CN103294726A - Method and equipment for processing video file - Google Patents
Method and equipment for processing video file Download PDFInfo
- Publication number
- CN103294726A CN103294726A CN2012100552135A CN201210055213A CN103294726A CN 103294726 A CN103294726 A CN 103294726A CN 2012100552135 A CN2012100552135 A CN 2012100552135A CN 201210055213 A CN201210055213 A CN 201210055213A CN 103294726 A CN103294726 A CN 103294726A
- Authority
- CN
- China
- Prior art keywords
- video file
- piecemeal
- computing node
- schedule information
- remainder
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Information Transfer Between Computers (AREA)
Abstract
The invention provides a control node, comprising a block number calculating unit, a block scheduling unit and a scheduling information transmitting unit, wherein the block number calculating unit is configured to calculate the number of blocks of a video file; the block scheduling unit is configured to perform modular computation on the number of the blocks by using the frame number of each frame of the video file to generate scheduling information comprising the block number and the specific remainder of the modular computation; and the scheduling information transmitting unit is configured to transmit the scheduling information to a computational node. The invention also provides a corresponding video file processing method. In addition, the invention also provides the computational node and a corresponding video file processing method. According to the invention, the processing expenditure of the control node can be reduced, and the distributed processing process of the video file is flexible.
Description
Technical field
The present invention relates to the distributed treatment of file, be specifically related to a kind of method and apparatus for the treatment of video file.
Background technology
Along with the development of computer system, the scale that data are handled becomes increasing.Usually, the expense that large-scale data is handled is very large, this be embodied in otherwise computing time very long, or need a large amount of computational resources.
Google company is at the large-scale data Treatment Design and developed special-purpose distributed computing framework: mapping-abbreviation (Map-Reduce).Mapping-abbreviation framework is being handled performance brilliance aspect the large-scale parallel task, comes into vogue in the Distributed Calculation field very soon.
In order to handle big video file at distributed platform, this video file need be cut into little file earlier.The concrete grammar of file cutting will directly influence the overall performance of parallel processing.A kind of cutting-converge system architecture of the distributed video file processing of (Split-Merge) that is called as has been proposed.In this system architecture, video file will at first be cut into a plurality of, and wherein the size of piece is by coded system and the common decision of coding parameter.Then, different pieces is distributed on the different computing nodes handles.At last, after all are finished dealing with, the result of each piece is converged the final complete results of formation.
Fig. 1 shows the block diagram of a kind of cutting-convergence type video file disposal system.As shown in Figure 1, this system comprises control node 10 and computing node 20.Need to prove that although only show a computing node 20 among Fig. 1, yet this only is schematic; It will be understood by those skilled in the art that and in actual conditions, may have a plurality of computing nodes 20.
As shown in Figure 1, control node 10 comprises file information analysis unit 110, block size computing unit 120, cutting scheduling unit 130 and schedule information transmitting element 140.Computing node 20 comprises schedule information reading unit 210, document alignment unit 220, processing unit 230 and output collector unit 240.
Particularly, control node 10 at first utilizes 110 pairs of these video files of file information analysis unit to analyze behind the video file that receives user's submission, obtains to comprise information such as file size, coding method and parameter.Then, block size computing unit 120 is according to the size of above-mentioned information calculations video piecemeal.After obtaining block size, cutting scheduling unit 130 is dispatched to different computing node 20 with each piecemeal, and schedule information transmitting element 140 comprises that to computing node 20 transmissions piecemeal is with respect to the schedule information of the side-play amount of file header and branch block size.
On the other hand, computing node 20 at first utilizes schedule information reading unit 210 to read in schedule information after receiving schedule information.Then, document alignment unit 220 obtains the appropriate section of video file according to schedule information.Next, 230 pairs of these parts of processing unit are handled.At last, the result of calculation that output collector unit 240 is collected this computing node 20 is to be used for gathering the result of calculation that obtains whole file.
As can be seen, in above-mentioned video file disposal system, at first to calculate the size of whole file and the size of each piecemeal, produce the information (comprising that each piecemeal is with respect to the side-play amount of file header and the size of piecemeal) of the piecemeal that each computing node need handle then based on this.Yet the process of calculating the size of the size of whole file and each piecemeal can consume considerable computational resource, thereby causes very big processing expenditure for the control node.On the other hand, more if the video file that the user submits to is handled request, then control the request that node is difficult to timely process user, thereby may cause delay.
Summary of the invention
Therefore, need a kind of cutting method of video file more flexibly and equipment, it can alleviate the processing expenditure of control node, and makes the distributed treatment process of video file more flexible.
According to an aspect of the present invention, provide a kind of control node, having comprised: piecemeal number computing unit is configured to calculate the piecemeal number of video file; The piecemeal scheduling unit is configured to the frame number of each frame in the video file is carried out modular arithmetic to described piecemeal number, comprises the schedule information of the specific remainder of piecemeal number and modular arithmetic with generation; And the schedule information transmitting element, be configured to send described schedule information to computing node.
Preferably, piecemeal number computing unit is configured to: calculate the number of available computing node, and with this piecemeal number as video file.
Preferably, piecemeal number computing unit is configured to: calculate the number of available computing node, and the number of the size of video file divided by available computing node; And if the result of this division divides block size greater than minimum, then with the number of available computing node as described piecemeal number; Otherwise the merchant who divides block size divided by minimum with the size of video file is as described piecemeal number, and wherein, the described minimum block size that divides is according to historical statistical data and predetermined.
Preferably, piecemeal number computing unit is configured to: calculate the size of the number of video file, minimum video file and the number of available computing node; And if the number of video file is less than the number of available computing node, then the size of video file divided by the merchant of the size of minimum video file as described piecemeal number; Otherwise the merchant who divides block size divided by minimum with the size of video file is as described piecemeal number, and wherein, the described minimum block size that divides is according to historical statistical data and predetermined.
Preferably, the piecemeal scheduling unit is configured to: the remainder of modular arithmetic is integrated into mean allocation between the computing node, thereby produces the schedule information of the remainder subclass that comprises piecemeal number and modular arithmetic.
Preferably, the piecemeal scheduling unit is configured to: according to the computing power of computing node, the remainder of modular arithmetic is integrated between the computing node distributes, thereby produce the schedule information of the remainder subclass that comprises piecemeal number and modular arithmetic.
According to another aspect of the present invention, provide a kind of computing node, having comprised: the schedule information reading unit, be configured to read schedule information from the control node, described schedule information comprises the piecemeal number of video file and the specific remainder of modular arithmetic; Filter element frame is configured to determine whether to handle according to the schedule information that reads the present frame of video file; Processing unit is configured to the present frame of video file is handled; And the output collector unit, the result that is configured to collect described processing unit, and result is sent to the control node.
Preferably, filter element is configured to: if the remainder that the frame number of the present frame of video file carries out modular arithmetic to described piecemeal number equals described specific remainder, then determine the present frame of described video file is handled.
According to another aspect of the present invention, provide a kind of method of carrying out at control node place for the treatment of video file, described method comprises: the piecemeal number of calculating video file; The frame number of each frame in the video file is carried out modular arithmetic to described piecemeal number, comprise the schedule information of the specific remainder of piecemeal number and modular arithmetic with generation; And send described schedule information to computing node.
Preferably, the step of calculating the piecemeal number of video file comprises: calculate the number of available computing node, and with this piecemeal number as video file.
Preferably, the step of calculating the piecemeal number of video file comprises: calculate the number of available computing node, and the number of the size of video file divided by available computing node; And if the result of this division divides block size greater than minimum, then with the number of available computing node as described piecemeal number; Otherwise the merchant who divides block size divided by minimum with the size of video file is as described piecemeal number, and wherein, the described minimum block size that divides is according to historical statistical data and predetermined.
Preferably, the step of the piecemeal number of calculating video file comprises: calculate the size of the number of video file, minimum video file and the number of available computing node; And if the number of video file is less than the number of available computing node, then the size of video file divided by the merchant of the size of minimum video file as described piecemeal number; Otherwise the merchant who divides block size divided by minimum with the size of video file is as described piecemeal number, and wherein, the described minimum block size that divides is according to historical statistical data and predetermined.
Preferably, the remainder of modular arithmetic is integrated into mean allocation between the computing node, thereby produces the schedule information of the remainder subclass that comprises piecemeal number and modular arithmetic.
Preferably, according to the computing power of computing node, the remainder of modular arithmetic is integrated between the computing node distributes, thereby produce the schedule information of the remainder subclass that comprises piecemeal number and modular arithmetic.
According to another aspect of the present invention, a kind of method of carrying out at the computing node place for the treatment of video file is provided, described method comprises: read schedule information from the control node, described schedule information comprises the piecemeal number of video file and the specific remainder of modular arithmetic; Determine whether to handle the present frame of video file according to the schedule information that reads; Handle the present frame of video file if desired, then the present frame of video file is handled; And the result of collecting described processing unit, and result is sent to the control node.
Preferably, if the remainder that the frame number of the present frame of video file carries out modular arithmetic to described piecemeal number equals described specific remainder, then determine the present frame of described video file is handled.
According to the present invention, do not need to calculate size and the accurate index information of each piecemeal in the video file, the number that only needs to determine piecemeal gets final product.In other words, control node according to the present invention does not need to calculate each video piecemeal with respect to side-play amount and the size of file header, and only need calculate the number of the whole video file being carried out piecemeal when scheduling video piecemeal.Thus, reduced the processing expenditure of control node, and made the distributed treatment process of video file more flexible.
Description of drawings
By detailed description with the accompanying drawing hereinafter, above-mentioned and further feature of the present invention will become more apparent, wherein:
Fig. 1 shows the block diagram of cutting of the prior art-convergence type video file disposal system.
Fig. 2 shows the block diagram of distributed video document handling system according to an embodiment of the invention.
Fig. 3 shows according to the process flow diagram of one embodiment of the invention by the video slicing method of control node execution.
Fig. 4 shows according to the process flow diagram of one embodiment of the invention by the method for processing video frequency of computing node execution.
Embodiment
Below, in conjunction with the drawings to the description of specific embodiments of the invention, principle of the present invention and realization will become obvious.Should be noted in the discussion above that the present invention should not be limited to specific embodiment hereinafter described.In addition, for for simplicity, omitted the detailed description of the known technology that has nothing to do with the present invention.
Fig. 2 shows the block diagram of distributed video document handling system according to an embodiment of the invention.As shown in Figure 2, this video file disposal system comprises control node 30 and computing node 40.Wherein, control node 30 is supvrs of whole distributed treatment process, and it is responsible for video file is carried out cutting and the video piecemeal after scheduling cutting between different computing nodes.Computing node 40 is nodes of being responsible for the video piecemeal is carried out actual treatment in the distributed system.
Need to prove that although only show a computing node 40 among Fig. 2, yet this only is schematic.It will be understood by those skilled in the art that and in actual conditions, may have a plurality of computing nodes 40.
Particularly, the control node 30 in the distributed video document handling system of present embodiment comprises: piecemeal number computing unit 310, piecemeal scheduling unit 320 and schedule information transmitting element 330.Below, 3 concrete operations of describing the control node 30 in the present embodiment in detail by reference to the accompanying drawings.
Fig. 3 shows according to the process flow diagram of one embodiment of the invention by the video slicing method 50 of control node 30 execution among Fig. 2.As shown in Figure 3, method 50 begins at step S510 place.
At step S520, after control node 30 obtained video file to be processed from the user, control node 30 at first calculated the size (for example totalframes M) of this video file.Then, at step S530, piecemeal number computing unit 310 calculates the piecemeal number N of this video file.Some examples of piecemeal number computing unit 310 calculating piecemeal number N have hereinafter been provided.Yet following example only is schematic description, those skilled in the art will appreciate that the computation process that can also adopt other calculates piecemeal number N.
In one example, piecemeal number computing unit 310 calculates the number P of current available computing node in the distributed systems, and with this as the piecemeal number N of video file (that is, N=P).The advantage of this example is: computation process is simple, can farthest save the computing cost of control node.
In another example, piecemeal number computing unit 310 calculates the number P of available computing node, and the number P of the big or small M of video file divided by available computing node.If the result of this division (M/P) divide block size S greater than minimum, then piecemeal number computing unit 310 with the number P of available computing node as described piecemeal number N (that is, N=P).Otherwise piecemeal number computing unit 310 divides block size with the size of video file divided by minimum merchant (M/S) as described piecemeal number N (that is, N=M/S).Wherein, minimum branch block size S is predetermined value, and this value can obtain according to historical statistical data, and perhaps the empirical data according to the Systems Operator arranges.
The advantage of this example is: computing node can produce overhead when calculating each piecemeal, such as reading of video file data, and collection of result etc.Therefore, the piecemeal number too much also can influence overall performance.By the minimum block size S that divides is set, can make the overall performance optimization of system.
In another example, if the user has submitted a plurality of video files to and wished to handle simultaneously, then piecemeal number computing unit 310 calculates number n, the big or small s of minimum video file of video file and the number P of available computing node.If the number n of video file is less than the number P of available computing node, then piecemeal number computing unit 310 the big or small M of video file divided by the merchant of the big or small s of minimum video file as described piecemeal number (that is, N=M/s).Otherwise piecemeal number computing unit 310 divides block size S with the big or small M of video file divided by minimum merchant as described piecemeal number (that is, N=M/S).Wherein, minimum branch block size S is predetermined value, and this value can obtain according to historical statistical data, and perhaps the empirical data according to the Systems Operator arranges.
Similarly, the advantage of this example is: by the minimum block size S that divides is set under the situation of handling a plurality of video files at the same time, can make the overall performance optimization of system.
Need to prove that in this article, " A " is actually divided by merchant's's (" A " and " B " is positive integer) of " B " implication: be not less than " A " divided by the merchant's (A/B) of " B " minimum positive integer.
Get back to Fig. 3, at step S540, piecemeal scheduling unit 320 is dispatched (distribution) with each piecemeal of video file to the computing node in the system according to the piecemeal number, to produce schedule information.According to the present invention, controller does not need to calculate each video piecemeal with respect to side-play amount and the size of file header, and only need calculate the number that the whole video file is carried out piecemeal when scheduling video piecemeal.
Below enumerated two kinds of concrete scheduling (distribution) process.Similarly, those skilled in the art will appreciate that scheduling (distribution) process that can also adopt other is next to each computing node scheduling (distribution) video piecemeal.
In one example, the piecemeal of piecemeal scheduling unit 320 mean allocation video file between each computing node.For example, if the piecemeal number that piecemeal number computing unit 310 calculates is 10 (namely, pending video file is divided into 10 video piecemeals), and 5 available computing nodes of existence in the system, then piecemeal scheduling unit 320 can be given 5 computing nodes with these 10 video piecemeal mean allocation, makes each computing node be responsible for handling 2 video piecemeals.
For example, the piecemeal number of the frame number i N of each frame carries out modular arithmetic and can obtain remainder n in the video file, the scope of this remainder n be [0, N).So, piecemeal scheduling unit 320 can N) be distributed to each computing node with remainder n[0.If piecemeal number N is 10, available computing node has 5, and piecemeal scheduling unit 320 can be with s1{0 so, 1}, s2{2, and 3}, s3{4,5}, s4{6,7} and s5{8,9} distribute to 5 computing nodes respectively.Like this, suppose that first computing node receives schedule information s1{0,1}, then only frame number i in the video file and 10 to be carried out remainder after the modular arithmetic be that 0 and 1 frame of video is handled to this first computing node, and do not handle other frame of video.Hereinafter with reference Fig. 4 is described in detail this.
In another example, piecemeal scheduling unit 320 distributes the piecemeal of video file according to the computing power of each computing node (for example the number of idle CPU, etc.) between computing node.Still to be example above, suppose that piecemeal number N is 10, available computing node has 5.Yet first computing node in these 5 computing nodes and the performance of second computing node are the strongest, and the 3rd computing node takes second place, and the performance of the 4th computing node and the 5th computing node is the most weak.So, piecemeal scheduling unit 320 can be with s1{0,1,2}, s2{3, and 4,5}, s3{6,7}, s4{8} and s5{9} distribute to these computing nodes successively.That is, first computing node is responsible for handling frame number i and 10 in the video file, and to carry out remainder after the modular arithmetic be 0,1 and 2 frame of video, and to carry out remainder after the modular arithmetic be 8 frame of video and the 4th computing node only is responsible for handling frame number i and 10 in the video file.As can be seen, by considering the computing power of each computing node, can realize the optimization of entire system performance.
Get back to Fig. 3, at step S550, schedule information transmitting element 330 sends schedule information to computing node.For example, (piecemeal number N is 10 in the above example, there are 5 and calculate computing node), the schedule information that sends to first computing node comprises s1 and N, and the schedule information that sends to second computing node comprises that s2 and N...... and the schedule information that sends to the 5th computing node comprise s5 and N.
At last, method 50 finishes at step S560 place.
On the other hand, get back to Fig. 2, computing node 40 comprises schedule information reading unit 410, filter element frame 420, processing unit 430 and output collector unit 440.Below, describe the concrete operations of the computing node 40 of present embodiment in detail with reference to figure 4.
Fig. 4 shows according to the process flow diagram of one embodiment of the invention by the method for processing video frequency 60 of computing node 40 execution among Fig. 2.As shown in Figure 4, method 60 begins at step S610 place.
At step S620, the schedule information reading unit 410 in the computing node 40 reads the schedule information that control node 30 sends.For example, in this example, this schedule information can comprise remainder subclass s and the piecemeal number N that distributes.
Next, at step S630, filter element frame 420 reads the frame number i of the present frame of video file.Then, at step S640,420 couples of frame number i of filter element frame and piecemeal number N carry out modular arithmetic (calculating x=i%N).At step S650, filter element frame 420 judges whether x falls into remainder subclass s.Particularly, if x falls into remainder subclass s, show that then present frame i should be handled by this computing node 40.Otherwise, show that present frame i can't help this computing node 40 and handles.
Need handle present frame i if judge computing node 40, then at step S660 place, carry out corresponding the processing by processing unit 430.
At step S670, judge whether present frame arrives the afterbody of video file.If not, then return step S630, continue to carry out corresponding operation by filter element frame 420.Otherwise, if arrive the afterbody of video file, then proceed to step S680.
At step S680, the output collector unit 440 collection and treatment results of computing node 40, and send to control node 30.Like this, after control node 30 obtains all video piecemeal results, can re-construct the result of whole video file, and return to the user.
At last, method 60 finishes at step S690 place.
Need to prove, although only described the processing procedure to a video file among Fig. 4, those skilled in the art will appreciate that if desired, can carry out same processing to a plurality of video files simultaneously.
According to the present invention, the control node does not need to calculate each video piecemeal with respect to side-play amount and the size of file header, and only need calculate the piecemeal number of whole video file when scheduling video piecemeal.Thus, greatly reduced the processing expenditure of control node, thereby made the distributed treatment process of video file more flexible.
Should be appreciated that the above embodiment of the present invention can realize by both combinations of software, hardware or software and hardware.For example, the various assemblies of the control node in above-described embodiment and computing node and their inside can be realized by multiple device, these devices include but not limited to: general processor, digital signal processing (DSP) circuit, programmable processor, special IC (ASIC), field programmable gate array (FPGA), programmable logic device (PLD) (CPLD), etc.
In addition, those skilled in the art will appreciate that video file that the user that describes in the embodiment of the invention submits to can be stored in user's the local data base, can be stored in the distributed data base or can be stored in the long-range private database.Equally, the result after the Video processing also can be stored in these databases.
In addition, embodiments of the invention disclosed herein can be realized at computer program.More specifically, this computer program is following a kind of product: have computer-readable medium, coding has computer program logic on the computer-readable medium, and when when computing equipment is carried out, this computer program logic provides relevant operation to realize technique scheme of the present invention.When at least one processor of computing system is carried out, computer program logic makes processor carry out the described operation of the embodiment of the invention (method).This set of the present invention typically be provided as on the computer-readable medium that arranges or be coded in for example light medium (for example CD-ROM), floppy disk or hard disk etc. software, code and/or other data structures or such as other media or the Downloadable software image in one or more module, the shared data bank etc. of the firmware on one or more ROM or RAM or the PROM chip or microcode.Software or firmware or this configuration can be installed on the computing equipment, so that the one or more processors in the computing equipment are carried out the described technical scheme of the embodiment of the invention.
Although below show the present invention in conjunction with the preferred embodiments of the present invention, one skilled in the art will appreciate that under the situation that does not break away from the spirit and scope of the present invention, can carry out various modifications, replacement and change to the present invention.Therefore, the present invention should not limited by above-described embodiment, and should be limited by claims and equivalent thereof.
Claims (16)
1. control node for one kind, comprising:
Piecemeal number computing unit is configured to calculate the piecemeal number of video file;
The piecemeal scheduling unit is configured to the frame number of each frame in the video file is carried out modular arithmetic to described piecemeal number, comprises the schedule information of the specific remainder of piecemeal number and modular arithmetic with generation; And
The schedule information transmitting element is configured to send described schedule information to computing node.
2. control node according to claim 1, wherein, described piecemeal number computing unit is configured to: calculate the number of available computing node, and with this piecemeal number as video file.
3. control node according to claim 1, wherein, described piecemeal number computing unit is configured to:
Calculate the number of available computing node, and the number of the size of video file divided by available computing node; And
If the result of this division divides block size greater than minimum, then with the number of available computing node as described piecemeal number; Otherwise the merchant who divides block size divided by minimum with the size of video file is as described piecemeal number,
Wherein, the described minimum block size that divides is according to historical statistical data and predetermined.
4. control node according to claim 1, wherein, described piecemeal number computing unit is configured to:
Calculate the size of the number of video file, minimum video file and the number of available computing node; And
If the number of video file is less than the number of available computing node, then the size of video file divided by the merchant of the size of minimum video file as described piecemeal number; Otherwise the merchant who divides block size divided by minimum with the size of video file is as described piecemeal number,
Wherein, the described minimum block size that divides is according to historical statistical data and predetermined.
5. control node according to claim 1, wherein, described piecemeal scheduling unit is configured to: the remainder of modular arithmetic is integrated into mean allocation between the computing node, thereby produces the schedule information of the remainder subclass that comprises piecemeal number and modular arithmetic.
6. control node according to claim 1, wherein, described piecemeal scheduling unit is configured to: according to the computing power of computing node, the remainder of modular arithmetic is integrated between the computing node distributes, thereby produce the schedule information of the remainder subclass that comprises piecemeal number and modular arithmetic.
7. computing node comprises:
The schedule information reading unit is configured to read schedule information from the control node, and described schedule information comprises the piecemeal number of video file and the specific remainder of modular arithmetic;
Filter element frame is configured to determine whether to handle according to the schedule information that reads the present frame of video file;
Processing unit is configured to the present frame of video file is handled; And
The output collector unit, the result that is configured to collect described processing unit, and result is sent to the control node.
8. computing node according to claim 7, wherein, described filter element is configured to:
If the remainder that the frame number of the present frame of video file carries out modular arithmetic to described piecemeal number equals described specific remainder, then determine the present frame of described video file is handled.
9. controlling the method for the treatment of video file that the node place carries out for one kind, described method comprises:
Calculate the piecemeal number of video file;
The frame number of each frame in the video file is carried out modular arithmetic to described piecemeal number, comprise the schedule information of the specific remainder of piecemeal number and modular arithmetic with generation; And
Send described schedule information to computing node.
10. method according to claim 9, wherein, the step of calculating the piecemeal number of video file comprises: calculate the number of available computing node, and with this piecemeal number as video file.
11. method according to claim 9, wherein, the step of calculating the piecemeal number of video file comprises:
Calculate the number of available computing node, and the number of the size of video file divided by available computing node; And
If the result of this division divides block size greater than minimum, then with the number of available computing node as described piecemeal number; Otherwise the merchant who divides block size divided by minimum with the size of video file is as described piecemeal number,
Wherein, the described minimum block size that divides is according to historical statistical data and predetermined.
12. method according to claim 9, wherein, the step of calculating the piecemeal number of video file comprises:
Calculate the size of the number of video file, minimum video file and the number of available computing node; And
If the number of video file is less than the number of available computing node, then the size of video file divided by the merchant of the size of minimum video file as described piecemeal number; Otherwise the merchant who divides block size divided by minimum with the size of video file is as described piecemeal number,
Wherein, the described minimum block size that divides is according to historical statistical data and predetermined.
13. method according to claim 9 wherein, is integrated into mean allocation between the computing node with the remainder of modular arithmetic, thereby produces the schedule information of the remainder subclass that comprises piecemeal number and modular arithmetic.
14. method according to claim 9 wherein, according to the computing power of computing node, is integrated into the remainder of modular arithmetic between the computing node and distributes, thereby produces the schedule information of the remainder subclass that comprises piecemeal number and modular arithmetic.
15. the method for carrying out at the computing node place for the treatment of video file, described method comprises:
Read schedule information from the control node, described schedule information comprises the piecemeal number of video file and the specific remainder of modular arithmetic;
Determine whether to handle the present frame of video file according to the schedule information that reads;
Handle the present frame of video file if desired, then the present frame of video file is handled; And
Collect the result of described processing, and result is sent to the control node.
16. method according to claim 15 wherein, if the remainder that the frame number of the present frame of video file carries out modular arithmetic to described piecemeal number equals described specific remainder, is then determined the present frame of described video file is handled.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2012100552135A CN103294726A (en) | 2012-03-05 | 2012-03-05 | Method and equipment for processing video file |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2012100552135A CN103294726A (en) | 2012-03-05 | 2012-03-05 | Method and equipment for processing video file |
Publications (1)
Publication Number | Publication Date |
---|---|
CN103294726A true CN103294726A (en) | 2013-09-11 |
Family
ID=49095612
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN2012100552135A Pending CN103294726A (en) | 2012-03-05 | 2012-03-05 | Method and equipment for processing video file |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103294726A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105302853A (en) * | 2015-09-17 | 2016-02-03 | 浪潮(北京)电子信息产业有限公司 | Data reconstruction method and apparatus |
CN106874277A (en) * | 2015-12-10 | 2017-06-20 | 北大方正集团有限公司 | Distributed data processing system and method |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20020039904A1 (en) * | 1999-01-08 | 2002-04-04 | Anderson Robert J. | Monitoring of call information in a wireless location system |
CN101119319A (en) * | 2007-09-19 | 2008-02-06 | 腾讯科技(深圳)有限公司 | Method, transmitting/receiving device and system against lost packet in data transmission process |
CN101198053A (en) * | 2006-12-08 | 2008-06-11 | 中兴通讯股份有限公司 | Video data self-adapting partition device in encoder |
CN101365000A (en) * | 2008-09-24 | 2009-02-11 | 清华大学 | Stream media data processing method and network node |
CN101472166A (en) * | 2007-12-26 | 2009-07-01 | 华为技术有限公司 | Method for caching and enquiring content as well as point-to-point medium transmission system |
-
2012
- 2012-03-05 CN CN2012100552135A patent/CN103294726A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20020039904A1 (en) * | 1999-01-08 | 2002-04-04 | Anderson Robert J. | Monitoring of call information in a wireless location system |
CN101198053A (en) * | 2006-12-08 | 2008-06-11 | 中兴通讯股份有限公司 | Video data self-adapting partition device in encoder |
CN101119319A (en) * | 2007-09-19 | 2008-02-06 | 腾讯科技(深圳)有限公司 | Method, transmitting/receiving device and system against lost packet in data transmission process |
CN101472166A (en) * | 2007-12-26 | 2009-07-01 | 华为技术有限公司 | Method for caching and enquiring content as well as point-to-point medium transmission system |
CN101365000A (en) * | 2008-09-24 | 2009-02-11 | 清华大学 | Stream media data processing method and network node |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105302853A (en) * | 2015-09-17 | 2016-02-03 | 浪潮(北京)电子信息产业有限公司 | Data reconstruction method and apparatus |
CN106874277A (en) * | 2015-12-10 | 2017-06-20 | 北大方正集团有限公司 | Distributed data processing system and method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Petra et al. | Real-time stochastic optimization of complex energy systems on high-performance computers | |
CN102063336B (en) | Distributed computing multiple application function asynchronous concurrent scheduling method | |
CN106202431A (en) | A kind of Hadoop parameter automated tuning method and system based on machine learning | |
CN106339351B (en) | A kind of SGD algorithm optimization system and method | |
CN102779223A (en) | Method and device for forecasting short-term power load | |
CN112685153A (en) | Micro-service scheduling method and device and electronic equipment | |
CN102314521B (en) | Distributed parallel Skyline inquiring method based on cloud computing environment | |
CN105373517A (en) | Spark-based distributed matrix inversion parallel operation method | |
Radchenko et al. | Micro-workflows: Kafka and kepler fusion to support digital twins of industrial processes | |
CN115099543A (en) | Path planning method, device and equipment for battery replacement and storage medium | |
Herda | Parallel genetic algorithm for capacitated p-median problem | |
CN103294726A (en) | Method and equipment for processing video file | |
Chen | A hybrid solution of fork/join synchronization in parallel queues | |
KR102477509B1 (en) | Method for generating weather prediction information and computing device for executing the method | |
CN101495978A (en) | Reduction of message flow between bus-connected consumers and producers | |
Marcos Alvarez et al. | Machine learning to balance the load in parallel branch-and-bound | |
Yasudo et al. | Performance estimation for exascale reconfigurable dataflow platforms | |
CN106844024A (en) | The GPU/CPU dispatching methods and system of a kind of self study run time forecast model | |
CN110175680A (en) | Utilize the internet of things data analysis method of the online machine learning of distributed asynchronous refresh | |
CN106446308A (en) | Sparse Bayesian-based fault locating method and system | |
Bani-Mohammad et al. | Comparative evaluation of contiguous allocation strategies on 3D mesh multicomputers | |
CN111858530B (en) | Real-time correlation analysis method and system based on mass logs | |
CN107274005A (en) | The Urban Storm Flood disaster alarm method and system of fractal algorithm based on big data | |
CN102833200A (en) | DPD (Digital Pre Distortion) self-adapting method and device based on symmetric multiprocessors | |
Yu et al. | A Predictive Dispatching Rule Assisted by Multi-Layer Perceptron for Scheduling Wafer Fabrication Lines |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20130911 |
|
RJ01 | Rejection of invention patent application after publication |