CN102662740A - Asymmetric multi-core system and realization method thereof - Google Patents
Asymmetric multi-core system and realization method thereof Download PDFInfo
- Publication number
- CN102662740A CN102662740A CN2012100879306A CN201210087930A CN102662740A CN 102662740 A CN102662740 A CN 102662740A CN 2012100879306 A CN2012100879306 A CN 2012100879306A CN 201210087930 A CN201210087930 A CN 201210087930A CN 102662740 A CN102662740 A CN 102662740A
- Authority
- CN
- China
- Prior art keywords
- nuclear
- task
- professional
- business
- control
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Landscapes
- Stored Programmes (AREA)
- Multi Processors (AREA)
Abstract
The invention relates to an information technology and provides a data central management system and a data central management method in order to solve the problem of high complexity of business programs in the existing AMP (asymmetric multi-core processing) system. A technical scheme can be summarized as follows: the invention provides an asymmetric multi-core system comprising a multi-core processor, and the multi-core processor is characterized by comprising at least one control core, wherein all cores of the multi-core processor, except the control core, are business cores; the control core corresponds to one task queue in which a plurality of tasks can be placed so as to run a mirror of an operation system; the operation system is used for scheduling the tasks in the task queue; and each business core corresponds to one task queue in which only one task is placed so as to run an exclusive task and the exclusive task monopolizes all resources of the business core and does not participate in task scheduling of the operation system. The method has the benefits of effectively improving the stability and being applicable to the asymmetric multi-core system.
Description
Technical field
The present invention relates to infotech, particularly be applicable to the technology of the asymmetric multiple nucleus system of the communications field.
Background technology
Along with the raising of communication system to performance requirement, the monokaryon system can not satisfy performance requirement gradually, has introduced multiple nucleus system thus.In the application of multiple nucleus system, use each nuclear to handle identical task usually, this system becomes symmetrical multiple core processing system (symmetric multiprocessing abbreviates SMP as).Ripe smp system has the characteristics of each task equity dispatching, each nuclear load balancing, and it is widely used in field of personal computers.But in the communications field, smp system is not a best solution.Reason is that communication system requires to have the higher data transmitting ability, just needs also to guarantee that the data forwarding inter-related task can fast as far as possible execution, and the execution time is also many as far as possible.This demand of communication system and smp system are afoul to the strategy of task equity dispatching, and therefore needing more, optimisation strategy reaches best performance.
In various multiple core processing systems, asymmetric multiple nucleus system is to satisfy a kind of good solution that communication system requires.Asymmetric multiple nucleus system (Asymmetric MultiProcessing abbreviates AMP as).The AMP system divides into groups each nuclear, and every group of nuclear task for processing has nothing in common with each other, and between different nuclear groups, task does not require equity dispatching, does not require load balancing yet.The AMP system has bigger flexibility, can carry out different customizations based on application scenarios.Specific to communication system, Task Distribution that can data forwarding is relevant goes up to one group of specific nuclear and carries out, and has so just guaranteed the exclusivity of these tasks to nuclear, effectively raises data forwarding performance.
The implementation of existing AMP system is as shown in Figure 1.The software section of AMP system is made up of operating system mirror image and professional mirror image two parts.The operation on control nuclear of operating system mirror image is responsible for each professional nuclear is carried out the most basic initialization, and is responsible for that professional mirror image is loaded into corresponding business nuclear and goes up operation.Under this application characteristic, because business procedure compiles mirror image separately, and separate with the operating system mirror image, this just causes physics nuclear directly to be exposed to business procedure.Business procedure needs bare nucleus initialization, memory management, process communication of dealing with complicated etc. to have increased the complexity of business procedure by the operating system task for processing under this environment.
Summary of the invention
The object of the invention is exactly to overcome the higher shortcoming of business procedure complexity in the present AMP system, and a kind of asymmetric multiple nucleus system and its implementation are provided.
The present invention solve the technical problem, and the technical scheme of employing is that a kind of asymmetric multiple nucleus system is provided; Comprise polycaryon processor, it is characterized in that said polycaryon processor comprises at least one control nuclear; Be all professional examining except that remaining nuclear of control nuclear in the polycaryon processor
Said control nuclear, the corresponding task queue that can place a plurality of tasks is used for the operation system image, by operating system the task in the said task queue is dispatched;
Said professional nuclear, corresponding task queue of only placing a task is used for operation and monopolizes task, and the said task of monopolizing is monopolized all resources of this business nuclear, the not task scheduling of Attended Operation system.
