CN101984414B - Method and device for scheduling central processing unit (CPU) resource - Google Patents
Method and device for scheduling central processing unit (CPU) resource Download PDFInfo
- Publication number
- CN101984414B CN101984414B CN 201010534278 CN201010534278A CN101984414B CN 101984414 B CN101984414 B CN 101984414B CN 201010534278 CN201010534278 CN 201010534278 CN 201010534278 A CN201010534278 A CN 201010534278A CN 101984414 B CN101984414 B CN 101984414B
- Authority
- CN
- China
- Prior art keywords
- task
- task groups
- cpu resource
- groups
- takies
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Debugging And Monitoring (AREA)
Abstract
The invention discloses a method and device for scheduling central processing unit (CPU) resource. The method comprises the following steps: dividing tasks in a system into task groups; carrying out statistics on the occupied situation of CPU resource currently occupied by the task groups; and when the occupied amount of the CUP resource currently occupied by the task group excesses a preset maximum value, releasing CPU resource of one or more tasks to ready tasks in other task groups except the task group to be operated. According to the invention, the stability of the system is strengthened and the performance of the system is improved.
Description
Technical field
The present invention relates to the computer operating system field, relate in particular to a kind of dispatching method and device of cpu resource.
Background technology
In computer operating system, cpu resource is shared by all tasks in the system as the core resource of total system, distributes cpu resource in order to give numerous tasks, has produced the dispatching method of a variety of cpu resources thus.In the common operating system, in order to satisfy the required concurrent and real-time requirement of system, between the task of different priorities, adopted the dispatching method seized based on priority, this dispatching method provides the assurance of cpu resource scheduling for the highest task of priority, as long as the highest Mission Ready of priority, it can seize the low priority task of present operation immediately, and takies cpu resource always and initiatively abandon use to CPU up to it.If the task long-play of high priority, the task of low priority can't get access to cpu resource and be died of hunger.
Single system for the task negligible amounts, can be by the accurate priority of dividing task function, adjusting task, guarantee that real-time task can not take cpu resource for a long time, and the low priority task that may take cpu resource for a long time is set is identical priority, and adopts the dispatching method of round-robin to avoid these tasks hungry to death.
But the inventor finds, for comprising tens even a hundreds of task, have the big system of hundreds of priority, above-mentioned method is difficult to accomplish, especially the software systems that have a lot of subsystems, these software subsystems are finished by different development teams, address this problem unusual difficult.Even if solved the problem of existing system by above-mentioned method, be accompanied by the continuous introducing of new demand, new subsystem is added in the system, and problem will occur again.
For example, in a communication system devices software, there are three software subsystems: Operation, Administration and Maintenance (Operation Administration and Maintenance abbreviates OAM as) subsystem, broadband subsystem and narrowband subsystem.These three subsystems itself are all finished specific business function by several task cooperations, the dispatching method based on preemptive priority according to routine, wherein the task of any one high priority takies cpu resource for a long time, all can cause other subsystem can't obtain cpu resource.Such as, broadband subsystem is rejected service (Denial of Service abbreviates DoS as) and attacks, and finally causes the OAM module can't obtain scheduling, can't respond the request of user's Operation and Maintenance, makes the user lose the control to equipment.
Summary of the invention
Fundamental purpose of the present invention is to provide a kind of scheduling scheme of cpu resource, because the task of high priority takies cpu resource for a long time, the task of causing low priority can't be obtained cpu resource and the problem died of hunger to solve at least in the above-mentioned correlation technique.
To achieve these goals, according to an aspect of the present invention, provide a kind of dispatching method of cpu resource.
Dispatching method according to cpu resource of the present invention may further comprise the steps: the task division in the system is become task groups; The current situation that takies to cpu resource of statistics task group; And in that task groups is current when taking of cpu resource surpassed default maximal value, the cpu resource that discharges the one or more tasks in the task groups is given the ready task run in other task groups except this task groups.
Further, before the current situation that takies to cpu resource of statistics task group, this method also comprises: set in advance the maximal value that task groups takies cpu resource, wherein, maximal value is the summation that all tasks take cpu resource in the task groups.
Further, setting in advance task groups comprises the maximal value that cpu resource takies: be the largest percentage that task groups distributes it that cpu resource is taken at each scheduled time slot.
Further, the current situation that takies to cpu resource of statistics task group comprises: the statistics task group in each scheduled time slot to the occupancy of cpu resource.
Further, the time precision of each predetermined amount of time is microsecond.
Further, the cpu resource that discharges the one or more tasks in the task groups comprises for the ready task run in other task groups except this task groups: if equal offhand ready task in other task groups, then the task in the task groups holds over cpu resource and moves; If in other task groups ready task is arranged, then discharge the cpu resource of the one or more tasks in the task groups to the ready task run in other task groups.
To achieve these goals, according to another aspect of the present invention, also provide a kind of dispatching device of cpu resource.
Dispatching device according to cpu resource of the present invention comprises: divide module, be used for the task division of system is become task groups; Statistical module is used for the current situation that takies to cpu resource of statistics task group; And scheduler module, being used in that task groups is current when taking of cpu resource surpassed default maximal value, the cpu resource that discharges the one or more tasks in the task groups is given the ready task run in other task groups except this task groups.
Further, this device also comprises: module is set, is used for setting in advance the maximal value that task groups takies cpu resource, wherein, maximal value is the summation that all tasks take cpu resource in the task groups.
Further, arranging that module also is used at each scheduled time slot is the largest percentage that task groups distributes it that cpu resource is taken; And statistical module also is used for statistics task group occupancy to cpu resource in each scheduled time slot.
Further, scheduler module comprises: holding unit is used for all under the situation of offhand ready task, keeping the task in the task groups to hold over the cpu resource operation in other task groups; And releasing unit, be used for having under the situation of ready task in other task groups, discharge the cpu resource of the one or more tasks in the task groups to the ready task run in other task groups.
By the present invention, employing discharges the mode that the shared cpu resource of wherein one or more tasks is given ready task in other task groups when task groups surpasses threshold value to taking of cpu resource, solved in the correlation technique because the task of high priority takies cpu resource for a long time, the task of causing low priority can't be obtained cpu resource and the problem died of hunger, strengthen the stability of system, improved the performance of system.
Description of drawings
Accompanying drawing described herein is used to provide further understanding of the present invention, constitutes the application's a part, and illustrative examples of the present invention and explanation thereof are used for explaining the present invention, do not constitute improper restriction of the present invention.In the accompanying drawings:
Fig. 1 is the process flow diagram according to the dispatching method of the cpu resource of the embodiment of the invention;
Fig. 2 is the structured flowchart according to the dispatching device of the cpu resource of the embodiment of the invention;
Fig. 3 is the structured flowchart of the dispatching device of cpu resource according to the preferred embodiment of the invention;
Fig. 4 is the synoptic diagram of dividing according to the task groups of the preferred embodiment of the present invention two;
Fig. 5 is the synoptic diagram according to the system operation time of the preferred embodiment of the present invention two;
Fig. 6 is the process flow diagram that takies the maximum duration of CPU according to the allocating task group of the preferred embodiment of the present invention two.
Embodiment
Hereinafter will describe the present invention with reference to the accompanying drawings and in conjunction with the embodiments in detail.Need to prove that under the situation of not conflicting, embodiment and the feature among the embodiment among the application can make up mutually.
According to embodiments of the invention, provide a kind of dispatching method of cpu resource.Fig. 1 is the process flow diagram according to the dispatching method of the cpu resource of the embodiment of the invention, and as shown in Figure 1, this method may further comprise the steps:
Step S102 becomes task groups with the task division in the system;
Step S104, the current situation that takies to cpu resource of statistics task group; And
Step S106, in that task groups is current when taking of cpu resource surpassed default maximal value, the cpu resource that discharges the one or more tasks in the task groups is given the ready task run in other task groups except this task groups.
Pass through above-mentioned steps, employing discharges the mode that the shared cpu resource of wherein one or more tasks is given ready task in other task groups when task groups surpasses threshold value to taking of cpu resource, solved in the correlation technique because the task of high priority takies cpu resource for a long time, the task of causing low priority can't be obtained cpu resource and the problem died of hunger, strengthen the stability of system, improved the performance of system.
Preferably, before step S104, can set in advance the maximal value that task groups takies cpu resource, wherein, maximal value is the summation that all tasks take cpu resource in the task groups.For example, can set in advance the maximal value that each task groups takies cpu resource.This method makes the user to be configured the situation that each task groups takies cpu resource, has strengthened the dirigibility of system, has improved user's experience.
Preferably, setting in advance task groups comprises the maximal value that cpu resource takies: be the largest percentage that task groups distributes it that cpu resource is taken at each scheduled time slot.For example, the largest percentage of distributing it that cpu resource is taken for each task groups at each scheduled time slot.This method can improve validity and the accuracy of system.
Preferably, in step S104, can the statistics task group in each scheduled time slot to the occupancy of cpu resource.This method realizes simple, workable.
Preferably, the time precision of each predetermined amount of time is microsecond.This method is conducive to accurate statistics task to the occupancy of CPU, has improved the precision of system.
Preferably, in step S106, if equal offhand ready task in other task groups, then the task in the task groups holds over the cpu resource operation; If in other task groups ready task is arranged, then discharge the cpu resource of the one or more tasks in the task groups to the ready task run in other task groups.
This preferred embodiment can (that is, CPU usage the task in) the task groups be dispatched, and has improved dirigibility and the processing power of system to surpassing default maximal value according to the current situation of system.
Corresponding to above-mentioned method, also provide a kind of dispatching device of cpu resource in the present embodiment.Fig. 2 is the structured flowchart according to the dispatching device of the cpu resource of the embodiment of the invention, and as shown in Figure 2, this device comprises: divide module 22, statistical module 24, scheduler module 26, below this structure is elaborated.
Divide module 22, be used for the task division of system is become task groups; Statistical module 24 is coupled to and divides module 22, is used for the current situation that takies to cpu resource of statistics task group; And scheduler module 26, be coupled to statistical module 24, be used in that task groups is current when taking of cpu resource surpassed default maximal value, the cpu resource that discharges the one or more tasks in the task groups is given the ready task run in other task groups except this task groups.
Pass through said apparatus, adopt scheduler module 26 when task groups surpasses threshold value to taking of cpu resource, to discharge the mode that the shared cpu resource of wherein one or more tasks is given ready task in other task groups, solved in the correlation technique because the task of high priority takies cpu resource for a long time, the task of causing low priority can't be obtained cpu resource and the problem died of hunger, strengthen the stability of system, improved the performance of system.
Fig. 3 is the structured flowchart of the dispatching device of cpu resource according to the preferred embodiment of the invention, as shown in Figure 3, this device also comprises: module 32 is set, be coupled to and divide module 22 and statistical module 24, be used for setting in advance the maximal value that task groups takies cpu resource, wherein, this maximal value is the summation that all tasks take cpu resource in the task groups.
Preferably, arranging that module 32 also is used at each scheduled time slot is the largest percentage that task groups distributes it that cpu resource is taken; And statistical module 24 also is used for statistics task group occupancy to cpu resource in each scheduled time slot.
For example, module 32 is set and distributes its maximum duration that cpu resource is taken and the threshold value of maximum duration at each scheduled time slot for each task groups.
Preferably, scheduler module 26 comprises: holding unit 262 is used for all under the situation of offhand ready task, keeping the task in the task groups to hold over the cpu resource operation in other task groups; And releasing unit 264, be used for having under the situation of ready task in other task groups, discharge the cpu resource of the one or more tasks in the task groups to the ready task run in other task groups.
Preferably, in releasing unit 264 discharges the cpu resource of the one or more tasks in the task groups to other task groups after the ready task run, statistical module 24 count on this task groups current to cpu resource take less than default maximal value the time, being scheduled away before can also recovering in this task groups of task is to the use of cpu resource.
Below each embodiment combine preferred embodiment above-mentioned.
Embodiment one
Present embodiment is that task division is become task groups, and its maximum cpu resource value of taking is set, when taking of the cpu resource of task groups surpassed the maximal value of setting, just the task scheduling in this task groups is gone out, abandon the use to cpu resource, be dispatched to the task run in other task groups, up to this task groups taking less than the maximal value of setting cpu resource.Wherein, the task groups here can comprise a plurality of tasks, also can only comprise a task.
In specific implementation process, at first need the task division in the system is become several task groups, can arrange according to the needs of system then the maximal value that task groups takies cpu resource (in the task groups all tasks to the cpu resource value of taking and), that is the largest percentage that cpu resource is taken.Because task is dynamic data to taking of cpu resource, the operation that is accompanied by system changes dynamically, has ageingly, takies the situation of CPU so the CPU occupation rate of certain task here all refers to this task in the past period.For example, the CPU occupation rate in the CPU occupation rate in 1 second, 30 seconds, 1 minute CPU occupation rate etc.That is to say that it is the cpu resource occupation rate in the section how long that the maximal value that the cpu resource of the task groups in the present embodiment takies need be specified usually.In addition, no matter be take peaked of all task groups cpu resources and, or individual task group cpu resource take peaked arrange available greater than 100%, also can be less than 100%.The peaked setting that certain individual task group cpu resource takies surpasses 100% without any meaning.
Set after the above condition, in the operating system scheduling task (can be here regularly scheduling, also can be the scheduling that system call causes) the statistics task group at the appointed time in the section to the situation that takies of cpu resource, if task groups has surpassed the maximal value of setting to taking of cpu resource, and other task groups has ready can moving of task, immediately the task scheduling of this task groups is gone out, discharge cpu resource and give ready task run in other task groups.If equal offhand ready task in other all task groups, this task groups holds over CPU operation (that is, task groups can surpass the maximal value of setting to taking of CPU in this case) so.
As seen, present embodiment becomes several task groups to task division all in the system, and the maximal value that need distribute each task groups that cpu resource is taken quantitatively according to system, in real time each task groups being taken the cpu resource situation then monitors, when taking of the cpu resource of task groups surpassed the maximal value of setting, task scheduling in this task groups is gone out, abandon the use to cpu resource, be dispatched to the task run in other task groups, up to this task groups taking less than the maximal value of setting cpu resource, thereby avoided in traditional real time operating system seizing scheduler task according to strict priority and causing low priority task problem hungry to death, improved the stability of system.
Embodiment two
Be example with the object lesson, describe the dispatching method of the cpu resource in the present embodiment in detail, this method can may further comprise the steps:
Step 1 is carried out the division of task groups to the task in the system.Fig. 4 is the synoptic diagram of dividing according to the task groups of the preferred embodiment of the present invention two, supposes that here amounting to 9 tasks by T1~T9 in certain system software constitutes, and is divided into A, B, three task groups of C.Wherein, task groups A comprises three tasks of T1-T3, and task groups B comprises T4 and two tasks of T5, and other task all is divided among the task groups C.
Step 2 arranges the maximal value that each task groups takies cpu resource.As shown in Figure 4, suppose that here the maximal value that the cpu resource of task groups A takies is set to i%, the maximal value that the cpu resource of task groups B takies is set to j%, and the maximal value that the cpu resource of task groups C takies is set to k%.
As seen, present embodiment is the task division to of each a subsystem task groups, and it is set cpu resource is taken maximal value, thereby avoided a subsystem to take cpu resource for a long time and the situation that causes other subsystem normally to move.
Need to prove, because the CPU occupation rate of task refers to that situation that this task in the past period at interval takies CPU (for example, the CPU occupation rate in 1 second, the CPU occupation rate in 30 seconds, 1 minute CPU occupation rate etc.), so, it is cpu resource occupation rate in how long at interval that the maximal value that the cpu resource of task groups takies need be specified, and can this time interval be set to m here.In addition, the CPU occupation rate at interval interior each task groups of accurate Calculation setting-up time in specific implementation process, need have a time source that ratio of precision is higher, can suppose that here time precision is microsecond.
