CN103888279A - Resource pool dynamic management method and system based on feedback control cycle - Google Patents
Resource pool dynamic management method and system based on feedback control cycle Download PDFInfo
- Publication number
- CN103888279A CN103888279A CN201210558692.2A CN201210558692A CN103888279A CN 103888279 A CN103888279 A CN 103888279A CN 201210558692 A CN201210558692 A CN 201210558692A CN 103888279 A CN103888279 A CN 103888279A
- Authority
- CN
- China
- Prior art keywords
- parameter
- resource pool
- output
- adjusted
- module
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Stored Programmes (AREA)
Abstract
The invention relates to a resource pool dynamic management method and a system based on feedback control cycle. The method comprises steps of: receiving operation data outputted by a resource pool and carrying out analysis on the output data to acquire practical performance parameters of a system; pre-setting pre-set system object performance parameters outputted by the system; comparing the practical performance parameters of the system with the pre-set system object performance parameters and calculating an object difference value; carrying out self-adaption calculation of the object different value and outputting adjusting parameters; based on the adjusting parameters, adjusting the system configuration parameters and taking the system parameters as cycle input parameters of the system. The system has a module employing the method. A feedback control cycle principle is utilized, configuration of the resource pool can be automatically adjusted according to the software and hardware environment and the request pressure during system operation in an operation process, utilization efficiency of the resource pool can be improved to a maximum degree, and the system performance is improved.
Description
Technical field
The application's patent application relates to the resource management field of resource pool, especially designs a kind of resource pool dynamic management approach and system based on FEEDBACK CONTROL circulation.
Background technology
Resource pool (Resource Pooling) be for meet public computing with the ability of selecting resource, the computer hardware that supplier must be of certain scale and internet resource are to be combined into resource pool, to allow user's Analysis freely.The target resource of some is provided by resource pool manager, in the time having this resource of request, resource pool is distributed to one, then give this resource identification for busy, be denoted as busy resource and can not be assigned with again use, after some resources use, resource pool is disposed the busy sign of relevant resource, to show that this resource can be used by next one request again.The example of resource pool has a lot, such as the thread pool of tomcat, the database connection pool of hibernate, XML resolves pond etc., and the application of resource pool in actual development is very extensive, by the size in resource allocation pond, the scheduling of CPU can be reached to optimum performance, internal memory uses and reaches minimum etc., and then system resource is efficiently utilized, and thread pool is exactly an example wherein.
Thread pool is a kind of multithreading processing form, in processing procedure, adds task to queue, then after establishment thread, automatically starts these tasks.Thread pool thread is all background thread.Each thread uses the storehouse size of acquiescence, with the priority operation of giving tacit consent to, and in multiple thread units.If certain thread is idle (as waited for certain event) in Managed Code, thread pool makes all processors keep busy another worker thread of insertion.If all thread pool threads all remain busy, but in queue, comprise the work of hang-up, thread pool will create another worker thread but the number of thread exceedes maximum never over time.Exceeding peaked thread can queue up, but they to wait until other threads complete after just start.
Thread pool mainly solves the problem of multiple thread execution in processor unit, can significantly reduce the standby time of processor unit, increases the handling capacity of processor unit.Suppose that the server required time that accomplishes a task is: t1 creates the thread time, and the time that t2 executes the task in thread, t3 destroys the thread time.Thread Pool Technology is paid close attention to just and how to be shortened or to adjust t1, the technology of t3 time, thus improve server program performance.It is t1, and t3 is arranged in respectively the startup of server program and the time period of end or time period that some are idle, in the time that server program is processed client requests, does not have t1 like this, the expense of t3.
In the development process of software systems and internet system, " resource pool " is because its utilance that can improve significantly system resource becomes a class object of often using." resource pool " conventionally need to carry out prior parameter configuration (as core pond size, maximum pond size, waiting list size etc.) and realize dynamic creation and use resource, as follows in the configuration of spring thread pool:
In spring thread pool, need configuration: the active Thread Count of thread pool, the maximum active Thread Count of thread pool, the heap(ed) capacity of queue, principle: in the time that a task is wanted to add thread pool to by execute (Runnable) method: if the quantity now in thread pool is less than core pond size (corePoolSize), even if the thread in thread pool all in idle condition, also will create new thread and process being added of task; If now the quantity in thread pool equals core pond size (corePoolSize), but buffer queue (workQueue) less than, task is placed into buffer queue so.If now the quantity in thread pool is greater than core pond size (corePoolSize), buffer queue workQueue is full, and the quantity in thread pool is less than the heap(ed) capacity (maximumPoolSize) of thread pool, builds new thread and processes being added of task.If now the quantity in thread pool is greater than core pond size, buffer queue workQueue is full, and the quantity in thread pool equals the heap(ed) capacity maximumPoolSize of thread pool, process this task by the specified strategy of processor (handler) so.Namely: the priority of Processing tasks is: kernel thread corePoolSize, task queue workQueue, maximum thread maximumPoolSize, if three has expired, use handler to process unaccepted task.In the time that the number of threads in thread pool is greater than corePoolSize, if certain thread free time exceedes enliven the time (keepAliveTime) of setting, thread will be terminated.Like this, thread pool can be adjusted the Thread Count in pond dynamically.But there are the following problems for this resource distribution mode presetting:
1, taking thread pool as example, reasonably thread pool configuration is significantly improved for the efficiency of considerable task processing, if but selecting inappropriate configuration (excessive or too small) seriously to reduce affects server performance." too small " is by the situation that goes out current task and can not process in time in theory, (some studies show that out that some undersized thread pool does well, this is because test has a lot of thread synchronization expenses in driving, and this expense can not be ignored with respect to the time that completes individual task.) " excessive " there will be the problem that cross-thread synchronization overhead is too large, and switch and consume very much the CPU time at cross-thread, may there is the mistake of " OutOfMemory ".
2, in exploitation, test environment and true line often there is very large difference in the aspect such as environment machine performance and pressure, is therefore difficult to specify on stream the size of optimized parameters.
3, machine dilatation, server increase or minimizing etc. all may form challenge to resource pool configuration before.
Development environment often runs into such problem: optimized configuration is difficult in prior accurate Calculation, such as in using spring thread pool, how to specify corePoolSize, maxPoolSize, the size of the several parameters of queueCapacity, configuring task too little cannot not do completely, " excessive " there will be cross-thread synchronization overhead too large, and the machine of development environment is often mutually far short of what is expected with the machine performance of production environment on line, the task scale facing is also different, and the task scale on line may constantly change equally, the system consumption of each task is also different, therefore in development process, cannot accurately simulate and estimate, often can only by virtue of experience guess configuration, and cannot dynamically adjust.
Summary of the invention
For the defect of above-mentioned prior art, the application utilizes FEEDBACK CONTROL circulation theory, and a kind of method and system of the resource pool dynamic management based on FEEDBACK CONTROL circulation is provided, thereby solves the problem of resource pool dynamic management.
A kind of resource pool dynamic management system based on FEEDBACK CONTROL circulation that the application provides, comprising: performance monitoring module, receive the runtime data that resource pool is exported, and to the described data analysis of output, obtain the actual performance parameter of system; Predetermined system target capabilities module, the predetermined system target capabilities parameter of initialization system output; Goal discrepancy is apart from computing module, and the actual performance parameter of more described system and predetermined system target capabilities parameter, calculate target gap value; Adaptive algorithm module, carries out self adaptation calculating by described target gap value, and parameter is adjusted in output; Configuration adjusting module, based on described adjustment parameter, adjusts system configuration parameter, as the circulation input parameter of described system.The method of a kind of resource pool dynamic management based on FEEDBACK CONTROL circulation of corresponding the application's system, comprising: receive the runtime data of resource pool output, to the described data analysis of output, obtain the actual performance parameter of system; Preset the predetermined system target capabilities parameter of system output; The actual performance parameter of more described system and described predetermined system target capabilities parameter, calculate target gap value; Described target gap value is carried out to self adaptation calculating, and parameter is adjusted in output; Based on described adjustment parameter, system configuration parameter is adjusted, as the circulation input parameter of described system.The scheme that the application provides, development phase distributing rationally without accurate allocated resource pond parameters, in the process of system operation, system is according to dynamically configurations in adjustresources pond automatically such as the hardware environment of actual motion, request pressure, utilization ratio that will maximized raising resource pool, and then improve systematic function.
Brief description of the drawings
In order to be illustrated more clearly in the technical scheme of the embodiment of the present application, below the accompanying drawing of required use during embodiment is described is briefly described, apparently, accompanying drawing in the following describes is only some embodiment of the application, for those of ordinary skill in the art, do not paying under the prerequisite of creative work, can also obtain according to these accompanying drawings other accompanying drawing.
Fig. 1 is the logical organization structure of feedback control system
Fig. 2 is the resource pool dynamic management working mechanism based on FEEDBACK CONTROL circulation
Fig. 3 is performance monitoring module workflow diagram
Fig. 4 is that goal discrepancy is apart from computing module workflow diagram.
Fig. 5 is adaptive algorithm flow chart.
Fig. 6 is the adjustment scheme of calculating single index.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present application, the technical scheme in the embodiment of the present application is clearly and completely described, obviously, described embodiment is only some embodiments of the present application, instead of whole embodiment.Based on the embodiment in the application, those of ordinary skill in the art are not making the every other embodiment obtaining under creative work prerequisite, all belong to the scope of the application's protection.
The application's embodiment is still example taking Spring thread pool as object.The application uses the principle of FEEDBACK CONTROL circulation, Spring thread pool is controlled to parameter and regulate.
FEEDBACK CONTROL circulation (Feedback Control Loop) is applied for many years at various engineering fields, and feedback control system is the automatic control system of setting up based on feedback principle.So-called feedback principle, controls according to the information of system exporting change exactly, by the deviation between comparison system behavior (output) and expected behavior, and eliminates deviation to obtain the systematic function of expection.As shown in Figure 1.In the FEEDBACK CONTROL circulatory system, initial configuration is inputted as system, belong to controllable parameter, interference and noise enter core system 101(as uncontrollable parameter and require the system of controlling), after core system 101 operations, generate observation output data, then yardstick observation module 102, to output data analysis, obtains system monitoring data.Then system is through calculating system performance information, and analysis module 103 compares the performance data of the output of system (having measured output) and system expected data (being reference input).Both if inconsistent (or gap exceeds the threshold values of setting), analysis module 103 is by various adaptive algorithms, propose to adjust strategy (adjustment trigger) to system, export to method of adjustment module 104, method of adjustment module 104 provides positive feedback or negative feedback, control parameter initial configuration parameter to core system 101 is adjusted, and makes the performance data of core system 101 consistent with system expected data, to reach the object of control.
The application uses for reference the principle of the FEEDBACK CONTROL circulatory system, to the various resource distributions of resource pool, according to the dynamic resource demand of system, the output of the system of the configuration of resource and expection relatively, if both deviations exceed acceptable scope, by the feedback information of system self, utilize adaptive algorithm, constantly the resource configuration parameter of system is regulated, to obtain the systematic function of expection.Below in conjunction with accompanying drawing, the application's execution mode is described.
Fig. 2 is the resource pool dynamic management working mechanism of the FEEDBACK CONTROL circulation based on Spring thread pool.The resource pool dynamic management of the FEEDBACK CONTROL circulation shown in Fig. 2, is the control system of a closed loop, comprising: initialization module 201, output system initial configuration; Client 203 is offered the challenge and is processed request to system, as a large amount of tasks is delivered to resource pool module 202(Spring thread pool); Resource pool module 202 is processed (as output run time behaviour data) to the data of client 203; Performance monitoring module 204, to the output data analysis of resource pool module 202, obtains relevant system performance parameter (analysis result); Predetermined system target capabilities module 205, sets the predetermined system target capabilities parameter that resource pool module 202 is exported; Goal discrepancy is apart from computing module 206, the actual system performance parameter of comparison system and this predetermined system target capabilities parameter, and parameter that output can be corrected in subsequent algorithm (correct algorithm enter ginseng), offers adaptive algorithm module 207; Adaptive algorithm module 207, carries out self adaptation calculating, and parameter is adjusted in output; Configuration adjusting module 208, obtain the output (as the parameter after adjusting) of adaptive algorithm module 207, system configuration parameter is adjusted, as configuration parameter delivered to resource pool module 202, as the circulation input parameter that carries out data processing of resource pool module 202.Adaptive algorithm module 207 also can be calculated actual performance parameter, the parameter after adjusting is offered to configuration adjusting module 208 and configure use.But it is not more excellent apart from the adaptive algorithm effect that enters to have joined of calculating rear increase correct algorithm by goal discrepancy.
The workflow of system is as follows:
1) first, initialization module 201 carries out initial configuration to spring thread pool 202 in advance in the development phase, comprises core pond size, maximum pond size, waiting list size, thread free time etc.
2) Spring thread pool 202 is after the task requests order of receiving client 203, production environment brings into operation on line, now have task unpredictable scale, that system consumption is different to be delivered in thread pool and carry out, the interference of correspondence system and noise, belong to uncontrollable parameter.
3) Spring thread pool 202, in executing the task, can be exported necessary daily record.
4) performance monitoring module 204 is collected these data, analyze, output analysis result, comprise: the time of task average waiting, the task that the unit interval is dropped, the Thread Count that the unit interval creates and the Thread Count of destruction (this two for weighing the overhead of thread creation, destruction).
5) goal discrepancy apart from computing module 206 by these results and the target capabilities parameter obtained from " predetermined system target capabilities module 205 " (
obtain target capabilities index), and the gap of outputting measurement value in calculating and step 4).
6) adaptive algorithm module 207 is calculated above-mentioned difference data by adaptive algorithm, draws adjustment scheme, and adjustment scheme can be positive feedback or negative-feedback information, and adaptive algorithm proposes parameter strategy.
7) configuration adjusting module 208 acts on spring thread pool 202 by this adjustment scheme and configures, and spring loads new configuration thread pool service is provided.
8) the 2nd)-7) process reruns, and forms FEEDBACK CONTROL circulation.
The FEEDBACK CONTROL circulation that above-mentioned flow process forms is adjusted in real time thread pool configuration according to instant situation, to realize the high efficiency utilization of thread pool and then can improve systematic function in running.
Fig. 3 has shown the workflow of performance monitoring module.Performance monitoring module is received the run time behaviour data of Spring thread pool, and step 301 is obtained the daily record of fixed time sheet, and this timeslice is generally the time interval (t* be current time) of t*-T to t*.Then step 302 carries out analytical calculation to daily record, and step 303 obtains the following data within the scope of detection time: the i) time T 1 of task average waiting; Ii) the number of tasks C1 being dropped; Iii) the Thread Count C2 creating; Iv) the Thread Count C3 destroying.
Wherein C2 and C3 are for weighing the overhead of thread creation and destruction.Then performance monitoring module is passed to goal discrepancy apart from computing module the performance parameter analysis result obtaining.
Goal discrepancy calculates the deviation of actual performance parameter and predetermined system target capabilities apart from computing module 206, its workflow as shown in Figure 4.First, goal discrepancy obtains default aims of systems performance parameter T1 ', C1 ', C2 ' and C3 ' (step S401) from predetermined system target capabilities module apart from computing module.Wherein, parameter preset is specified in realization, is the expectation of system operation.Step S402 obtains the measured data that performance detection module obtains, and then compares in the measured data of step 403 parameter preset and performance detection module acquisition, calculates goal discrepancy distance.This calculates gap using the input data as adaptive algorithm module.Corresponding goal discrepancy apart from computing formula is: Δ T1=T1 '-T1; Δ C1=C1 '-C1; Δ C2=C2 '-C2; Δ C3=C3 '-C3.
Adaptive algorithm module obtains after goal discrepancy distance, will be according to target T1, and C1, C2, may there is some irrational place in the higher or on the low side configuration that reflects thread pool of the several indexs of C3.Concrete relation is as follows:
Then, the result that adaptive algorithm module reflects according to data obtains the corresponding strategy of adjusting, and is adaptive algorithm.The flow chart of adaptive algorithm as shown in Figure 5.First, whether four described indexs of adaptive algorithm module judgement have exceeded threshold values (step S501), if do not had, do not need to adjust any parameter, and adaptive algorithm module finishes to adjust.If there is goal discrepancy to exceed threshold values apart from Δ, adaptive algorithm module is adjusted (step S502) to each index performance parameter, until complete the adjustment (step S503) to all single performance parameters that exceed threshold values.The adjustment scheme of the performance parameter that need to adjust each is described in detail in Fig. 6.Adjusting after all performance parameters that exceeds threshold values, adaptive algorithm module is given configuration adjusting module 208 in step S504 output adjustment scheme.
To exceeding the adjustment process of described single performance parameter of threshold values, calculate as shown in Figure 6 the flow process of the adjustment scheme of single index.If Δ >0, finely tunes the corresponding configuration parameter in resource pool being reduced to 1 unit in configuration.Otherwise, if corresponding configuration parameter value in configuration is added to 1 to finely tune.
Taking example above as example:
Wherein, corePoolSize, maxPoolSize, queueCapacity three's unit is " number of thread ", and keepAliveSeconds unit is millisecond.Such as, if Δ T1>0, the time T 1 of task average waiting of output is on the low side the time of default task average waiting (lower than), according to the relation in upper table, illustrate that maximum resource tankage (maxPoolSize) and core pond size arrange arranging of (corePoolSize) excessive, so need to finely tune maxPoolSize and corePoolSize, reduce by a unit, reduce separately by a thread.
It should be noted that, regulation rule also not necessarily will be according to identical rule, can select the stride of adjusting according to the character of different parameters, for example keepAliveSeconds can increase at every turn or reduce 10ms, and using increase and decrease 1 is herein to be a process for fine setting in order to embody adjustment..
Because circulating in, FEEDBACK CONTROL continues to carry out, will soon be repaired appears after deviation in each index, so deviate there will not be very much larger situation, each adjustment adds 1 or subtract 1 just passable, even if there is relatively large deviation, circulation several times also just can have been adjusted.The performance consumption of bringing in order to reduce the configuration of amendment thread pool itself, all carries out parameter adjustment without each circulation, can set in advance a threshold values for the deviation of each index, modifies when deviate exceedes this threshold values again.
The adjustment scheme that configuration adjusting module 208 is exported according to adaptive algorithm module 207, the configuration of amendment spring thread pool, the new configuration of spring thread module loading provides thread pool service.
Each embodiment in this specification is general, and the mode of going forward one by one that adopts is described, and what each embodiment stressed is and the difference of other embodiment, between each embodiment identical similar part mutually referring to.
The application can describe in the general context of computer executable instructions, for example program module or unit.Usually, program module or unit can comprise and carry out particular task or realize routine, program, object, assembly, data structure of particular abstract data type etc.In general, program module or unit can be realized by software, hardware or both combinations.Also can in distributed computing environment (DCE), put into practice the application, in these distributed computing environment (DCE), be executed the task by the teleprocessing equipment being connected by communication network.In distributed computing environment (DCE), program module or unit can be arranged in the local and remote computer-readable storage medium including memory device.
Finally, also it should be noted that, term " comprises ", " comprising " or its any other variant are intended to contain comprising of nonexcludability, thereby the process, method, commodity or the equipment that make to comprise a series of key elements not only comprise those key elements, but also comprise other key elements of clearly not listing, or be also included as the intrinsic key element of this process, method, commodity or equipment.The in the situation that of more restrictions not, the key element being limited by statement " comprising ... ", and be not precluded within process, method, commodity or the equipment that comprises described key element and also have other identical element.
Those skilled in the art should understand, the application's embodiment can be provided as method, system or computer program.Therefore, the application can adopt complete hardware implementation example, completely implement software example or the form in conjunction with the embodiment of software and hardware aspect.And the application can adopt the form at one or more upper computer programs of implementing of computer-usable storage medium (including but not limited to magnetic disc store, CD-ROM, optical memory etc.) that wherein include computer usable program code.
Applied principle and the execution mode of specific case to the application herein and set forth, the explanation of above embodiment is just for helping to understand the application's method and main thought thereof; , for one of ordinary skill in the art, according to the application's thought, all will change in specific embodiments and applications, in sum, this description should not be construed as the restriction to the application meanwhile.
Claims (14)
1. the resource pool dynamic management system based on FEEDBACK CONTROL circulation, is characterized in that, comprising:
Performance monitoring module, receives the runtime data that resource pool is exported, and to the described data analysis of output, obtains the actual performance parameter of system;
Predetermined system target capabilities module, the predetermined system target capabilities parameter of initialization system output;
Goal discrepancy is apart from computing module, and the actual performance parameter of more described system and predetermined system target capabilities parameter, calculate target gap value;
Adaptive algorithm module, carries out self adaptation calculating by described target gap value, and parameter is adjusted in output;
Configuration adjusting module, based on described adjustment parameter, adjusts system configuration parameter, as the circulation input parameter of described system.
2. the system as claimed in claim 1, is characterized in that, in the runtime data of described resource pool output, also comprises the daily record of work of described resource pool.
3. system as claimed in claim 2, is characterized in that, described performance monitoring module also comprises: the described data analysis of output is comprised and analyzes described daily record of work, to obtain the data within the scope of monitoring time.
4. the system as claimed in claim 1, is characterized in that, described self adaptation is calculated and also comprised: when described target gap value exceeds default threshold values, calculate and adjust, parameter is adjusted in output.
5. system as claimed in claim 4, is characterized in that, described self adaptation is calculated and also comprised: described target gap value comprises one or more indexs; Finish computing and do not export adjustment parameter when described one or more indexs all do not exceed described threshold value; Calculate adjustment when described one or more indexs one or more exceed described threshold value, parameter is adjusted in output.
6. system as claimed in claim 5, it is characterized in that, described self adaptation is calculated and is also comprised: during described calculating is adjusted, traveling through the whole of described one or more indexs does not finish, calculate one by one the adjustment of single index, finish until travel through the whole of described one or more indexs, after completing, parameter is adjusted in output.
7. the system as described in one of claim 1-5, is characterized in that, also comprises:
Resource pool module, utilizes resource pool to process the task requests of client, and exports runtime data.
8. a method for the resource pool dynamic management based on FEEDBACK CONTROL circulation, is characterized in that, comprising:
Receive the runtime data of resource pool output, to the described data analysis of output, obtain the actual performance parameter of system;
Preset the predetermined system target capabilities parameter of system output;
The actual performance parameter of more described system and described predetermined system target capabilities parameter, calculate target gap value;
Described target gap value is carried out to self adaptation calculating, and parameter is adjusted in output; Based on described adjustment parameter, system configuration parameter is adjusted, as the circulation input parameter of described system.
9. method as claimed in claim 8, is characterized in that, in the runtime data of described resource pool output, also comprises the daily record of work of described resource pool.
10. method as claimed in claim 9, is characterized in that, the described data analysis of output is comprised: analyze described daily record of work, to obtain the data within the scope of monitoring time.
11. methods as claimed in claim 8, is characterized in that, described self adaptation is calculated and also comprised: when described target gap value exceeds default threshold values, calculate and adjust, parameter is adjusted in output.
12. methods as claimed in claim 11, is characterized in that, described self adaptation is calculated and also comprised: described target gap value comprises one or more indexs; Finish computing and do not export adjustment parameter when described one or more indexs all do not exceed described threshold value; Calculate adjustment when described one or more indexs one or more exceed described threshold value, parameter is adjusted in output.
13. methods as claimed in claim 12, it is characterized in that, described self adaptation is calculated and is also comprised: during described calculating is adjusted, traveling through the whole of described one or more indexs does not finish, calculate one by one the adjustment of single index, finish until travel through the whole of described one or more indexs, after completing, parameter is adjusted in output.
14. methods as described in one of claim 8-13, is characterized in that, also comprise:
Utilize described resource pool to process the task requests of client, and export runtime data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210558692.2A CN103888279A (en) | 2012-12-20 | 2012-12-20 | Resource pool dynamic management method and system based on feedback control cycle |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210558692.2A CN103888279A (en) | 2012-12-20 | 2012-12-20 | Resource pool dynamic management method and system based on feedback control cycle |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103888279A true CN103888279A (en) | 2014-06-25 |
Family
ID=50957006
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210558692.2A Pending CN103888279A (en) | 2012-12-20 | 2012-12-20 | Resource pool dynamic management method and system based on feedback control cycle |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103888279A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104461845A (en) * | 2014-11-17 | 2015-03-25 | 中国航天科工集团第二研究院七〇六所 | Self-adaption method of thread pool of log collection system |
| CN107463441A (en) * | 2017-06-30 | 2017-12-12 | 大唐软件技术股份有限公司 | A kind of number of threads control method and equipment |
| CN109753416A (en) * | 2018-12-11 | 2019-05-14 | 东软集团股份有限公司 | Agent agent capabilities control method, device, storage medium and electronic equipment |
| CN109766143A (en) * | 2019-01-16 | 2019-05-17 | 广州虎牙信息科技有限公司 | Object loading method, device, terminal and storage medium |
| CN111324644A (en) * | 2018-11-28 | 2020-06-23 | 中国移动通信集团浙江有限公司 | Method and device for monitoring database connection storm under large-scale micro-service architecture |
| CN112039934A (en) * | 2019-06-03 | 2020-12-04 | 大唐移动通信设备有限公司 | Information feedback method, feedback information processing method and device |
| CN112817719A (en) * | 2021-01-28 | 2021-05-18 | 平安普惠企业管理有限公司 | Method, device and equipment for adjusting parameters of thread pool and readable storage medium |
| CN114265637A (en) * | 2021-12-24 | 2022-04-01 | 中电信数智科技有限公司 | Archive management system configuration method, system, electronic device and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101043389A (en) * | 2007-04-20 | 2007-09-26 | 北京航空航天大学 | Control system of grid service container |
| CN101364185A (en) * | 2008-09-02 | 2009-02-11 | 中国科学院软件研究所 | Thread pool size self-adapting regulating method and concurrency control method applying server |
| CN102043675A (en) * | 2010-12-06 | 2011-05-04 | 北京华证普惠信息股份有限公司 | Thread pool management method based on task quantity of task processing request |
| US20110161972A1 (en) * | 2009-12-24 | 2011-06-30 | International Business Machines Corporation | Goal oriented performance management of workload utilizing accelerators |
| CN102147746A (en) * | 2010-03-05 | 2011-08-10 | 微软公司 | Dynamic thread pool management |
| CN102360310A (en) * | 2011-09-28 | 2012-02-22 | 中国电子科技集团公司第二十八研究所 | Multitask process monitoring method and system in distributed system environment |
-
2012
- 2012-12-20 CN CN201210558692.2A patent/CN103888279A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101043389A (en) * | 2007-04-20 | 2007-09-26 | 北京航空航天大学 | Control system of grid service container |
| CN101364185A (en) * | 2008-09-02 | 2009-02-11 | 中国科学院软件研究所 | Thread pool size self-adapting regulating method and concurrency control method applying server |
| US20110161972A1 (en) * | 2009-12-24 | 2011-06-30 | International Business Machines Corporation | Goal oriented performance management of workload utilizing accelerators |
| CN102147746A (en) * | 2010-03-05 | 2011-08-10 | 微软公司 | Dynamic thread pool management |
| CN102043675A (en) * | 2010-12-06 | 2011-05-04 | 北京华证普惠信息股份有限公司 | Thread pool management method based on task quantity of task processing request |
| CN102360310A (en) * | 2011-09-28 | 2012-02-22 | 中国电子科技集团公司第二十八研究所 | Multitask process monitoring method and system in distributed system environment |
Non-Patent Citations (3)
| Title |
|---|
| 唐伟: "《基于多阶段的公共I/O服务的研究与实现》", 《信息科技辑》 * |
| 唐国钟: "《基于SEDA的自适应性能优化技术》", 《万方数据企业知识服务平台》 * |
