CN102023897A - Dynamic self-adapting parallel decoding method and system at multi-core environment - Google Patents

Dynamic self-adapting parallel decoding method and system at multi-core environment Download PDF

Info

Publication number
CN102023897A
CN102023897A CN2010105364843A CN201010536484A CN102023897A CN 102023897 A CN102023897 A CN 102023897A CN 2010105364843 A CN2010105364843 A CN 2010105364843A CN 201010536484 A CN201010536484 A CN 201010536484A CN 102023897 A CN102023897 A CN 102023897A
Authority
CN
China
Prior art keywords
task
decoding
module
parallel decoding
dynamic self
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN2010105364843A
Other languages
Chinese (zh)
Other versions
CN102023897B (en
Inventor
刘立
何丁丁
蓝发财
许建卫
蔡华元
万伟
刘兴彬
王泼
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Dawning Information Industry Beijing Co Ltd
Dawning Information Industry Co Ltd
Original Assignee
Dawning Information Industry Beijing Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Dawning Information Industry Beijing Co Ltd filed Critical Dawning Information Industry Beijing Co Ltd
Priority to CN 201010536484 priority Critical patent/CN102023897B/en
Publication of CN102023897A publication Critical patent/CN102023897A/en
Application granted granted Critical
Publication of CN102023897B publication Critical patent/CN102023897B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Landscapes

  • Compression Or Coding Systems Of Tv Signals (AREA)

Abstract

The invention provides dynamic multi-resolution self-adapting parallel video coding and decoding method and system at a multi-core environment. The dynamic multi-resolution self-adapting parallel video coding and decoding system comprises an initialization module, an input module, a decoding module, an output module and a statistics and state monitoring module. The invention can dynamically regulate the running tasks of cores according to code stream formats and the resolution distribution of the system, thereby obtaining sufficient decoding computing resources for different code stream formats and different resolutions; and in addition, the invention ensures the quality and the property of the system.

Description

The parallel decoding method and system of dynamic self-adapting under a kind of many nuclear environments
Technical field
The present invention relates to the coding and decoding video field, adaptive parallel method of coding and decoding video dynamic multiple resolution ratio and system under particularly a kind of many nuclear environments.
Background technology
Video decode application need under the complex environment is handled mass data simultaneously, these data have multiple code stream form, multiple resolution, adopt traditional DSP, FPGA, ASIC hardware cell, risc processor etc. all to be difficult to satisfy performance and the requirement of multi-format decoding function.
Adopt the many-core processor technology of 64 cores, utilize the abundant computational resource of many nuclears, carry out video decode by software and handle, can satisfy and separate the various video form simultaneously, the application demand of multiple code stream resolution.
Summary of the invention
The purpose of this invention is to provide a kind of parallel method and system that controls the multiple nucleus system computational resource allocation, code stream form and resolution according to current decoding, the dynamic computational resource allocation of adjusting, the feasible computational resource that makes full use of system satisfies various different-formats and the resolution demand to the decoding computational resource.
The parallel decoding method of dynamic self-adapting comprises the steps: under a kind of many nuclear environments
A, system initialization are according to system task attribute configuration table default configuration, distributes calculation resources and IO resource;
B, data input module are deposited into raw data in the shared data field of all modules after receiving packet, and each decoder module obtains decoding task to sharing data area;
C, decoder module inquiry system task attribute allocation list if there is one road video to need multinuclear to assist parallel decoding, are then revised system task attribute configuration table, and register module computational resource participates in this road decoding;
D, request if nothing assistance is decoded, then whether inquiry has new task, as if having, obtains new task;
E, decoder module are resolved new task form and resolution, and judging whether needs to assist decoding, if need to assist decoding, then the decoding request is assisted in issue, revises system task attribute configuration table; After new core adds, distribute decoding task to give new core, parallel decoding is finished this road video decode task, after task is finished, gives back the resource of adding, recovers on-the-spot.
A kind of optimal technical scheme of the present invention is: described system task attribute configuration table is done dynamic assignment according to operating code stream form of current system and resolution situation to computational resource.
The parallel decoding system of dynamic self-adapting under a kind of many nuclear environments is characterized in that: comprise system initialization module, load module, decoder module, output module and statistics and state monitoring module.
A kind of optimal technical scheme of the present invention is: each computing unit of described decoder module is analyzed current system task attribute configuration table, decides the decoding of getting new calculation task or certain promoter's of participation big resolution video from task list according to system task attribute configuration table.
Another optimal technical scheme of the present invention is: described statistics and state monitoring module, and generation system task attribute allocation list, this allocation list is dynamically adjusted resources allocation according to operating code stream form of system and resolution situation.
An also optimal technical scheme of the present invention is: described system initialization module is used for distributes calculation resources and IO resource, creates each calculation task and IO task; Described load module receives the data block that needs decoding, generates decoding task, puts into task list; Data behind the output module output decoder.
The operation task that the present invention can be dynamically adjusts each core according to the code stream form in the system and resolution distribution, thus for different code streams form and different resolution can both obtain the computational resource of decoding fully, guarantee the quality and the performance of system.
Description of drawings
Fig. 1 is a dynamic multiple resolution ratio self-adaptive parallel algorithm flow chart.
Fig. 2 is the decode system configuration diagram of dynamic self-adapting
Specific embodiments
Below in conjunction with the drawings and specific embodiments method of the present utility model is described.
During system initialization, according to the default configuration of system task attribute configuration table, distributes calculation resources and IO resource.
Data input module is deposited into raw data in the shared data field of all modules after receiving packet, and each decoder module obtains decoding task to sharing data area.
Decoder module was inquired about current system task attribute configuration table before obtaining new task, seeing if there is certain road video needs multinuclear to assist parallel decoding, if have, revise system task attribute configuration epiphase and close project, the computational resource of register module participates in the parallel decoding on this road.
If can obtain new task, get new task, resolve the form and the resolution of new task, whether decision needs to initiate to assist the decoding request.Other cores are assisted decoding if desired, and can issue new decoding and assist request, revise system task attribute configuration epiphase so and close project, and request is assisted in the issue decoding.Behind other cores registration computational resources, the distribution decoding task is given the core that adds, and parallel decoding is finished the decoding task of this road video.After task is finished, give back the computational resource of adding, recover on-the-spot.