Step 3 calculates the time value (that is duration) that in each time interval m each task groups can take CPU.This time value is exactly that m is multiplied by the maximal value setting that the task groups cpu resource takies, that is, task groups A is exactly that m is multiplied by i%.Here suppose that the duration that task groups A is assigned with is that (precision is equal to above-mentioned time source to I=m*i% in time interval m, here suppose that precision is microsecond), the duration that task groups B is assigned with in time interval m is J=m*j%, and the duration that task groups C is assigned with in time interval m is K=m*k%.
Fig. 5 is the synoptic diagram according to the system operation time of the preferred embodiment of the present invention two, as shown in Figure 5, the start time of system's operation is 0, system can carry out the distribution of working time of task groups cpu resource in each time interval m, the value of distributing is exactly the duration that above-mentioned task groups is assigned with in time interval m, for example, I, J and K.
Step 4 is distributed it for each task groups and is taken the maximum duration of CPU.
Fig. 6 is the process flow diagram that takies the maximum duration of CPU according to the allocating task group of the preferred embodiment of the present invention two, as shown in Figure 6, may further comprise the steps:
Step S602 arranges the time span (that is maximum duration) that task groups can take CPU.For example, x, y, three variablees of z can be set, corresponding A, three task groups of B, C CPU time length (precision is equal to above-mentioned time source, supposes that here precision is microsecond) that can take respectively, wherein, x=I, y=Y, z=K.
Whether step S604 detects the predetermined duration that distributes and reaches.If arrive this distribution duration, enter step S606, if do not distribute duration to this, then continue to detect.
Step S606 upgrades the time span that each task groups can take CPU.For example, the duration that sets that can distribute its correspondence for each task groups at each time interval m (namely, three values of above-mentioned I, J and K), namely, the time span that each time interval m time integral point, task groups can take CPU is updated to the above-mentioned distribution duration that the maximum duration of this task groups this moment adds this task groups.Be example with task groups A, at each time interval m, the duration of the taken CPU of its correspondence add up I, i.e. x=x+I.Need to prove, if some task groups (are not for example moved for a long time, the user interactions task groups, if the user never uses this equipment, corresponding task groups does not have thing to do, will dormancy), corresponding maximum duration just can not be consumed, and is added to a very big value always, in case this task groups obtains carrying out, will take processor for a long time, cause other task groups to can not get carrying out.So, the threshold values of a maximum duration can be set for each task groups here, after maximum duration surpassed this threshold values, the duration of not reallocating changed task groups to this, up to its maximum duration less than this threshold values.
In specific implementation process, when task is scheduled each time, being scheduled and entering of task, obtain the time t1 of this point by above-mentioned high precision time source, when this task is scheduled away, obtain the time t2 of this point by above-mentioned high precision time source, it is exactly this task run time that t2 deducts t1.In the scheduler task operation, judge that at first the task groups under this task can take CPU time length (namely, maximum duration), whether deduct this working time smaller or equal to 0, if, just forbid that this task scheduling carries out, select other task run, belonging to task groups up to this task, can to take CPU time length be greater than 0.For example, this task belongs to task groups A, then x=I-(t2-t1).
In sum, pass through above-described embodiment, employing discharges the mode that the shared cpu resource of wherein one or more tasks is given ready task in other task groups when task groups surpasses threshold value to taking of cpu resource, the problem of having avoided single (perhaps a plurality of) task to take cpu resource for a long time and having caused other task to be died of hunger, strengthen the stability of system, improved the performance of system.
Obviously, those skilled in the art should be understood that, above-mentioned each module of the present invention or each step can realize with the general calculation device, they can concentrate on the single calculation element, perhaps be distributed on the network that a plurality of calculation elements form, alternatively, they can be realized with the executable program code of calculation element, thereby, they can be stored in the memory storage and be carried out by calculation element, and in some cases, can carry out step shown or that describe with the order that is different from herein, perhaps they are made into each integrated circuit modules respectively, perhaps a plurality of modules in them or step are made into the single integrated circuit module and realize.Like this, the present invention is not restricted to any specific hardware and software combination.