| 罗海南: "《多阶段事件驱动架构性能调优机制的研究》", 《万方数据企业知识服务平台》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104461845A (en) * | 2014-11-17 | 2015-03-25 | 中国航天科工集团第二研究院七〇六所 | Self-adaption method of thread pool of log collection system |
| CN104461845B (en) * | 2014-11-17 | 2017-05-17 | 中国航天科工集团第二研究院七〇六所 | Self-adaption method of thread pool of log collection system |
| CN107463441A (en) * | 2017-06-30 | 2017-12-12 | 大唐软件技术股份有限公司 | A kind of number of threads control method and equipment |
| CN107463441B (en) * | 2017-06-30 | 2020-10-27 | 大唐软件技术股份有限公司 | Thread quantity control method and equipment |
| CN111324644A (en) * | 2018-11-28 | 2020-06-23 | 中国移动通信集团浙江有限公司 | Method and device for monitoring database connection storm under large-scale micro-service architecture |
| CN111324644B (en) * | 2018-11-28 | 2023-10-27 | 中国移动通信集团浙江有限公司 | Method and device for monitoring database connection storm under large-scale micro-service architecture |
| CN109753416A (en) * | 2018-12-11 | 2019-05-14 | 东软集团股份有限公司 | Agent agent capabilities control method, device, storage medium and electronic equipment |
| CN109766143A (en) * | 2019-01-16 | 2019-05-17 | 广州虎牙信息科技有限公司 | Object loading method, device, terminal and storage medium |
| CN112039934A (en) * | 2019-06-03 | 2020-12-04 | 大唐移动通信设备有限公司 | Information feedback method, feedback information processing method and device |
| CN112817719A (en) * | 2021-01-28 | 2021-05-18 | 平安普惠企业管理有限公司 | Method, device and equipment for adjusting parameters of thread pool and readable storage medium |
| CN114265637A (en) * | 2021-12-24 | 2022-04-01 | 中电信数智科技有限公司 | Archive management system configuration method, system, electronic device and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103888279A (en) | Resource pool dynamic management method and system based on feedback control cycle | |
| US11080095B2 (en) | Scheduling of work interval objects in an AMP architecture using a closed loop performance controller | |
| US10261818B2 (en) | Optimizing virtual machine synchronization for application software | |
| US10444812B2 (en) | Power shifting in multicore platforms by varying SMT levels | |
| Boyer et al. | Load balancing in a changing world: dealing with heterogeneity and performance variability | |
| KR101953906B1 (en) | Apparatus for scheduling task | |
| CN102521055B (en) | Virtual machine resource allocating method and virtual machine resource allocating system | |
| CN106557369A (en) | A kind of management method and system of multithreading | |
| CN102043675A (en) | Thread pool management method based on task quantity of task processing request | |
| CN110300959B (en) | Method, system, device, apparatus and medium for dynamic runtime task management | |
| CN103268247A (en) | Method and device for executing task and adjusting number of remaining threads in thread pool | |
| TW201342030A (en) | Instruction that specifies an application thread performance state | |
| CN106293947B (en) | GPU-CPU (graphics processing Unit-Central processing Unit) mixed resource allocation system and method in virtualized cloud environment | |
| CN111090505B (en) | Task scheduling method and system in multiprocessor system | |
| EP2960787A1 (en) | A method of executing an application on a computer system, a resource manager and a high performance computer system | |
| CN101685335A (en) | Application server based on SEDA as well as energy-saving device and method thereof | |
| Xu et al. | Multi resource scheduling with task cloning in heterogeneous clusters | |
| Singla et al. | Task Scheduling Algorithms for Grid Computing with Static Jobs: A Review | |
| Dokulil et al. | Adaptive scheduling of collocated applications using a task-based runtime system | |
| US8352723B2 (en) | Automatic maintenance of a computing system in a steady state using correlation | |
| Golchin | Control based tickless scheduling | |
| Sethi | Learning Energy-Aware Transaction Scheduling in Database Systems | |
| Kunas et al. | Managing the residual load imbalance and clock frequency to save energy on iterative applications | |
| WO2023082723A1 (en) | Processor circuit, power supply control method and terminal device | |
| Tan et al. | On the Energy-efficient Scheduling for Coal Mine Heterogeneous Computing System |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20140625 |