Further, close all interruptions on the said professional nuclear, executive system is not called, and does not move any timer.
Concrete, the task of monopolizing that said professional nuclear is gone up operation is an Agent, said Agent is that each professional nuclear is bound the different service program.
Concrete, the task that said professional nuclear is gone up operation is concrete business procedure.
The implementation method of asymmetric multiple nucleus system is characterized in that, may further comprise the steps:
Step 1. system powers on, and control nuclear is as main nuclear operation and beginning initialization, and professional nuclear is in halted state;
The professional nuclear of startup was accomplished hardware and the relevant initialization of operating system after step 2. control nuclear was accomplished necessary initialization;
Step 3. system is tied to corresponding service nuclear with task and goes up independent operating.
Concrete, step 1 may further comprise the steps: system powers on, and control nuclear moves and jumps to the setup code of specific address executive operating system, accomplishes the initialization that comprises necessity of internal storage access and interruption on the control nuclear.
Concrete, step 2 may further comprise the steps:
After step 21. control nuclear is accomplished necessary initialization, for necessary preliminary work is done in the operation of business nuclear;
All business nuclear was interrupted in one of transmission after step 22. control nuclear had been carried out necessary preliminary work, accomplished the startup of professional nuclear;
Jump to control nuclear after the professional nuclear of step 23. starts and begin to carry out specific program, at first accomplish further initialization action in this specific program, operation Agent after the initialization of professional nuclear is accomplished business nuclear for the address of its setting.
Further, in the step 21, the preliminary work of said necessity comprises the necessary mapping relations of setting up internal storage access, Interrupt Process is set, the program entry address and the task stack address of professional nuclear operation are set.
Concrete, in the step 23, the setup code of said further initialization action adopt operating system be SMP provide from the nuclear setup code, close all interruptions on the professional nuclear, responding system does not call, and does not open any timer.
Further again, may further comprise the steps in the step 3:
Step 31. user uses operating system task to be placed in the task queue of control nuclear; The task queue of said control nuclear is a chained list; The corresponding professional nuclear of each node has comprised the entry address of the nuclear number and the business procedure of corresponding professional nuclear at least on this node;
Agent traversal task queue on the professional nuclear of step 32. number is selected concrete task to be placed in the task queue of this business nuclear through coupling nuclear and is moved.
The invention has the beneficial effects as follows,, business procedure has been shielded physics nuclear through above-mentioned asymmetric multiple nucleus system and its implementation; Peel off the complicated operations system works, reduced the complexity of business procedure, can effectively improve stability; And avoided the interruption on the professional nuclear; Improve handling capacity, used Agent, increased the flexibility ratio that business nuclear is used.
Description of drawings
Fig. 1 is traditional AMP system model synoptic diagram.
Fig. 2 is the asymmetric multiple nucleus system model of a present invention synoptic diagram.
Fig. 3 is the asymmetric multiple nucleus system initialization flowchart of the embodiment of the invention.
Fig. 4 is a business-binding synoptic diagram in the asymmetric multiple nucleus system of the embodiment of the invention.
Embodiment
Below in conjunction with embodiment and accompanying drawing, describe technical scheme of the present invention in detail.
The asymmetric multiple nucleus system model of the present invention synoptic diagram is referring to Fig. 2.Asymmetric multiple nucleus system of the present invention comprises polycaryon processor, and said polycaryon processor comprises at least one control nuclear; All the other are all professional nuclear except that control nuclear in the polycaryon processor; Control nuclear is used for the operation system, and professional nuclear is used to move concrete task, in this asymmetric multiple nucleus system; Business procedure does not compile mirror image separately, is tied to specific business nuclear as a task of operating system and goes up operation.In the implementation method of the asymmetric multiple nucleus system of the present invention, at first system powers on, operation of control nuclear and beginning initialization; Professional nuclear is in halted state; After control nuclear is accomplished necessary initialization, start professional nuclear then and accomplish hardware and the relevant initialization of operating system, final system is tied to corresponding service nuclear with task and goes up independent operating, in this asymmetric multiple nucleus system; Operating system shields physics nuclear; Business procedure need not realized the complicated operations system logic, has effectively reduced complexity, has increased reliability.
Embodiment
Asymmetric multiple nucleus system model synoptic diagram is referring to Fig. 2 in the embodiment of the invention, and the asymmetric multiple nucleus system initialization flowchart of the embodiment of the invention is referring to Fig. 3, and the business-binding synoptic diagram is referring to Fig. 4 in the asymmetric multiple nucleus system of the embodiment of the invention.The asymmetric multiple nucleus system of present embodiment; Comprise polycaryon processor; Said polycaryon processor comprises at least one control nuclear; All the other are all professional nuclear except that control nuclear in the polycaryon processor; Control nuclear is used for the operation system, and professional nuclear is used to move concrete task, all has a task queue in control nuclear and the professional nuclear; And the task queue of control nuclear can be placed a plurality of tasks, is dispatched by operating system; Only place a task in the task queue of professional nuclear, after task brings into operation, monopolize all resources of this business nuclear, the task scheduling of Attended Operation system not, the task that professional here nuclear is gone up operation can be concrete business procedure or Agent; And on business nuclear, close all interruptions, executive system is not called, and does not move any timer.
The implementation method of the asymmetric multiple nucleus system of present embodiment may further comprise the steps:
A. system powers on and initialization, and the starting point of system initialization powers on from polycaryon processor and begins to accomplish up to system start-up, i.e. the os-task that brings into operation is examined in control, the professional service agent program that brings into operation of examining.
Polycaryon processor can start a nuclear operation earlier usually when powering on, other nuclear is in halted state.The nuclear of operation is called main nuclear at first, i.e. control in present embodiment nuclear, other nuclear are called from nuclear, i.e. the nuclear of business in the present embodiment.When powering on, main nuclear jumps to the setup code of specific address executive operating system, accomplishes main nuclear and goes up necessary initialization, comprises internal storage access, initialization such as interruption.After main nuclear is accomplished necessary initialization, just begin mainly to refer to set up the necessary mapping relations of internal storage access for doing necessary preliminary work from the nuclear operation, Interrupt Process is set, be provided with from the program entry address of nuclear operation, task stack address etc.When for after sufficient preparation has been carried out in nuclear operation, main authorizing send one to interrupt allly from nuclear, accomplishes the startup from nuclear.
Begin to carry out specific program from jumping to main nuclear after nuclear starts for the address of its setting.The specific program of this section is at first accomplished the further initialization action from nuclear.This part setup code can adopt operating system be SMP provide from the nuclear setup code, also can use user's oneself setup code.Having adopted in operating system in an embodiment of the present invention is the general mode of transforming from the basis of nuclear setup code that SMP provides, and promptly closes from all interruptions of examining, and responding system does not call, and does not open any timer.
From examining the Agent that just brings into operation after initialization is accomplished, so far this AMP system has accomplished initialization procedure.
B. business-binding is to specific business nuclear.In operational process, above specific business nuclear, detailed process is as shown in Figure 4 with business-binding for the operable specific interface of user.
The user can be placed into business procedure n (also can be Agent n) in the business procedure formation (i.e. the task queue of control nuclear) through the interface that provides; The business procedure formation can be a chained list; The corresponding professional nuclear of each node; The nuclear number that has comprised professional nuclear on this node, necessary informations such as the entry address of business procedure.Agent traversal business procedure formation on the professional nuclear number is selected concrete business to be placed in oneself the task queue through coupling nuclear and is moved.Because closed all interruptions on the professional nuclear, executive system is not called, the task scheduling of Attended Operation system not, the business procedure resources that all business are examined of just can perfecting monopoly have effectively guaranteed service throughput like this.
Though this routine Fig. 3 has only described the model of four karyonides systems, the implementation method of this asymmetric multiple nucleus system is applicable to the multiple nucleus system of other number.This asymmetric multiple nucleus system divides into groups to nuclear, and the quantity of control nuclear also not only is confined to 1, also can form smp system performance control function by a plurality of control nuclears.
Above-described specific embodiment; To the object of the invention; Technical scheme and beneficial effect have carried out further explain; Institute it should be noted that the above is merely specific embodiment of the present invention, and those skilled in the art can carry out various changes and modification and not break away from the spirit and scope of the present invention the present invention.
Claims (10)
1. asymmetric multiple nucleus system comprises polycaryon processor, it is characterized in that, said polycaryon processor comprises at least one control nuclear, is all professional examining except that remaining nuclear of control nuclear in the polycaryon processor,
Said control nuclear, the corresponding task queue that can place a plurality of tasks is used for the operation system image, by operating system the task in the said task queue is dispatched;
Said professional nuclear, corresponding task queue of only placing a task is used for operation and monopolizes task, and the said task of monopolizing is monopolized all resources of this business nuclear, the not task scheduling of Attended Operation system.
2. asymmetric multiple nucleus system as claimed in claim 1 is characterized in that, closes all interruptions on the said professional nuclear, and executive system is not called, and does not move any timer.
3. asymmetric according to claim 1 or claim 2 multiple nucleus system is characterized in that, the task of monopolizing that said professional nuclear is gone up operation is an Agent, and said Agent is that each professional nuclear is bound the different service program.
4. according to claim 1 or claim 2 asymmetric multiple nucleus system is characterized in that, the task that said professional nuclear is gone up operation is concrete business procedure.
5. the implementation method of asymmetric multiple nucleus system is characterized in that, may further comprise the steps:
Step 1. system powers on, and control nuclear is as main nuclear operation and beginning initialization, and professional nuclear is in halted state;
The professional nuclear of startup was accomplished hardware and the relevant initialization of operating system after step 2. control nuclear was accomplished necessary initialization;
Step 3. system is tied to corresponding service nuclear with task and goes up independent operating.
6. the implementation method of asymmetric multiple nucleus system as claimed in claim 5 is characterized in that, step 1 comprises:
System powers on, and control nuclear moves and jump to the setup code of specific address executive operating system, accomplishes the initialization that comprises necessity of internal storage access and interruption on the control nuclear.
7. the implementation method of asymmetric multiple nucleus system as claimed in claim 5 is characterized in that, step 2 may further comprise the steps:
After step 21. control nuclear is accomplished necessary initialization, for necessary preliminary work is done in the operation of business nuclear;
All business nuclear was interrupted in one of transmission after step 22. control nuclear had been carried out necessary preliminary work, accomplished the startup of professional nuclear;
Jump to control nuclear after the professional nuclear of step 23. starts and begin to carry out specific program, at first accomplish further initialization action in this specific program, operation Agent after the initialization of professional nuclear is accomplished business nuclear for the address of its setting.
8. the implementation method of asymmetric multiple nucleus system as claimed in claim 7; It is characterized in that; In the step 21, the preliminary work of said necessity comprises the necessary mapping relations of setting up internal storage access, Interrupt Process is set, the program entry address and the task stack address of professional nuclear operation are set.
9. the implementation method of asymmetric multiple nucleus system as claimed in claim 7; It is characterized in that; In the step 23, the setup code of said further initialization action adopt operating system be SMP provide from the nuclear setup code, close all interruptions on the professional nuclear; Responding system does not call, and does not open any timer.
10. want the implementation method of each described asymmetric multiple nucleus system of 6-10 like right, it is characterized in that, may further comprise the steps in the step 3:
Step 31. user uses operating system task to be placed in the task queue of control nuclear; The task queue of said control nuclear is a chained list; The corresponding professional nuclear of each node has comprised the entry address of the nuclear number and the business procedure of corresponding professional nuclear at least on this node;
Agent traversal task queue on the professional nuclear of step 32. number is selected concrete task to be placed in the task queue of this business nuclear through coupling nuclear and is moved.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210087930.6A CN102662740B (en) | 2012-03-29 | 2012-03-29 | Asymmetric multi-core system and realization method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210087930.6A CN102662740B (en) | 2012-03-29 | 2012-03-29 | Asymmetric multi-core system and realization method thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102662740A true CN102662740A (en) | 2012-09-12 |
| CN102662740B CN102662740B (en) | 2014-12-10 |
Family
ID=46772239
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210087930.6A Active CN102662740B (en) | 2012-03-29 | 2012-03-29 | Asymmetric multi-core system and realization method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102662740B (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103019856A (en) * | 2012-11-23 | 2013-04-03 | 上海寰创通信科技股份有限公司 | Asymmetric dispatch method for multi-core processor |
| CN105718320A (en) * | 2016-01-18 | 2016-06-29 | 华为技术有限公司 | Clock task processing method, device and facility |
| CN106254578A (en) * | 2016-09-28 | 2016-12-21 | 深圳震有科技股份有限公司 | Ethernet message transmission method and system under double-core AMP system framework |
| CN107357666A (en) * | 2017-06-26 | 2017-11-17 | 西安微电子技术研究所 | A kind of multi-core parallel concurrent system processing method based on hardware protection |
| CN108259369A (en) * | 2018-01-26 | 2018-07-06 | 迈普通信技术股份有限公司 | The retransmission method and device of a kind of data message |
| WO2020063040A1 (en) * | 2018-09-26 | 2020-04-02 | Oppo广东移动通信有限公司 | Processor core configuration method and apparatus, terminal, and storage medium |
| CN111078301A (en) * | 2018-10-22 | 2020-04-28 | 致茂电子(苏州)有限公司 | Multi-core arithmetic device and operation method thereof |
| WO2021022964A1 (en) * | 2019-08-02 | 2021-02-11 | 中兴通讯股份有限公司 | Task processing method, device, and computer-readable storage medium based on multi-core system |
| CN112395056A (en) * | 2020-10-21 | 2021-02-23 | 国电南瑞科技股份有限公司 | Embedded asymmetric real-time system and electric power secondary equipment |
| CN112527404A (en) * | 2020-11-11 | 2021-03-19 | 贵州电网有限责任公司 | Configuration method, device, equipment and medium for chip relay protection general system |
| CN114090498A (en) * | 2021-11-23 | 2022-02-25 | 云南电网有限责任公司电力科学研究院 | Method for realizing mutual decoupling of multi-core starting and service of embedded SOC (system on chip) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1910553A (en) * | 2004-01-08 | 2007-02-07 | 皇家飞利浦电子股份有限公司 | Method and apparatus for scheduling task in multi-processor system based on memory requirements |
| CN101013388A (en) * | 2007-01-26 | 2007-08-08 | 浙江大学 | Heterogeneous multi-core system-oriented process scheduling method |
| 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 |
| CN101788920A (en) * | 2010-02-05 | 2010-07-28 | 北京航空航天大学 | CPU virtualization method based on processor partitioning technology |
-
2012
- 2012-03-29 CN CN201210087930.6A patent/CN102662740B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1910553A (en) * | 2004-01-08 | 2007-02-07 | 皇家飞利浦电子股份有限公司 | Method and apparatus for scheduling task in multi-processor system based on memory requirements |
| CN101013388A (en) * | 2007-01-26 | 2007-08-08 | 浙江大学 | Heterogeneous multi-core system-oriented process scheduling method |
| 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 |
| CN101788920A (en) * | 2010-02-05 | 2010-07-28 | 北京航空航天大学 | CPU virtualization method based on processor partitioning technology |
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103019856A (en) * | 2012-11-23 | 2013-04-03 | 上海寰创通信科技股份有限公司 | Asymmetric dispatch method for multi-core processor |
| CN105718320B (en) * | 2016-01-18 | 2020-11-06 | 华为技术有限公司 | Clock task processing method, device and equipment |
| CN105718320A (en) * | 2016-01-18 | 2016-06-29 | 华为技术有限公司 | Clock task processing method, device and facility |
| WO2017124717A1 (en) * | 2016-01-18 | 2017-07-27 | 华为技术有限公司 | Clock task processing method, apparatus and device |
| US10891171B2 (en) | 2016-01-18 | 2021-01-12 | Huawei Technologies Co., Ltd. | Method, apparatus and device for transitioning between data and control core and migrating clock task from data core to control core |
| CN106254578A (en) * | 2016-09-28 | 2016-12-21 | 深圳震有科技股份有限公司 | Ethernet message transmission method and system under double-core AMP system framework |
| CN106254578B (en) * | 2016-09-28 | 2019-07-12 | 深圳震有科技股份有限公司 | Ethernet message transmission method and system under double-core AMP system framework |
| CN107357666A (en) * | 2017-06-26 | 2017-11-17 | 西安微电子技术研究所 | A kind of multi-core parallel concurrent system processing method based on hardware protection |
| CN108259369A (en) * | 2018-01-26 | 2018-07-06 | 迈普通信技术股份有限公司 | The retransmission method and device of a kind of data message |
| CN110955499A (en) * | 2018-09-26 | 2020-04-03 | Oppo广东移动通信有限公司 | Processor core configuration method, device, terminal and storage medium |
| WO2020063040A1 (en) * | 2018-09-26 | 2020-04-02 | Oppo广东移动通信有限公司 | Processor core configuration method and apparatus, terminal, and storage medium |
| CN111078301A (en) * | 2018-10-22 | 2020-04-28 | 致茂电子(苏州)有限公司 | Multi-core arithmetic device and operation method thereof |
| CN111078301B (en) * | 2018-10-22 | 2023-11-21 | 致茂电子(苏州)有限公司 | Multi-core computing device and operating method thereof |
| WO2021022964A1 (en) * | 2019-08-02 | 2021-02-11 | 中兴通讯股份有限公司 | Task processing method, device, and computer-readable storage medium based on multi-core system |
| CN112395056A (en) * | 2020-10-21 | 2021-02-23 | 国电南瑞科技股份有限公司 | Embedded asymmetric real-time system and electric power secondary equipment |
| CN112395056B (en) * | 2020-10-21 | 2022-08-26 | 国电南瑞科技股份有限公司 | Embedded asymmetric real-time system and electric power secondary equipment |
| CN112527404A (en) * | 2020-11-11 | 2021-03-19 | 贵州电网有限责任公司 | Configuration method, device, equipment and medium for chip relay protection general system |
| CN112527404B (en) * | 2020-11-11 | 2021-12-17 | 贵州电网有限责任公司 | Configuration method, device, equipment and medium for chip relay protection general system |
| CN114090498A (en) * | 2021-11-23 | 2022-02-25 | 云南电网有限责任公司电力科学研究院 | Method for realizing mutual decoupling of multi-core starting and service of embedded SOC (system on chip) |
| CN114090498B (en) * | 2021-11-23 | 2023-07-28 | 云南电网有限责任公司电力科学研究院 | Method for realizing multi-core starting and service mutual decoupling of embedded SOC (system on chip) system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102662740B (en) | 2014-12-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102662740B (en) | Asymmetric multi-core system and realization method thereof | |
| EP3425502B1 (en) | Task scheduling method and device | |
| US11036547B2 (en) | File operation task optimization | |
| Calheiros et al. | Energy-efficient scheduling of urgent bag-of-tasks applications in clouds through DVFS | |
| RU2454704C2 (en) | Method and system for executing program applications and machine-readable medium | |
| US9772879B2 (en) | System and method for isolating I/O execution via compiler and OS support | |
| CN113641457B (en) | Container creation method, device, apparatus, medium, and program product | |
| US8413158B2 (en) | Processor thread load balancing manager | |
| US9973512B2 (en) | Determining variable wait time in an asynchronous call-back system based on calculated average sub-queue wait time | |
| US8849888B2 (en) | Candidate set solver with user advice | |
| CN103793255B (en) | Starting method for configurable multi-main-mode multi-OS-inner-core real-time operating system structure | |
| WO2015101091A1 (en) | Distributed resource scheduling method and device | |
| US8924754B2 (en) | Quality of service targets in multicore processors | |
| US20240160474A1 (en) | Multi-core processor task scheduling method, and device and storage medium | |
| CN103902355A (en) | Quick medical image loading method | |
| WO2023174037A1 (en) | Resource scheduling method, apparatus and system, device, medium, and program product | |
| CN104978174A (en) | Method and system for switching network cards in plurality of operating systems | |
| CN109002286A (en) | Data asynchronous processing method and device based on synchronous programming | |
| US10248459B2 (en) | Operating system support for game mode | |
| US20180329758A1 (en) | Method and device for transmitting managing instruction and managing automatic scaling function of vnf | |
| Henrio et al. | Declarative scheduling for active objects | |
| CN113391821A (en) | Asymmetric multiprocessor embedded operating system | |
| CN102681881B (en) | Across machine dispatching method and system thereof | |
| CN108694571A (en) | A kind of method for processing business and device based on multi-process | |
| Huang et al. | A novel fined-grained GPU sharing mechanism for artificial intelligence computing-resource automatic scheduling |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |