CN106888400B - A kind of method and system for realizing transcoding task schedule - Google Patents
A kind of method and system for realizing transcoding task schedule Download PDFInfo
- Publication number
- CN106888400B CN106888400B CN201510930440.1A CN201510930440A CN106888400B CN 106888400 B CN106888400 B CN 106888400B CN 201510930440 A CN201510930440 A CN 201510930440A CN 106888400 B CN106888400 B CN 106888400B
- Authority
- CN
- China
- Prior art keywords
- total
- gpu card
- decoding capability
- gpu
- ability
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 28
- 238000006243 chemical reaction Methods 0.000 abstract description 3
- 230000007812 deficiency Effects 0.000 description 8
- 238000010586 diagram Methods 0.000 description 6
- 230000001771 impaired effect Effects 0.000 description 4
- 230000003472 neutralizing effect Effects 0.000 description 2
- 238000010187 selection method Methods 0.000 description 2
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/44—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream or rendering scenes according to encoded video stream scene graphs
- H04N21/4402—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream or rendering scenes according to encoded video stream scene graphs involving reformatting operations of video signals for household redistribution, storage or real-time display
- H04N21/440236—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream or rendering scenes according to encoded video stream scene graphs involving reformatting operations of video signals for household redistribution, storage or real-time display by media transcoding, e.g. video is transformed into a slideshow of still pictures, audio is converted into text
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
- Compression, Expansion, Code Conversion, And Decoders (AREA)
Abstract
The invention discloses a kind of method and systems for realizing transcoding task schedule, are related to video code conversion field.Method therein includes: to count GPU card set current total coding ability Es and total decoding capability Ds；Judge whether total coding ability Es matches with total decoding capability Ds；If total coding ability Es is not matched that with total decoding capability Ds, the code capacity or decoding capability of corresponding GPU card in GPU card set are adjusted, so that the total coding ability Es of GPU card set matches with total decoding capability Ds.The present invention combines the relative influence relationship of coding/decoding capability by being based on coding/decoding capability matching principle for the coding of GPU card and decoding capability separate computations, carries out task schedule, can maximally utilise the ability of GPU, reach the maximum throughput rate of transcoding task.
Description
Technical field
Video code conversion field of the present invention more particularly to a kind of method and system for realizing transcoding task schedule.
Background technique
Decoding and coding are the producer and Consumer relationships in video code conversion task, and decoding must phase with code capacity
Match.And there are coding/decoding capability asymmetry problems, some GPU cards, and coding or decoded ability is individually performed for GPU card in reality
Than being 1:3, and some GPU card code capacities for being configured with own coding circuit may be suitable or even stronger with decoding capability.And
And GPU card, when executing encoding and decoding at the same time, there are influence relationships for coding/decoding capability.According to test result, GPU card compiles solution simultaneously
When code, the influence relationship decoded to coding is as shown in Figure 1.Abscissa is kd, ordinate ke.When solution code amount is smaller, decoding pair
Coding does not have an impact, and influences coefficient k at this timeeIt is 1.When solving, code amount is larger, reaches certain threshold value, then can generate shadow to coding
It rings, influences coefficient ke<1.When decoding reaches maximum value, code capacity is impaired the most serious, remembers k at this timeeFor kmin.Due to
GPU card coding/decoding capability is asymmetric, and when leading to carry out the distribution of transcoding task as unit of card, processing capacity is limited.
Summary of the invention
The present invention is to solve asymmetrical more asymmetric problems of GPU card coding/decoding capability.
According to an aspect of the present invention, a kind of method for realizing transcoding task schedule is proposed, comprising: statistics GPU card collection
Close current total coding ability Es and total decoding capability Ds；Judge whether total coding ability Es matches with total decoding capability Ds；
If total coding ability Es is not matched that with total decoding capability Ds, the code capacity or solution of corresponding GPU card in GPU card set are adjusted
Code ability, so that the total coding ability Es of GPU card set matches with total decoding capability Ds.
Further, it if total coding ability Es is not matched that with total decoding capability Ds, adjusts corresponding in GPU card set
If the step of code capacity or decoding capability of GPU card includes: that total decoding capability is insufficient, phase is selected from GPU card set
The decoding capability of selected GPU card is set as maximizing by the GPU card for answering quantity, so as to the total coding ability Es of GPU card set
Match with total decoding capability Ds.
Further, it if total coding ability Es is not matched that with total decoding capability Ds, adjusts corresponding in GPU card set
The step of code capacity or decoding capability of GPU card further include: if total coding scarce capacity, selected from GPU card set
The code capacity of selected GPU card is set as maximizing by the GPU card of respective numbers, so as to the total coding ability of GPU card set
Es matches with total decoding capability Ds.
Further, the GPU card that respective numbers are selected from GPU card set sets the code capacity of selected GPU card
Being set to maximized step includes: in GPU card set, from small to large according to the ratio between the code capacity of GPU card and decoding capability
Sequence, successively select GPU card, the code capacity of selected GPU card be set as maximizing, until GPU card set is current
Total coding ability Es matches with total decoding capability Ds.
Further, judge that the step of whether total coding ability Es and total decoding capability Ds match includes: judgment formula
Whether Ds < Es × k is true, and wherein k is system parameter；If formula Ds < Es × k is set up, determine that total decoding capability is insufficient；If public
Formula Ds < Es × k is invalid, then determines total coding scarce capacity.
According to another aspect of the present invention, it is also proposed that a kind of system for realizing transcoding task schedule, comprising: statistics is single
Member, for counting GPU card set current total coding ability Es and total decoding capability Ds；Judging unit, for judging total coding
Whether ability Es matches with total decoding capability Ds；Adjustment unit, if being used for total coding ability Es and total decoding capability Ds not phase
Matching, then adjust the code capacity or decoding capability of corresponding GPU card in GPU card set, so as to the total coding ability of GPU card set
Es matches with total decoding capability Ds.
Further, if adjustment unit is used for total decoding capability deficiency, respective numbers are selected from GPU card set
The decoding capability of selected GPU card is set as maximizing by GPU card, so that the total coding ability Es of GPU card set is decoded with total
Ability Ds matches.
Further, if adjustment unit is used for total coding scarce capacity, respective numbers are selected from GPU card set
The code capacity of selected GPU card is set as maximizing by GPU card, so that the total coding ability Es of GPU card set is decoded with total
Ability Ds matches.
Further, adjustment unit is used in GPU card set, according to the ratio between the code capacity of GPU card and decoding capability
Sequence from small to large, successively selects GPU card, and the code capacity of selected GPU card is set as maximizing, until GPU card collection
Current total coding ability Es is closed to match with total decoding capability Ds.
Further, whether judging unit is true for judgment formula Ds < Es × k, and wherein k is system parameter；If formula
Ds < Es × k is set up, then determines that total decoding capability Ds is insufficient；If formula Ds < Es × k is invalid, total coding ability Es is determined not
Foot.
Compared with prior art, the present invention is by being based on encoding and decoding energy for the coding of GPU card and decoding capability separate computations
Power matching principle, and the relative influence relationship of coding/decoding capability is combined, task schedule is carried out, can maximally utilise GPU's
Ability reaches the maximum throughput rate of transcoding task.
By referring to the drawings to the detailed description of exemplary embodiment of the present invention, other feature of the invention and its
Advantage will become apparent.
Detailed description of the invention
The attached drawing for constituting part of specification describes the embodiment of the present invention, and together with the description for solving
Release the principle of the present invention.
The present invention can be more clearly understood according to following detailed description referring to attached drawing, in which:
Fig. 1 is influence relational graph of the decoding to coding.
Fig. 2 is flow diagram of the present invention for realizing one embodiment of the method for transcoding task schedule.
Fig. 3 is flow diagram of the present invention for realizing another embodiment of the method for transcoding task schedule.
Fig. 4 is structural schematic diagram of the present invention for realizing one embodiment of the system of transcoding task schedule.
Specific embodiment
Carry out the various exemplary embodiments of detailed description of the present invention now with reference to attached drawing.It should also be noted that unless in addition having
Body explanation, the unlimited system of component and the positioned opposite of step, numerical expression and the numerical value otherwise illustrated in these embodiments is originally
The range of invention.
Simultaneously, it should be appreciated that for ease of description, the size of various pieces shown in attached drawing is not according to reality
Proportionate relationship draw.
Be to the description only actually of at least one exemplary embodiment below it is illustrative, never as to the present invention
And its application or any restrictions used.
Technology, method and apparatus known to person of ordinary skill in the relevant may be not discussed in detail, but suitable
In the case of, the technology, method and apparatus should be considered as authorizing part of specification.
It is shown here and discuss all examples in, any occurrence should be construed as merely illustratively, without
It is as limitation.Therefore, the other examples of exemplary embodiment can have different values.
It should also be noted that similar label and letter indicate similar terms in following attached drawing, therefore, once a certain Xiang Yi
It is defined in a attached drawing, then in subsequent attached drawing does not need that it is further discussed.
To make the objectives, technical solutions, and advantages of the present invention clearer, below in conjunction with specific embodiment, and reference
Attached drawing, the present invention is described in more detail.
Fig. 2 is flow diagram of the present invention for realizing one embodiment of the method for transcoding task schedule.This method
The following steps are included:
In step 210, GPU card set current total coding ability Es and total decoding capability Ds are counted.
It include multiple asymmetrical GPU in GPU (Graphics Processing Unit, graphics processor) card set
Card.
In step 220, judge whether total coding ability Es matches with total decoding capability Ds.If total coding ability Es and total
Decoding capability Ds is not matched that, thens follow the steps 230, otherwise, executes step 240.
The process can be judged according to whether formula Ds < Es × k is true, if formula Ds < Es × k is set up, be determined
Total decoding capability is insufficient；If formula Ds < Es × k is invalid, total coding scarce capacity is determined, wherein k is system parameter.
In step 230, the code capacity or decoding capability of corresponding GPU card in GPU card set are adjusted, so as to GPU card set
Total coding ability Es match with total decoding capability Ds.
In step 240, without adjustment.
In this embodiment, the total coding ability Es and total decoding capability Ds current by statistics GPU card set；Judgement is total
Whether code capacity Es matches with total decoding capability Ds；If total coding ability Es is not matched that with total decoding capability Ds, adjust
The code capacity or decoding capability of corresponding GPU card in whole GPU card set, so that the total coding ability Es of GPU card set is solved with total
Code ability Ds matches.Due to being based on encoding and decoding using the code capacity of GPU card and decoding capability as two dimension separate computations
The matched principle of ability, while fully considering while the relative influence relationship of encoding and decoding, carry out encoding tasks and decoding task tune
Degree, to reach the maximum task throughput of system.
Fig. 3 is flow diagram of the present invention for realizing another embodiment of the method for transcoding task schedule.The party
Method the following steps are included:
In step 310, GPU card set current total coding ability Es and total decoding capability Ds are counted.
In step 320, whether judgment formula Ds < Es × k is true, and wherein k is system parameter.If so, it thens follow the steps
330, otherwise, execute step 360.
If formula Ds < Es × k is set up, determine that total decoding capability is insufficient；If formula Ds < Es × k is invalid, determine total
Code capacity is insufficient.
In step 330, i-th of GPU card is chosen, the decoding capability of the GPU card is set as maximizing.
Since the 1st card select, can an optional GPU card, the decoding capability of the GPU card is set as maximizing, directly
To meeting step 340.
In step 340, judge whether the total coding ability Es of GPU card set matches with total decoding capability Ds.If phase
With then terminating the process.Otherwise, step 350 is executed.
When decoding capability deficiency, decoding capability decision systems ability, the desirable k of impact factormin, the selection of coding card
Code capacity summation need to only be met equal to Ds.
In step 350, i+1 card is chosen.
As unit of card, the GPU card of respective numbers is selected, the decoding capability of selected GPU card is set as maximizing, directly
Total coding ability Es to GPU card set matches with total decoding capability Ds.It is subsequent then not execute the present embodiment other steps
Suddenly.
In step 360, in GPU card set, according to the ratio between code capacity and decoding capability of GPU card from small to large suitable
Sequence.
It is every select a GPU card for maximize decoding, the code capacity of the card can be by EkIt is reduced to Ek*kmin, system total coding
Ability is also with impaired.To make system overall capacity most strong, i.e., it is minimum to be damaged ratio for code capacity, therefore, it is preferential choose by
The smallest GPU card of damage ratio.Therefore, the sequence of the ratio between the code capacity of GPU card and decoding capability from small to large is sorted.
In step 370, i-th of GPU card is successively selected, the decoding capability of the GPU card is set as maximizing.
By the ratio between the code capacity of GPU card and decoding capability sequence from small to large as a result, being selected since the 1st GPU card,
Until meeting step 380.
In step 380, judge whether the total coding ability Es of GPU card set matches with total decoding capability Ds.If phase
With then terminating the process.Otherwise, step 390 is executed.
In step 390, i+1 card is chosen.
As unit of card, GPU card is successively selected, the decoding capability of selected GPU card is set as maximizing, until GPU
The total coding ability Es of card set matches with total decoding capability Ds.
In this embodiment, GPU card selection method when task schedule, in code capacity deficiency, GPU card is by coding energy
Power/decoding capability ratio sorts from small to large to be chosen.And in decoding capability deficiency, all card is for maximum neutralizing
Code, and coding card then selects enough cards.The present invention is using the code capacity of GPU card and decoding capability as two dimensions point
Calculating is opened, the matched principle of coding/decoding capability is based on, while fully considering while the relative influence relationship of encoding and decoding, is encoded
Task and decoding task are dispatched, to reach the maximum task throughput of system.
Fig. 4 is structural schematic diagram of the present invention for realizing one embodiment of the system of transcoding task schedule.The system
Including statistic unit 410, judging unit 420 and adjustment unit 430.
Statistic unit 410 is for counting GPU card set current total coding ability Es and total decoding capability Ds.
It include multiple asymmetrical GPU in GPU (Graphics Processing Unit, graphics processor) card set
Card.
Judging unit 420 is for judging whether total coding ability Es matches with total decoding capability Ds.
Judging unit 420 according to whether formula Ds < Es × k is true for being judged, if formula Ds < Es × k is set up,
Determine that total decoding capability is insufficient；If formula Ds < Es × k is invalid, total coding scarce capacity is determined, wherein k is system parameter.
If adjustment unit 430 is not matched that for total coding ability Es and total decoding capability Ds, adjust in GPU card set
The code capacity or decoding capability of corresponding GPU card, so as to GPU card set total coding ability Es with total decoding capability Ds phase
Match.
In this embodiment, the total coding ability Es and total decoding capability Ds current by statistics GPU card set；Judgement is total
Whether code capacity Es matches with total decoding capability Ds；If total coding ability Es is not matched that with total decoding capability Ds, adjust
The code capacity or decoding capability of corresponding GPU card in whole GPU card set, so that the total coding ability Es of GPU card set is solved with total
Code ability Ds matches.Due to being based on encoding and decoding using the code capacity of GPU card and decoding capability as two dimension separate computations
The matched principle of ability, while fully considering while the relative influence relationship of encoding and decoding, carry out encoding tasks and decoding task tune
Degree, to reach the maximum task throughput of system.
Another embodiment of the invention, statistic unit 410 for count the current total coding ability Es of GPU card set and
Total decoding capability Ds.
Judging unit 420 according to whether formula Ds < Es × k is true for being judged, if formula Ds < Es × k is set up,
Determine that total decoding capability is insufficient；If formula Ds < Es × k is invalid, total coding scarce capacity is determined, wherein k is system parameter.
When decoding capability deficiency, decoding capability decision systems ability.The desirable k of impact factormin, the selection of coding card
Code capacity summation need to only be met equal to Ds.
The decoding capability of the GPU card is set as maximizing by adjustment unit 430 for choosing i-th of GPU card.Judge GPU
Whether the total coding ability Es of card set matches with total decoding capability Ds, if it does not match, choosing i+1 card.
Can an optional GPU card, the decoding capability of the GPU card is set as maximizing.As unit of card, selection is corresponding
The decoding capability of selected GPU card is set as maximizing by the GPU card of quantity, until the total coding ability Es of GPU card set is same
Total decoding capability Ds matches.
When code capacity deficiency, code capacity decision systems ability, it is every select a GPU card for maximize decode, this
The code capacity of card can be by EkIt is reduced to Ek*kmin, system total coding ability is also with impaired.To make system overall capacity most strong,
I.e. code capacity is damaged ratio minimum, therefore, preferential to choose the impaired the smallest GPU card of ratio.
Adjustment unit 430 be used in GPU card set, according to the ratio between the code capacity of GPU card and decoding capability from it is small to
Big sequence.I-th of GPU card is successively selected, the decoding capability of the GPU card is set as maximizing.
As unit of card, GPU card is successively selected, the decoding capability of selected GPU card is set as maximizing, until GPU
The total coding ability Es of card set matches with total decoding capability Ds.
In this embodiment, GPU card selection method when task schedule, in code capacity deficiency, GPU card is by coding energy
Power/decoding capability ratio sorts from small to large to be chosen.And in decoding capability deficiency, all card is for maximum neutralizing
Code, and coding card then selects enough cards.The present invention is using the code capacity of GPU card and decoding capability as two dimensions point
Calculating is opened, the matched principle of coding/decoding capability is based on, while fully considering while the relative influence relationship of encoding and decoding, is encoded
Task and decoding task are dispatched, to reach the maximum task throughput of system.
So far, the present invention is described in detail.In order to avoid covering design of the invention, it is public that this field institute is not described
The some details known.Those skilled in the art as described above, completely it can be appreciated how implementing technology disclosed herein
Scheme.
Method and device of the invention may be achieved in many ways.For example, can by software, hardware, firmware or
Person's software, hardware, firmware any combination realize method and device of the invention.The step of for the method it is above-mentioned
Sequence is merely to be illustrated, and the step of method of the invention is not limited to sequence described in detail above, unless with other sides
Formula illustrates.In addition, in some embodiments, the present invention can be also embodied as recording program in the recording medium, these
Program includes for realizing machine readable instructions according to the method for the present invention.Thus, the present invention also covers storage for executing
The recording medium of program according to the method for the present invention.
Although some specific embodiments of the invention are described in detail by example, the skill of this field
Art personnel it should be understood that above example merely to being illustrated, the range being not intended to be limiting of the invention.The skill of this field
Art personnel are it should be understood that can without departing from the scope and spirit of the present invention modify to above embodiments.This hair
Bright range is defined by the following claims.
Claims (8)
1. a kind of method for realizing transcoding task schedule characterized by comprising
Count the current total coding ability Es and total decoding capability Ds of GPU card set, wherein total coding ability Es and always decode energy
Power Ds includes task throughput；
Judge whether total coding ability Es matches with total decoding capability Ds, wherein whether judgment formula Ds < Es × k is true,
Middle k is system parameter, if formula Ds < Es × k is set up, determines that total decoding capability is insufficient, if formula Ds < Es × k is invalid,
Determine total coding scarce capacity；
If total coding ability Es is not matched that with total decoding capability Ds, the code capacity of corresponding GPU card in GPU card set is adjusted
Or decoding capability, so that the total coding ability Es of GPU card set matches with total decoding capability Ds.
2. the method according to claim 1, wherein
If total coding ability Es is not matched that with total decoding capability Ds, the code capacity of corresponding GPU card in GPU card set is adjusted
Or the step of decoding capability, includes:
If total decoding capability is insufficient, the GPU card of respective numbers is selected from GPU card set, by the decoding of selected GPU card
Capabilities setting is to maximize, so that the total coding ability Es of GPU card set matches with total decoding capability Ds.
3. the method according to claim 1, wherein
If total coding ability Es is not matched that with total decoding capability Ds, the code capacity of corresponding GPU card in GPU card set is adjusted
Or the step of decoding capability further include:
If total coding scarce capacity selects the GPU card of respective numbers from GPU card set, by the coding of selected GPU card
Capabilities setting is to maximize, so that the total coding ability Es of GPU card set matches with total decoding capability Ds.
4. according to the method described in claim 3, it is characterized in that,
The GPU card that respective numbers are selected from GPU card set sets maximized for the code capacity of selected GPU card
Step includes:
In GPU card set, according to the sequence of the ratio between the code capacity of GPU card and decoding capability from small to large, GPU is successively selected
Card, the code capacity of selected GPU card is set as maximizing, until the current total coding ability Es of GPU card set is the same as total solution
Code ability Ds matches.
5. a kind of system for realizing transcoding task schedule characterized by comprising
Statistic unit, for counting GPU card set current total coding ability Es and total decoding capability Ds, wherein total coding energy
Power Es and total decoding capability Ds includes task throughput；
Judging unit, for judging whether total coding ability Es matches with total decoding capability Ds, wherein judging unit is for sentencing
Whether disconnected formula Ds < Es × k is true, and wherein k is system parameter；If formula Ds < Es × k is set up, total decoding capability Ds is determined
It is insufficient；If formula Ds < Es × k is invalid, determine that total coding ability Es is insufficient；
Adjustment unit adjusts corresponding in GPU card set if not matching that for total coding ability Es and total decoding capability Ds
The code capacity or decoding capability of GPU card, so that the total coding ability Es of GPU card set matches with total decoding capability Ds.
6. system according to claim 5, which is characterized in that
If it is insufficient that adjustment unit is used for total decoding capability, the GPU card of respective numbers is selected from GPU card set, it will be selected
The decoding capability for selecting GPU card is set as maximizing, so as to GPU card set total coding ability Es with total decoding capability Ds phase
Match.
7. system according to claim 5, which is characterized in that
If adjustment unit is used for total coding scarce capacity, the GPU card of respective numbers is selected from GPU card set, it will be selected
The code capacity for selecting GPU card is set as maximizing, so as to GPU card set total coding ability Es with total decoding capability Ds phase
Match.
8. system according to claim 7, which is characterized in that
Adjustment unit is used in GPU card set, according to the ratio between code capacity and decoding capability of GPU card from small to large suitable
Sequence successively selects GPU card, and the code capacity of selected GPU card is set as maximizing, until the current editor-in-chief of GPU card set
Code ability Es matches with total decoding capability Ds.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510930440.1A CN106888400B (en) | 2015-12-15 | 2015-12-15 | A kind of method and system for realizing transcoding task schedule |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510930440.1A CN106888400B (en) | 2015-12-15 | 2015-12-15 | A kind of method and system for realizing transcoding task schedule |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106888400A CN106888400A (en) | 2017-06-23 |
| CN106888400B true CN106888400B (en) | 2019-05-10 |
Family
ID=59173474
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510930440.1A Active CN106888400B (en) | 2015-12-15 | 2015-12-15 | A kind of method and system for realizing transcoding task schedule |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106888400B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109213593B (en) * | 2017-07-04 | 2022-05-10 | 阿里巴巴集团控股有限公司 | Resource allocation method, device and equipment for panoramic video transcoding |
| CN108040258B (en) * | 2017-11-21 | 2020-05-15 | 北京臻迪科技股份有限公司 | Encoding and decoding method, device and system |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7511714B1 (en) * | 2003-11-10 | 2009-03-31 | Nvidia Corporation | Video format conversion using 3D graphics pipeline of a GPU |
| CN102273205A (en) * | 2008-11-04 | 2011-12-07 | 先进微装置公司 | Software video transcoder with gpu acceleration |
| CN102541640A (en) * | 2011-12-28 | 2012-07-04 | 厦门市美亚柏科信息股份有限公司 | Cluster GPU (graphic processing unit) resource scheduling system and method |
| CN104768026A (en) * | 2015-04-17 | 2015-07-08 | 中国工商银行股份有限公司 | Multi-channel audio and video transcoding device |
| CN104871132A (en) * | 2012-10-18 | 2015-08-26 | 超威半导体公司 | Media hardware resource allocation |
| US9142004B2 (en) * | 2012-12-20 | 2015-09-22 | Vmware, Inc. | Dynamic allocation of physical graphics processing units to virtual machines |
| CN104980752A (en) * | 2015-06-11 | 2015-10-14 | 武汉大千信息技术有限公司 | Method for realizing multipath self-adaptive parallel transcoding through CPU and GPU and system thereof |
-
2015
- 2015-12-15 CN CN201510930440.1A patent/CN106888400B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7511714B1 (en) * | 2003-11-10 | 2009-03-31 | Nvidia Corporation | Video format conversion using 3D graphics pipeline of a GPU |
| CN102273205A (en) * | 2008-11-04 | 2011-12-07 | 先进微装置公司 | Software video transcoder with gpu acceleration |
| CN102541640A (en) * | 2011-12-28 | 2012-07-04 | 厦门市美亚柏科信息股份有限公司 | Cluster GPU (graphic processing unit) resource scheduling system and method |
| CN104871132A (en) * | 2012-10-18 | 2015-08-26 | 超威半导体公司 | Media hardware resource allocation |
| US9142004B2 (en) * | 2012-12-20 | 2015-09-22 | Vmware, Inc. | Dynamic allocation of physical graphics processing units to virtual machines |
| CN104768026A (en) * | 2015-04-17 | 2015-07-08 | 中国工商银行股份有限公司 | Multi-channel audio and video transcoding device |
| CN104980752A (en) * | 2015-06-11 | 2015-10-14 | 武汉大千信息技术有限公司 | Method for realizing multipath self-adaptive parallel transcoding through CPU and GPU and system thereof |
Non-Patent Citations (1)
| Title |
|---|
| 基于GPU 的视频转码技术研究;黄兴等;《电视技术》;20120102;第36卷(第1期);第26-29页 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106888400A (en) | 2017-06-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104980752B (en) | The method and system of multichannel self-adaptive parallel transcoding are realized using CPU and GPU | |
| CN106888400B (en) | A kind of method and system for realizing transcoding task schedule | |
| CN105812804B (en) | The method and device of optimum quantization value during a kind of determining RDOQ | |
| US11062210B2 (en) | Method and apparatus for training a neural network used for denoising | |
| Campelo et al. | A modified immune network algorithm for multimodal electromagnetic problems | |
| CN109389212A (en) | A kind of restructural activation quantization pond system towards low-bit width convolutional neural networks | |
| Merhav et al. | Optimal watermark embedding and detection strategies under limited detection resources | |
| MY168806A (en) | Linear prediction based audio coding using improved probability distribution estimation | |
| Choi et al. | Tokenmixup: Efficient attention-guided token-level data augmentation for transformers | |
| CN105451020A (en) | Video compression method and device | |
| CN113691812B (en) | Hongmon system-based distributed video processing method, terminal and readable medium | |
| CN107240100A (en) | A kind of image partition method and system based on genetic algorithm | |
| Heindel et al. | Fast CU split decisions for HEVC inter coding using support vector machines | |
| CN103609114B (en) | Code device, decoding apparatus, coding/decoding system, coded method and coding/decoding method | |
| CN106201712B (en) | The method of adjustment of target identification frequency, device and mobile terminal in augmented reality | |
| CN115941112B (en) | Portable hidden communication method, computer equipment and storage medium | |
| Wang et al. | The upper and lower bounds of the prediction accuracies of ensemble methods for binary classification | |
| CN103281532A (en) | Quantification adjustment method and system based on exercise intensity detection | |
| Cogranne et al. | Reliable detection of hidden information based on a non-linear local model | |
| EP2798614A1 (en) | Method of and apparatus for low-complexity detection of periodic textures | |
| CN111628528A (en) | Method and device for eliminating power flow out-of-limit during wind power participation system recovery | |
| Gao et al. | Modified differential evolution for the integer programming problems | |
| CN111291602A (en) | Video detection method and device, electronic equipment and computer readable storage medium | |
| CN107749989A (en) | The coding method of image and device | |
| CN113810695A (en) | Video encoding method, apparatus and computer-readable 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |