logical block is made to realize various method or steps described above.The invention further relates to
For storing the storage medium of procedure above, such as hard disk, disk, CD, DVD, flash memory.
The many features and advantage of these embodiments are clear according to the detailed description, therefore appended claims are intended to
Cover all these feature and advantage of these embodiments fallen into its true spirit and range.Further, since this field
Technical staff is readily apparent that many modifications and changes, therefore is not meant to for the embodiment of the present invention to be limited to illustrated and description essence
Really structurally and operationally, but all suitable modifications and the equivalent fallen within the scope of its can be covered.
Unspecified part of the present invention is known to the skilled person technology.
Claims (10)
1. a kind of data dispatching method, which is characterized in that this method comprises:
Receive video and/or image data stream;
Received data stream is stored according to the state of buffer area each in the first Double buffer based on writing commands, is based on reading order
The data flow stored according to the state output of buffer area each in the first Double buffer;
Pretreatment is formatted to the data flow of output;
Based on writing commands according to the pretreated data flow of state storage formatting of buffer area each in the second Double buffer, it is based on
The pretreated data flow of formatting that reading order is stored according to the state output of buffer area each in the second Double buffer;
Data distribution is carried out to the pretreated data flow of the formatting of output.
2. the method according to claim 1, wherein
Storing received data stream according to the state of buffer area each in the first Double buffer based on writing commands includes: described the
In one Double buffer state be idle state buffer area in store received data flow;
It according to the data flow that the state output of buffer area each in the first Double buffer stores include: from described first based on reading order
State in Double buffer be idle state buffer area in export the data flow of storage.
3. according to the method described in claim 2, it is characterized in that, this method further include:
The first response signal is received, first response signal includes the mark that the output of the data flow of instruction storage is completed;
The read-write state switching of two buffer zones in first Double buffer is controlled based on first response signal.
4. the method according to claim 1, wherein
Include: according to the pretreated data flow of state storage formatting of buffer area each in the second Double buffer based on writing commands
The pretreated data flow of storage formatting in the buffer area that the state in second Double buffer is idle state;
The pretreated data of formatting stored based on reading order according to the state output of buffer area each in the second Double buffer
Stream includes: pretreated for the formatting for exporting storage in the buffer area of idle state from the state in second Double buffer
Data flow.
5. according to the method described in claim 4, it is characterized in that, this method further include:
The second response signal is received, second response signal includes the defeated of the pretreated data flow of formatting of instruction storage
The mark completed out;
The read-write state switching of two buffer zones in second Double buffer is controlled based on second response signal.
6. a kind of data scheduling device, which is characterized in that the device includes:
First Double buffer and the second Double buffer;
Data flow accesses queue management module, for receiving video and/or image data stream;
First Double buffer scheduler module, for being stored based on writing commands according to the state of each buffer area in first Double buffer
Received data stream;
The first Double buffer scheduler module is also used to the shape based on reading order according to each buffer area in first Double buffer
The data flow of state output storage;
Preprocessing module is formatted, for being formatted pretreatment to the data flow of output;
Second Double buffer scheduler module, for being stored based on writing commands according to the state of each buffer area in second Double buffer
Format pretreated data flow;
The second Double buffer scheduler module is also used to the shape based on reading order according to each buffer area in second Double buffer
The pretreated data flow of formatting of state output storage;
Data stream distribution queue management module, for carrying out data distribution to the pretreated data flow of the formatting of output.
7. device according to claim 6, which is characterized in that
The first Double buffer scheduler module is stored based on writing commands according to the state of each buffer area in first Double buffer
Received data stream include: in first Double buffer state be idle state buffer area in store received data
Stream;
The first Double buffer scheduler module is based on reading order according to the state output of each buffer area in first Double buffer
The data flow of storage includes: that the data of storage are exported from the buffer area that the state in first Double buffer is idle state
Stream.
8. device according to claim 7, which is characterized in that
The first Double buffer scheduler module is also used to receive the first response signal, and first response signal includes instruction storage
Data flow output complete mark;
The first Double buffer scheduler module is also used to control two in first Double buffer based on first response signal
The read-write state of a buffer zone switches.
9. device according to claim 6, which is characterized in that
The second Double buffer scheduler module is stored based on writing commands according to the state of each buffer area in second Double buffer
Formatting pretreated data flow includes: to store lattice in the buffer area that the state in second Double buffer is idle state
The pretreated data flow of formulaization;
The second Double buffer scheduler module is based on reading order according to the state output of each buffer area in second Double buffer
The pretreated data flow of formatting of storage includes: from the buffer area that the state in second Double buffer is idle state
Export the pretreated data flow of formatting of storage.
10. device according to claim 9, which is characterized in that
The second Double buffer scheduler module is also used to receive the second response signal, and second response signal includes instruction storage
The pretreated data flow of formatting the mark completed of output;
The second Double buffer scheduler module is also used to control two in second Double buffer based on second response signal
The read-write state of a buffer zone switches.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711317045.1A CN109918017A (en) | 2017-12-12 | 2017-12-12 | Data dispatching method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711317045.1A CN109918017A (en) | 2017-12-12 | 2017-12-12 | Data dispatching method and device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109918017A true CN109918017A (en) | 2019-06-21 |
Family
ID=66957222
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711317045.1A Pending CN109918017A (en) | 2017-12-12 | 2017-12-12 | Data dispatching method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109918017A (en) |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6085288A (en) * | 1997-04-14 | 2000-07-04 | International Business Machines Corporation | Dual cache directories with respective queue independently executing its content and allowing staggered write operations |
CN101917772A (en) * | 2010-08-26 | 2010-12-15 | 张若南 | Media access control method based on double-buffer area mixed type media access control protocol |
CN102426553A (en) * | 2011-11-11 | 2012-04-25 | 中国科学技术大学 | Method and device for transmitting data to user based on double-cache pre-reading |
US20140215156A1 (en) * | 2013-01-30 | 2014-07-31 | Electronics And Telecommunications Research Institute | Prioritized dual caching method and apparatus |
CN104750858A (en) * | 2015-04-16 | 2015-07-01 | 成都影泰科技有限公司 | Network-based data storage method |
CN104952032A (en) * | 2015-06-19 | 2015-09-30 | 清华大学 | Graph processing method and device as well as rasterization representation and storage method |
CN105530158A (en) * | 2015-12-16 | 2016-04-27 | 清华大学 | Inter-domain intercommunication control system of heterogeneous optical networks |
CN105912478A (en) * | 2016-04-06 | 2016-08-31 | 中国航空无线电电子研究所 | Dual-buffer mechanism based real-time system multi-task data sharing method |
-
2017
- 2017-12-12 CN CN201711317045.1A patent/CN109918017A/en active Pending
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6085288A (en) * | 1997-04-14 | 2000-07-04 | International Business Machines Corporation | Dual cache directories with respective queue independently executing its content and allowing staggered write operations |
CN101917772A (en) * | 2010-08-26 | 2010-12-15 | 张若南 | Media access control method based on double-buffer area mixed type media access control protocol |
CN102426553A (en) * | 2011-11-11 | 2012-04-25 | 中国科学技术大学 | Method and device for transmitting data to user based on double-cache pre-reading |
US20140215156A1 (en) * | 2013-01-30 | 2014-07-31 | Electronics And Telecommunications Research Institute | Prioritized dual caching method and apparatus |
CN104750858A (en) * | 2015-04-16 | 2015-07-01 | 成都影泰科技有限公司 | Network-based data storage method |
CN104952032A (en) * | 2015-06-19 | 2015-09-30 | 清华大学 | Graph processing method and device as well as rasterization representation and storage method |
CN105530158A (en) * | 2015-12-16 | 2016-04-27 | 清华大学 | Inter-domain intercommunication control system of heterogeneous optical networks |
CN105912478A (en) * | 2016-04-06 | 2016-08-31 | 中国航空无线电电子研究所 | Dual-buffer mechanism based real-time system multi-task data sharing method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102915254B (en) | task management method and device | |
US7844760B2 (en) | Schedule and data caching for wireless transmission | |
US8266393B2 (en) | Coordination among multiple memory controllers | |
JP2022545103A (en) | Distributed storage system and data processing method | |
CN109697122A (en) | Task processing method, equipment and computer storage medium | |
CN101150485A (en) | A management method for network data transmission of zero copy buffer queue | |
CN103262002A (en) | Optimizing communication of system call requests | |
CN107122244A (en) | A kind of diagram data processing system and method based on many GPU | |
CN101150486A (en) | A management method for receiving network data of zero copy buffer queue | |
CN109388590A (en) | Promote the dynamic buffering block management method and device of multi-channel DMA access performance | |
US20200250525A1 (en) | Lightweight, highspeed and energy efficient asynchronous and file system-based ai processing interface framework | |
CN110032450A (en) | A kind of extensive deep learning method and system based on solid-state disk exented memory | |
CN103164440B (en) | The Spatial Data Engine method of Virtual reality | |
CN109301936B (en) | Intelligent substation operation and maintenance information monitoring system based on container scheduling framework | |
CN111653317B (en) | Gene comparison acceleration device, method and system | |
CN117251275B (en) | Multi-application asynchronous I/O request scheduling method, system, equipment and medium | |
Narantuya et al. | Multi-Agent Deep Reinforcement Learning-Based Resource Allocation in HPC/AI Converged Cluster. | |
CN116069480B (en) | Processor and computing device | |
CN102487401B (en) | A kind of document down loading method and device | |
CN109918017A (en) | Data dispatching method and device | |
CN104657326A (en) | Double-CPU communication method and system and system-on-chip chip | |
CN101634956B (en) | Dispatching method and dispatcher of information of polynuclear processor | |
US11734551B2 (en) | Data storage method for speech-related DNN operations | |
Pineiro et al. | Rad-flows: Buffering for predictable communication | |
CN115509644A (en) | Calculation force unloading method and device, electronic equipment and storage medium |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190621 |