The above is the preferred embodiments of the present invention only, is not limited to the present invention, and for a person skilled in the art, the present invention can have various changes and variation.Within the spirit and principles in the present invention all, any modification of doing, be equal to replacement, improvement etc., all should be included within protection scope of the present invention.
Claims (8)
1. the dispatching method of a cpu resource is characterized in that, may further comprise the steps:
Task division in the system is become task groups;
Add up the current situation that takies to cpu resource of described task groups; And
In that described task groups is current when taking of described cpu resource surpassed default maximal value, discharge the described cpu resource of the one or more tasks in the described task groups to the ready task run in other task groups except this task groups, comprise: if all do not have described ready task in described other task groups, then the task in the described task groups holds over described cpu resource operation; If in described other task groups described ready task is arranged, then discharge the described cpu resource of the one or more tasks in the described task groups to the described ready task run in described other task groups.
2. method according to claim 1 is characterized in that, adds up before the current situation that takies to described cpu resource of described task groups, also comprises:
Set in advance the described maximal value that described task groups takies described cpu resource, wherein, described maximal value is the summation that all tasks take described cpu resource in the described task groups.
3. method according to claim 2 is characterized in that, sets in advance described task groups the described maximal value that described cpu resource takies is comprised:
Be the largest percentage that described task groups distributes it that described cpu resource is taken at each scheduled time slot.
4. method according to claim 3 is characterized in that, adds up the current situation that takies to cpu resource of described task groups and comprises:
Add up described task groups occupancy to described cpu resource in described each scheduled time slot.
5. according to claim 3 or 4 described methods, it is characterized in that the time precision of described each predetermined amount of time is microsecond.
6. the dispatching device of a cpu resource is characterized in that, comprising:
Divide module, be used for the task division of system is become task groups;
Statistical module is used for the current situation that takies to cpu resource of the described task groups of statistics; And
Scheduler module, be used in that described task groups is current when taking of described cpu resource surpassed default maximal value, the described cpu resource that discharges the one or more tasks in the described task groups is given the ready task run in other task groups except this task groups; Wherein, described scheduler module comprises: holding unit, be used for all not having under the situation of described ready task in described other task groups, and keep the task in the described task groups to hold over described cpu resource operation; And releasing unit, be used for having under the situation of described ready task in described other task groups, discharge the described cpu resource of the one or more tasks in the described task groups to the described ready task run in described other task groups.
7. device according to claim 6 is characterized in that, also comprises:
Module is set, is used for setting in advance the described maximal value that described task groups takies described cpu resource, wherein, described maximal value is the summation that all tasks take described cpu resource in the described task groups.
8. device according to claim 7 is characterized in that,
It is described that to arrange that module also is used at each scheduled time slot be the largest percentage that described task groups distributes it that described cpu resource is taken; And
Described statistical module also is used for the described task groups of statistics occupancy to described cpu resource in described each scheduled time slot.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN 201010534278 CN101984414B (en) | 2010-11-05 | 2010-11-05 | Method and device for scheduling central processing unit (CPU) resource |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN 201010534278 CN101984414B (en) | 2010-11-05 | 2010-11-05 | Method and device for scheduling central processing unit (CPU) resource |
Publications (2)
Publication Number | Publication Date |
---|---|
CN101984414A CN101984414A (en) | 2011-03-09 |
CN101984414B true CN101984414B (en) | 2013-08-07 |
Family
ID=43641583
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN 201010534278 Expired - Fee Related CN101984414B (en) | 2010-11-05 | 2010-11-05 | Method and device for scheduling central processing unit (CPU) resource |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN101984414B (en) |
Families Citing this family (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103810048B (en) * | 2014-03-11 | 2017-01-18 | 国家电网公司 | Automatic adjusting method and device for thread number aiming to realizing optimization of resource utilization |
CN105992360B (en) * | 2015-02-16 | 2020-05-26 | 电信科学技术研究院 | Resource allocation method and device |
CN106598740B (en) * | 2016-12-15 | 2020-11-27 | 苏州浪潮智能科技有限公司 | System and method for limiting CPU utilization rate occupied by multithreading program |
CN107193647A (en) * | 2017-05-25 | 2017-09-22 | 深信服科技股份有限公司 | Sandbox environment task processing method and system under a kind of NUMA architecture |
CN107665425B (en) * | 2017-09-27 | 2021-07-30 | 曙光信息产业(北京)有限公司 | Real-time charging method and system based on operation control |
CN109992403B (en) * | 2017-12-30 | 2021-06-01 | ***通信集团福建有限公司 | Optimization method and device for multi-tenant resource scheduling, terminal equipment and storage medium |
CN108984305B (en) * | 2018-07-13 | 2021-03-09 | 威创集团股份有限公司 | Task processing method, device and system |
CN109144725B (en) * | 2018-08-02 | 2020-11-17 | 北京达佳互联信息技术有限公司 | File processing method and device, electronic equipment and storage medium |
CN112395075A (en) * | 2019-08-15 | 2021-02-23 | 阿里巴巴集团控股有限公司 | Resource processing method and device and resource scheduling system |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101533362A (en) * | 2009-04-15 | 2009-09-16 | 南京联创科技股份有限公司 | Inter-process CPU resource balance scheduling method |
CN101706743A (en) * | 2009-12-07 | 2010-05-12 | 北京航空航天大学 | Dispatching method of virtual machine under multi-core environment |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7676809B2 (en) * | 2003-10-09 | 2010-03-09 | International Business Machines Corporation | System, apparatus and method of enhancing priority boosting of scheduled threads |
-
2010
- 2010-11-05 CN CN 201010534278 patent/CN101984414B/en not_active Expired - Fee Related
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101533362A (en) * | 2009-04-15 | 2009-09-16 | 南京联创科技股份有限公司 | Inter-process CPU resource balance scheduling method |
CN101706743A (en) * | 2009-12-07 | 2010-05-12 | 北京航空航天大学 | Dispatching method of virtual machine under multi-core environment |
Also Published As
Publication number | Publication date |
---|---|
CN101984414A (en) | 2011-03-09 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN101984414B (en) | Method and device for scheduling central processing unit (CPU) resource | |
Naik et al. | Performance analysis of job scheduling policies in parallel supercomputing environments | |
CN107018175B (en) | Scheduling method and device of mobile cloud computing platform | |
CN102457906B (en) | Load balancing control method and system of message queues | |
CN1989738A (en) | Propagation of minimum guaranteed scheduling rates | |
CN102902587A (en) | Distribution type task scheduling method, distribution type task scheduling system and distribution type task scheduling device | |
CN102611735A (en) | Load balancing method and system of application services | |
CN102761832B (en) | Message distribution method and device | |
CN101951411A (en) | Cloud scheduling system and method and multistage cloud scheduling system | |
CN103793272A (en) | Periodical task scheduling method and periodical task scheduling system | |
CN113422842B (en) | Distributed power utilization information data acquisition system considering network load | |
CN100562850C (en) | Multi-processor load distribution-regulation method | |
CN103297499A (en) | Scheduling method and system based on cloud platform | |
US10425293B2 (en) | Network resource allocation proposals | |
CN103530185A (en) | Resource optimization method and device | |
CN102521057A (en) | Resource scheduling method and device | |
WO2016101996A1 (en) | Allocating cloud computing resources in a cloud computing environment | |
CN112888005B (en) | MEC-oriented distributed service scheduling method | |
Wang et al. | Utopia: Near-optimal coflow scheduling with isolation guarantee | |
Zhang et al. | Reservation-based resource scheduling and code partition in mobile cloud computing | |
Jang et al. | K-LZF: An efficient and fair scheduling for edge computing servers | |
CN108280018B (en) | Method and system for analyzing and optimizing communication overhead efficiency of node workflow | |
CN105786447A (en) | Method and apparatus for processing data by server and server | |
CN102420741B (en) | The method of dispatching communication flow and device in based on the equipment of ATCA | |
Lin et al. | {RingLeader}: efficiently Offloading {Intra-Server} Orchestration to {NICs} |
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 | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20130807 Termination date: 20191105 |
|
CF01 | Termination of patent right due to non-payment of annual fee |