Claims (6)

1. the parallel decoding method of dynamic self-adapting under many nuclear environments is characterized in that: comprise the steps:
A, system initialization are according to system task attribute configuration table default configuration, distributes calculation resources and IO resource;
B, data input module are deposited into raw data in the prosperous altogether data field of all modules after receiving packet, and each decoder module obtains decoding task to sharing data area;
C, decoder module inquiry system task attribute allocation list if there is one road video to need multinuclear to assist parallel decoding, are then revised system task attribute configuration table, and register module computational resource participates in this road decoding;
D, request if nothing assistance is decoded, then whether inquiry has new task, as if having, obtains new task;
E, decoder module are resolved new task form and resolution, and judging whether needs to assist decoding, if need to assist decoding, then the decoding request is assisted in issue, revises system task attribute configuration table; After new core adds, distribute decoding task to give new core, parallel decoding is finished this road video decode task, after task is finished, gives back the resource of adding, recovers on-the-spot.
2. the parallel decoding method of dynamic self-adapting under a kind of according to claim 1 many nuclear environments is characterized in that: described system task attribute configuration table is done dynamic assignment according to operating code stream form of current system and resolution situation to computational resource.
3. the parallel decoding system of dynamic self-adapting under many nuclear environments is characterized in that: comprise system initialization module, load module, decoder module, output module and statistics and state monitoring module.
4. as the parallel decoding system of dynamic self-adapting under a kind of complex environment as described in the claim 3, it is characterized in that: each computing unit of described decoder module is analyzed current system task attribute configuration table, decides the decoding of getting new calculation task or certain promoter's of participation big resolution video from task list according to system task attribute configuration table.
5. as the parallel decoding system of dynamic self-adapting under a kind of many nuclear environments as described in the claim 3, it is characterized in that: described statistics and state monitoring module, generation system task attribute allocation list, this allocation list is dynamically adjusted resources allocation according to operating code stream form of system and resolution situation.
6. as the parallel decoding system of dynamic self-adapting under a kind of many nuclear environments as described in the claim 3, it is characterized in that: described system initialization module is used for distributes calculation resources and IO resource, creates each calculation task and IO task; Described load module receives the data block that needs decoding, generates decoding task, puts into task list; Data behind the output module output decoder.
CN 201010536484 2010-11-04 2010-11-04 Dynamic multi-resolution self-adapting parallel method and system for video encoding and decoding at multi-core environment Active CN102023897B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN 201010536484 CN102023897B (en) 2010-11-04 2010-11-04 Dynamic multi-resolution self-adapting parallel method and system for video encoding and decoding at multi-core environment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN 201010536484 CN102023897B (en) 2010-11-04 2010-11-04 Dynamic multi-resolution self-adapting parallel method and system for video encoding and decoding at multi-core environment

Publications (2)

Publication Number Publication Date
CN102023897A true CN102023897A (en) 2011-04-20
CN102023897B CN102023897B (en) 2013-03-13

Family

ID=43865219

Family Applications (1)

Application Number Title Priority Date Filing Date
CN 201010536484 Active CN102023897B (en) 2010-11-04 2010-11-04 Dynamic multi-resolution self-adapting parallel method and system for video encoding and decoding at multi-core environment

Country Status (1)

Country Link
CN (1) CN102023897B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112203057A (en) * 2020-10-10 2021-01-08 重庆紫光华山智安科技有限公司 Analysis task creating method, device, server and computer-readable storage medium

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101150726A (en) * 2007-09-30 2008-03-26 四川长虹电器股份有限公司 Decoding task allocation method for dual core video decoder
CN101261591A (en) * 2008-04-28 2008-09-10 艾诺通信系统(苏州)有限责任公司 Multi- nuclear DSP system self-adapting task scheduling method
US20090083516A1 (en) * 2006-03-28 2009-03-26 Radisys Canada, Inc. Multimedia processing in parallel multi-core computation architectures
CN101561923A (en) * 2008-04-16 2009-10-21 上海锐动信息技术有限公司 Coordinative render control system and control method thereof

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090083516A1 (en) * 2006-03-28 2009-03-26 Radisys Canada, Inc. Multimedia processing in parallel multi-core computation architectures
CN101150726A (en) * 2007-09-30 2008-03-26 四川长虹电器股份有限公司 Decoding task allocation method for dual core video decoder
CN101561923A (en) * 2008-04-16 2009-10-21 上海锐动信息技术有限公司 Coordinative render control system and control method thereof
CN101261591A (en) * 2008-04-28 2008-09-10 艾诺通信系统(苏州)有限责任公司 Multi- nuclear DSP system self-adapting task scheduling method

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112203057A (en) * 2020-10-10 2021-01-08 重庆紫光华山智安科技有限公司 Analysis task creating method, device, server and computer-readable storage medium
CN112203057B (en) * 2020-10-10 2022-06-03 重庆紫光华山智安科技有限公司 Analysis task creating method, device, server and computer-readable storage medium

Also Published As

Publication number Publication date
CN102023897B (en) 2013-03-13

Similar Documents

Publication Publication Date Title
CN109656703B (en) Method for assisting vehicle task unloading through mobile edge calculation
CN109508432B (en) High-performance FAAS system based on block chain
Li et al. Adaptive resource allocation based on the billing granularity in edge-cloud architecture
CN102625108B (en) Multi-core-processor-based H.264 decoding method
CN110198339B (en) QoE (quality of experience) perception-based edge computing task scheduling method
CN103152613A (en) Distributed code conversion task scheduling system and method based on video image group segmentation
CN102802024A (en) Transcoding method and transcoding system realized in server
CN102143386A (en) Streaming media server acceleration method based on graphics processing unit
CN110149377A (en) A kind of video service node resource allocation methods, system, device and storage medium
CN102609307A (en) Multi-core multi-thread dual-operating system network equipment and control method thereof
CN103338387A (en) Data packet scheduling method for performing combined optimization on energy consumption and video quality under cloud computation
CN102111237A (en) Cloud transcoding system and cloud transcoding method thereof
CN108920273B (en) Computing unloading excitation method for edge cloud
CN102023897B (en) Dynamic multi-resolution self-adapting parallel method and system for video encoding and decoding at multi-core environment
CN104980764A (en) Parallel coding/decoding method, device and system based on complexity balance
CN102073548A (en) Method for executing task, and system thereof
CN105323593A (en) Multimedia transcoding scheduling method and multimedia transcoding scheduling device
CN101256668B (en) Method for implementing video filtering to working balanced multiple nucleus
CN202309988U (en) Dynamic self-adaptive parallel decoding system under multi-core environment
CN112835691A (en) Edge data processing method and system for communication of Internet of things
CN102025990B (en) Video coding and decoding dynamic multiresolution self-adaption paralleling method under multicore environment
Deng et al. Cloud computing based high-performance platform in enabling scalable services in power system
CN105338371A (en) Multimedia transcoding scheduling method and apparatus
CN101616319B (en) Method for processing data and service equipment
CN110908800B (en) Scheduling optimization method based on scheduling entropy in edge cloud collaborative environment

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C53 Correction of patent of invention or patent application
CB03 Change of inventor or designer information

Inventor after: Liu Li

Inventor after: He Dingding

Inventor after: Lan Facai

Inventor after: Xu Jianwei

Inventor after: Cai Huayuan

Inventor after: Wan Wei

Inventor after: Liu Xingbin

Inventor after: Wang Po

Inventor after: Shao Zongyou

Inventor before: Liu Li

Inventor before: He Dingding

Inventor before: Lan Facai

Inventor before: Xu Jianwei

Inventor before: Cai Huayuan

Inventor before: Wan Wei

Inventor before: Liu Xingbin

Inventor before: Wang Po

COR Change of bibliographic data

Free format text: CORRECT: INVENTOR; FROM: LIU LI HE DINGDING LAN FACAI XU JIANWEI CAI HUAYUAN WAN WEI LIU XINGBIN WANG PO TO: LIU LI HE DINGDING LAN FACAI XU JIANWEI CAI HUAYUAN WAN WEI LIU XINGBIN WANG PO SHAO ZONGYOU

C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
TR01 Transfer of patent right

Effective date of registration: 20220727

Address after: 100193 No. 36 Building, No. 8 Hospital, Wangxi Road, Haidian District, Beijing

Patentee after: Dawning Information Industry (Beijing) Co.,Ltd.

Patentee after: DAWNING INFORMATION INDUSTRY Co.,Ltd.

Address before: 100084 Beijing Haidian District City Mill Street No. 64

Patentee before: Dawning Information Industry (Beijing) Co.,Ltd.

TR01 Transfer of patent right