CN109144693A - A kind of power adaptive method for scheduling task and system - Google Patents

A kind of power adaptive method for scheduling task and system Download PDF

Info

Publication number
CN109144693A
CN109144693A CN201810882835.2A CN201810882835A CN109144693A CN 109144693 A CN109144693 A CN 109144693A CN 201810882835 A CN201810882835 A CN 201810882835A CN 109144693 A CN109144693 A CN 109144693A
Authority
CN
China
Prior art keywords
processor
state
type
task
scheduling
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201810882835.2A
Other languages
Chinese (zh)
Other versions
CN109144693B (en
Inventor
冯国富
舒玉娟
陈明
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai Agriculture Information Co ltd
Original Assignee
Shanghai Maritime University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shanghai Maritime University filed Critical Shanghai Maritime University
Priority to CN201810882835.2A priority Critical patent/CN109144693B/en
Publication of CN109144693A publication Critical patent/CN109144693A/en
Application granted granted Critical
Publication of CN109144693B publication Critical patent/CN109144693B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/48—Program initiating; Program switching, e.g. by interrupt
    • G06F9/4806—Task transfer initiation or dispatching
    • G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
    • G06F9/4881—Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
    • G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2209/00—Indexing scheme relating to G06F9/00
    • G06F2209/50—Indexing scheme relating to G06F9/50
    • G06F2209/5017—Task decomposition
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2209/00—Indexing scheme relating to G06F9/00
    • G06F2209/50—Indexing scheme relating to G06F9/50
    • G06F2209/5018—Thread allocation

Landscapes

  • Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Supply And Distribution Of Alternating Current (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

本发明公开了一种功率自适应任务调度方法及系统。该方法包括：获取当前的任务信息；根据大数据编程模型MapReduce对所述任务信息进行划分和统计，得到多个可并行运行的子任务；获取输出功率的变动状态以及任务负载状态，得到状态信息；根据所述状态信息调整处理器的核数及类型；根据调整核数及类型后的处理器，将所述子任务调度到各工作者线程。本方法或系统能够解决目前能量收集系统在任务调度过程中没有考虑处理器工作功率需求和能量收集单元输出功率的不确定性造成的任务执行中断以及效能低等问题。

The invention discloses a power adaptive task scheduling method and system. The method includes: acquiring current task information; dividing and counting the task information according to a big data programming model MapReduce to obtain a plurality of subtasks that can be run in parallel; ; adjust the number and type of cores of the processor according to the state information; schedule the subtasks to each worker thread according to the processor after the number and type of cores are adjusted. The method or system can solve the problems of task execution interruption and low efficiency caused by the current energy harvesting system not considering the processor working power requirement and the uncertainty of the output power of the energy harvesting unit in the task scheduling process.

Description

A kind of power adaptive method for scheduling task and system
Technical field
The present invention relates to task schedule fields, more particularly to a kind of power adaptive method for scheduling task and system.
Background technique
In recent years, the development that technology of Internet of things is advanced by leaps and bounds, high-performance, low-power consumption have become internet of things equipment Important trend.High performance demand means the promotion of system energy consumption, and the development speed of current battery is much fallen Afterwards in the growth of energy requirements, and battery power supply still remains the problem that volume weight is big and maintenance cost is high.Using novel The various environmental energy (such as sunlight, wind, tide and vibration etc.) that nature is widely present are converted electricity by energy collection technology It can and store and utilize a kind of effective way for being solution energy problem.However, due to external environment resource have it is intermittent, with The features such as machine and uncertainty, energy collecting system output-power fluctuation, therefore, Internet of things node are needed by rationally utilizing Output power, and reasonable task schedule is carried out according to the power demand of different loads, so that power usage efficiency optimizes.
The method of existing combination task schedule research energy collecting system, main focus utilization power conditioning technology is in energy Amount constrains lower maximum system performance, or minimizes system energy consumption under the constraint of energy collection unit output power.For example, Using dynamic frequency pressure regulation, to reduce processor energy consumption.Currently, there are many power regulations in managing power consumption hardware view Technology, as dynamic voltage frequency adjustment (DVFS) can reduce system power dissipation by adjusting voltage and clock frequency；CPU hot plug (CPU Hotplug) technology adjusts system processor power consumption by increasing or decreasing processor nucleus number online.Opposite hardware view Fast development to managing power consumption, managing power consumption software technology research relatively lag behind, these technical applications are in energy collecting system In need software and hardware mutual cooperation competence exertion maximal efficiency, and task is basic in the case where no enough output powers It cannot execute.Therefore, energy collecting system allows for dynamically according to processor operating power requirements and energy collection unit Output power manages and scheduler task.
Summary of the invention
The object of the present invention is to provide a kind of power adaptive method for scheduling task and systems, receive to solve current energy Collecting system do not accounted in task scheduling process processor operating power requirements and energy collection unit output power not really The problems such as task execution caused by qualitative is interrupted and efficiency is low.
To achieve the above object, the present invention provides following schemes:
A kind of power adaptive method for scheduling task, which comprises
Obtain current mission bit stream；
The mission bit stream is divided and is counted according to big data programming model MapReduce, obtain it is multiple can be simultaneously The subtask of row operation；
The upset condition and task load state for obtaining output power, obtain status information；
The nucleus number and type of processor are adjusted according to the state information；
According to the processor after adjustment nucleus number and type, by the subtask scheduling to each worker thread.
Optionally, described that the mission bit stream is divided and counted according to big data programming model MapReduce, it obtains To multiple subtasks that can be run parallel, specifically include:
The mission bit stream is divided by Map function；
The mission bit stream after division is counted by Reduce function, obtains multiple subtasks that can be run parallel.
Optionally, the nucleus number and type for adjusting processor according to the state information, specifically includes:
Obtain the operating power requirements of processor；
According to the operating power requirements and the status information, the core of CPU hot plug technology adjustment processor is utilized Several and type.
Optionally, the nucleus number and type according to processor adjusted, by the subtask scheduling to each worker Thread specifically includes:
The state of processor after obtaining adjustment nucleus number and type, the state includes presence and off-line state；
Obtain adjustment nucleus number and type after processor change conditions, the change conditions include processor heat remove with And processor heat insertion；
According to the state and change conditions of the processor, by the subtask scheduling to each worker thread.
Optionally, successively by the subtask scheduling to each worker thread in a manner of first in, first out.
A kind of power adaptive task scheduling system, the system comprises:
Mission bit stream obtains module, for obtaining current mission bit stream；
Division and statistical module, for being divided according to big data programming model MapReduce to the mission bit stream And statistics, obtain multiple subtasks that can be run parallel；
State information acquisition module obtains state for obtaining the upset condition and task load state of output power Information；
Module is adjusted, for adjusting the nucleus number and type of processor according to the state information；
Scheduler module, for according to adjustment nucleus number and type after processor, by the subtask scheduling to each worker Thread.
Optionally, the division and statistical module specifically include:
Division unit, for being divided the mission bit stream by Map function；
Statistic unit, for the mission bit stream after division to be counted by Reduce function, obtain it is multiple can be parallel The subtask of operation.
Optionally, the adjustment module specifically includes:
Operating power requirements acquiring unit, for obtaining the operating power requirements of processor；
Adjustment unit, for utilizing CPU hot plug technology tune according to the operating power requirements and the status information The nucleus number and type of whole processor.
Optionally, the scheduler module specifically includes:
State acquiring unit, for obtaining the state of the processor after adjusting nucleus number and type, the state includes online State and off-line state；
Change conditions acquiring unit, for obtaining the change conditions of the processor after adjusting nucleus number and type, the variation Situation includes the removal of processor heat and the insertion of processor heat；
Scheduling unit, for the state and change conditions according to the processor, by the subtask scheduling to each work Author's thread.
Optionally, the scheduler module is in a manner of first in, first out successively by the subtask scheduling to each worker's line Journey.
Compared with prior art, the present invention has following technical effect that the present invention in the case where output power changes, is led to On-line tuning system processor nucleus number and type are crossed, system processor operating power requirements is made to adapt to energy collection unit output work The dynamic change of rate, and the migration of binding operation system task and thread affinity managing and serving system thread, are dispatched using phoenix Strategy by task schedule to worker thread, thus avoid output power insufficient and output power surplus to system effectiveness not Benefit influences.This method is suitable for the occasion of energy collecting system collection of energy output-power fluctuation, and efficiency is substantially better than biography System dispatching method.
Detailed description of the invention
It in order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, below will be to institute in embodiment Attached drawing to be used is needed to be briefly described, it should be apparent that, the accompanying drawings in the following description is only some implementations of the invention Example, for those of ordinary skill in the art, without any creative labor, can also be according to these attached drawings Obtain other attached drawings.
Fig. 1 is the flow chart of power adaptive of embodiment of the present invention method for scheduling task；
Fig. 2 is managing and serving system thread schematic diagram；
Fig. 3 is the structural block diagram of self-adapting task scheduling of embodiment of the present invention system.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other Embodiment shall fall within the protection scope of the present invention.
In order to make the foregoing objectives, features and advantages of the present invention clearer and more comprehensible, with reference to the accompanying drawing and specific real Applying mode, the present invention is described in further detail.
As shown in Figure 1, a kind of power adaptive method for scheduling task the following steps are included:
Step 101: obtaining current mission bit stream.
Step 102: the mission bit stream being divided and counted according to big data programming model MapReduce, is obtained Multiple subtasks that can be run parallel.Entire data handling procedure is attributed to two functions: Map and Reduce.Map be responsible for by Data processing load be divided into it is multiple can Parallel Scheduling task, and by all task piecemeals processing.Reduce is responsible for will be each Task processing result carries out data aggregate.
Step 103: obtaining the upset condition and task load state of output power, obtain status information.
Step 104: adjusting the nucleus number and type of processor according to the state information.
Core is adjusted automatically according to the change conditions of energy collection unit power output.Automatic tune core strategy is according to Optimum Matching The processor nucleus number and type that algorithms selection and energy collection unit output power most match finally provide CPU heat by OS and insert Technology (CPU Hotplug) is pulled out to realize.Current system is worked using CPU hot plug technology processor nucleus number and type into Row adjustment, and system processor operating power under different situations is tested respectively, come analysis processor operating power requirements and system Relationship between current processor nucleus number and type.Power adaptive tune core module perceives output according to power monitoring unit Power, according to the analysis to relationship between processor operating power requirements and system current processor nucleus number and type as a result, It makes adjustment to system current processor nucleus number and type.
Step 105: according to the processor after adjustment nucleus number and type, by the subtask scheduling to each worker thread.
As shown in Fig. 2, giving full play to the parallel processing capability of multi-core processor system using multithreading task schedule.According to The change conditions of current processor nucleus number, the migration of binding operation system process and thread affinity managing and serving system thread, and benefit With Phoenix scheduling strategy by task schedule to worker thread.It, will be in threadiness according to the online situation of current processor core Thread affinity is arranged in the corresponding worker thread of the processor core of state, is bundled on the core；In addition, being in off-line state The corresponding worker thread of processor core be not provided with thread affinity, and the suspend mode thread.
According to the change conditions of current processor core, the migration of binding operation system process and thread affinity managing and serving system Thread, multithreading task schedule are based on following rule and are adjusted: when certain processor core is heat-removed, executing in processor core Upper worker thread will be migrated by OS to running on other online processing device cores, be carrying out on the worker thread with emptying Task, then the suspend mode worker thread, makes it that can not continue to obtain task from task queue and executes；When certain processor When core is inserted by heat, then relevant work person's thread is waken up, and thread affinity is set and is tied on the processor core.
Task dispatcher is in a manner of first in first out successively by task schedule to each worker thread.Worker thread is appointed After the completion of business executes, without completing until all working person thread whole task execution, task dispatcher is directly by new task tune Spend the worker thread.
The specific embodiment provided according to the present invention, the invention discloses following technical effects: the present invention is in output power In the case where variation, by on-line tuning system processor nucleus number and type, system processor operating power requirements is made to adapt to energy Measure the dynamic change of collector unit output power, and the migration of binding operation system task and thread affinity managing and serving system line Journey, using phoenix scheduling strategy by task schedule to worker thread, to avoid output power insufficient and output power Adverse effect of the surplus to system effectiveness.This method is suitable for the occasion of energy collecting system collection of energy output-power fluctuation, And efficiency is substantially better than conventional scheduling method.
As shown in figure 3, a kind of power adaptive task scheduling system includes:
Mission bit stream obtains module 301, for obtaining current mission bit stream.
Division and statistical module 302, for being drawn according to big data programming model MapReduce to the mission bit stream Divide and count, obtains multiple subtasks that can be run parallel.
The division and statistical module 302 specifically include:
Division unit, for being divided the mission bit stream by Map function；
Statistic unit, for the mission bit stream after division to be counted by Reduce function, obtain it is multiple can be parallel The subtask of operation.
State information acquisition module 303 obtains shape for obtaining the upset condition and task load state of output power State information.
Module 304 is adjusted, for adjusting the nucleus number and type of processor according to the state information.
The adjustment module 304 specifically includes:
Operating power requirements acquiring unit, for obtaining the operating power requirements of processor；
Adjustment unit, for utilizing CPU hot plug technology tune according to the operating power requirements and the status information The nucleus number and type of whole processor.
Scheduler module 305, for according to adjustment nucleus number and type after processor, by the subtask scheduling to each work Person's thread.The scheduler module is in a manner of first in, first out successively by the subtask scheduling to each worker thread.
The scheduler module 305 specifically includes:
State acquiring unit, for obtaining the state of the processor after adjusting nucleus number and type, the state includes online State and off-line state；
Change conditions acquiring unit, for obtaining the change conditions of the processor after adjusting nucleus number and type, the variation Situation includes the removal of processor heat and the insertion of processor heat；
Scheduling unit, for the state and change conditions according to the processor, by the subtask scheduling to each work Author's thread.
Each embodiment in this specification is described in a progressive manner, the highlights of each of the examples are with other The difference of embodiment, the same or similar parts in each embodiment may refer to each other.For system disclosed in embodiment For, since it is corresponded to the methods disclosed in the examples, so being described relatively simple, related place is said referring to method part It is bright.
Used herein a specific example illustrates the principle and implementation of the invention, and above embodiments are said It is bright to be merely used to help understand method and its core concept of the invention；At the same time, for those skilled in the art, foundation Thought of the invention, there will be changes in the specific implementation manner and application range.In conclusion the content of the present specification is not It is interpreted as limitation of the present invention.

Claims (10)

1.一种功率自适应任务调度方法，其特征在于，所述方法包括：1. A power adaptive task scheduling method, wherein the method comprises: 获取当前的任务信息；Get current task information; 根据大数据编程模型MapReduce对所述任务信息进行划分和统计，得到多个可并行运行的子任务；According to the big data programming model MapReduce, the task information is divided and counted, and multiple subtasks that can be run in parallel are obtained; 获取输出功率的变动状态以及任务负载状态，得到状态信息；Obtain the changing state of the output power and the task load state, and obtain the state information; 根据所述状态信息调整处理器的核数及类型；Adjust the core number and type of the processor according to the state information; 根据调整核数及类型后的处理器，将所述子任务调度到各工作者线程。The subtasks are scheduled to each worker thread according to the adjusted core number and type of the processor. 2.根据权利要求1所述的调度方法，其特征在于，所述根据大数据编程模型MapReduce对所述任务信息进行划分和统计，得到多个可并行运行的子任务，具体包括：2. The scheduling method according to claim 1, wherein the task information is divided and counted according to the big data programming model MapReduce to obtain a plurality of subtasks that can run in parallel, specifically including: 通过Map函数将所述任务信息进行划分；Divide the task information through the Map function; 通过Reduce函数将划分后的任务信息进行统计，得到多个可并行运行的子任务。Through the Reduce function, the divided task information is counted, and multiple subtasks that can be run in parallel are obtained. 3.根据权利要求1所述的调度方法，其特征在于，所述根据所述状态信息调整处理器的核数及类型，具体包括：3 . The scheduling method according to claim 1 , wherein the adjusting the number and type of cores of the processor according to the state information specifically includes: 3 . 获取处理器的工作功率需求；Get the working power requirements of the processor; 根据所述工作功率需求以及所述状态信息，利用CPU热插拔技术调整处理器的核数及类型。According to the working power requirement and the status information, the number and type of processor cores are adjusted by using the CPU hot-plug technology. 4.根据权利要求1所述的调度方法，其特征在于，所述根据调整后的处理器的核数及类型，将所述子任务调度到各工作者线程，具体包括：4. The scheduling method according to claim 1, wherein the subtask is scheduled to each worker thread according to the adjusted core number and type of the processor, specifically comprising: 获取调整核数及类型后的处理器的状态，所述状态包括在线状态和离线状态；Obtain the state of the processor after adjusting the number and type of cores, the state includes an online state and an offline state; 获取调整核数及类型后的处理器的变动情况，所述变动情况包括处理器热移除以及处理器热插入；Obtain the changes of processors after adjusting the number and type of cores, the changes include processor hot removal and processor hot insertion; 根据所述处理器的状态以及变动情况，将所述子任务调度到各工作者线程。The subtask is scheduled to each worker thread according to the state and change of the processor. 5.根据权利要求1所述的调度方法，其特征在于，以先入先出的方式依次将所述子任务调度到各工作者线程。5 . The scheduling method according to claim 1 , wherein the subtasks are sequentially scheduled to each worker thread in a first-in, first-out manner. 6 . 6.一种功率自适应任务调度系统，其特征在于，所述系统包括：6. A power adaptive task scheduling system, wherein the system comprises: 任务信息获取模块，用于获取当前的任务信息；The task information acquisition module is used to obtain the current task information; 划分和统计模块，用于根据大数据编程模型MapReduce对所述任务信息进行划分和统计，得到多个可并行运行的子任务；The division and statistics module is used to divide and count the task information according to the big data programming model MapReduce, so as to obtain a plurality of subtasks that can be run in parallel; 状态信息获取模块，用于获取输出功率的变动状态以及任务负载状态，得到状态信息；The state information acquisition module is used to acquire the changing state of the output power and the task load state, and obtain the state information; 调整模块，用于根据所述状态信息调整处理器的核数及类型；an adjustment module, configured to adjust the number and type of cores of the processor according to the state information; 调度模块，用于根据调整核数及类型后的处理器，将所述子任务调度到各工作者线程。The scheduling module is configured to schedule the subtasks to each worker thread according to the adjusted core number and type of the processor. 7.根据权利要求6所述的调度系统，其特征在于，所述划分和统计模块具体包括：7. The scheduling system according to claim 6, wherein the division and statistics module specifically comprises: 划分单元，用于通过Map函数将所述任务信息进行划分；a dividing unit, used for dividing the task information by the Map function; 统计单元，用于通过Reduce函数将划分后的任务信息进行统计，得到多个可并行运行的子任务。The statistical unit is used to count the divided task information through the Reduce function to obtain multiple subtasks that can run in parallel. 8.根据权利要求6所述的调度系统，其特征在于，所述调整模块具体包括：8. The scheduling system according to claim 6, wherein the adjustment module specifically comprises: 工作功率需求获取单元，用于获取处理器的工作功率需求；a working power requirement obtaining unit, used to obtain the working power requirement of the processor; 调整单元，用于根据所述工作功率需求以及所述状态信息，利用CPU热插拔技术调整处理器的核数及类型。The adjustment unit is configured to adjust the number and type of processor cores by using the CPU hot-plug technology according to the working power requirement and the state information. 9.根据权利要求6所述的调度系统，其特征在于，所述调度模块具体包括：9. The scheduling system according to claim 6, wherein the scheduling module specifically comprises: 状态获取单元，用于获取调整核数及类型后的处理器的状态，所述状态包括在线状态和离线状态；a state acquisition unit, configured to acquire the state of the processor after adjusting the number and type of cores, and the state includes an online state and an offline state; 变动情况获取单元，用于获取调整核数及类型后的处理器的变动情况，所述变动情况包括处理器热移除以及处理器热插入；a change condition obtaining unit, used for obtaining the change condition of the processor after adjusting the number of cores and the type, and the change condition includes the hot removal of the processor and the hot insertion of the processor; 调度单元，用于根据所述处理器的状态以及变动情况，将所述子任务调度到各工作者线程。and a scheduling unit, configured to schedule the subtasks to each worker thread according to the state and change of the processor. 10.根据权利要求6所述的调度系统，其特征在于，所述调度模块以先入先出的方式依次将所述子任务调度到各工作者线程。10 . The scheduling system according to claim 6 , wherein the scheduling module sequentially schedules the subtasks to each worker thread in a first-in, first-out manner. 11 .
CN201810882835.2A 2018-08-06 2018-08-06 Power self-adaptive task scheduling method and system Active CN109144693B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810882835.2A CN109144693B (en) 2018-08-06 2018-08-06 Power self-adaptive task scheduling method and system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810882835.2A CN109144693B (en) 2018-08-06 2018-08-06 Power self-adaptive task scheduling method and system

Publications (2)

Publication Number Publication Date
CN109144693A true CN109144693A (en) 2019-01-04
CN109144693B CN109144693B (en) 2020-06-23

Family

ID=64791566

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810882835.2A Active CN109144693B (en) 2018-08-06 2018-08-06 Power self-adaptive task scheduling method and system

Country Status (1)

Country Link
CN (1) CN109144693B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110187966A (en) * 2019-05-13 2019-08-30 珠海全志科技股份有限公司 A kind of method, apparatus, system and program product locking CPU optimizing scheduling
CN113407313A (en) * 2020-11-27 2021-09-17 上海交通大学 Resource demand-aware multi-queue scheduling method, system and server

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103049245A (en) * 2012-10-25 2013-04-17 浪潮电子信息产业股份有限公司 Software performance optimization method based on central processing unit (CPU) multi-core platform
US20140149990A1 (en) * 2010-09-30 2014-05-29 International Business Machines Corporation Scheduling threads
CN104317658A (en) * 2014-10-17 2015-01-28 华中科技大学 MapReduce based load self-adaptive task scheduling method
US20150186184A1 (en) * 2013-12-26 2015-07-02 Electronics And Telecommunications Research Institute Apparatus and method for optimizing system performance of multi-core system
CN105700959A (en) * 2016-01-13 2016-06-22 南京邮电大学 Multi-core platform oriented multithreaded division and static balancing scheduling policy
CN107172149A (en) * 2017-05-16 2017-09-15 成都四象联创科技有限公司 Big data instant scheduling method

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140149990A1 (en) * 2010-09-30 2014-05-29 International Business Machines Corporation Scheduling threads
CN103049245A (en) * 2012-10-25 2013-04-17 浪潮电子信息产业股份有限公司 Software performance optimization method based on central processing unit (CPU) multi-core platform
US20150186184A1 (en) * 2013-12-26 2015-07-02 Electronics And Telecommunications Research Institute Apparatus and method for optimizing system performance of multi-core system
CN104317658A (en) * 2014-10-17 2015-01-28 华中科技大学 MapReduce based load self-adaptive task scheduling method
CN104317658B (en) * 2014-10-17 2018-06-12 华中科技大学 A kind of loaded self-adaptive method for scheduling task based on MapReduce
CN105700959A (en) * 2016-01-13 2016-06-22 南京邮电大学 Multi-core platform oriented multithreaded division and static balancing scheduling policy
CN107172149A (en) * 2017-05-16 2017-09-15 成都四象联创科技有限公司 Big data instant scheduling method

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
STEFANO CONOCI, ET AL.: "Adaptive Performance Optimization under Power Constraint in Multi-thread Applications with Diverse Scalability", 《ARXIV》 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110187966A (en) * 2019-05-13 2019-08-30 珠海全志科技股份有限公司 A kind of method, apparatus, system and program product locking CPU optimizing scheduling
CN113407313A (en) * 2020-11-27 2021-09-17 上海交通大学 Resource demand-aware multi-queue scheduling method, system and server
CN113407313B (en) * 2020-11-27 2022-05-17 上海交通大学 Resource demand-aware multi-queue scheduling method, system and server

Also Published As

Publication number Publication date
CN109144693B (en) 2020-06-23

Similar Documents

Publication Publication Date Title
CN105912401B (en) A kind of distributed data batch processing system and method
CN102360310B (en) Multitask process monitoring method in distributed system environment
CN102043675B (en) Thread pool management method based on task quantity of task processing request
CN105656973B (en) Method for scheduling task and system in a kind of distributed node group
CN102063336B (en) A Distributed Computing Multi-Application Function Asynchronous Concurrent Scheduling Method
Chen et al. Green-aware workload scheduling in geographically distributed data centers
Cheng et al. Heterogeneity-aware workload placement and migration in distributed sustainable datacenters
CN104808770B (en) Data center energy consumption management method and system based on dynamic frequency modulation
CN105302638B (en) MPP cluster task dispatching method based on system load
Hatzopoulos et al. Dynamic virtual machine allocation in cloud server facility systems with renewable energy sources
CN104317658A (en) MapReduce based load self-adaptive task scheduling method
CN104991830A (en) YARN resource allocation and energy-saving scheduling method and system based on service level agreement
CN108154317B (en) Workflow group scheduling method based on example self-adaptive distribution integration in multi-cloud environment
CN103399626B (en) Towards Parallel application dispatching system and the method for the power-aware of hybrid compute environment
CN104298550A (en) Hadoop-oriented dynamic scheduling method
CN102681889A (en) Scheduling method of cloud computing open platform
Fan Job scheduling in high performance computing
CN108574600A (en) Quality of Service Guarantee Method for Collaborative Control of Power Consumption and Resource Competition of Cloud Computing Servers
CN102339233A (en) Cloud computing centralized management platform
CN103888279A (en) Resource pool dynamic management method and system based on feedback control cycle
CN101685335A (en) Application server based on SEDA as well as energy-saving device and method thereof
CN108429784B (en) An energy efficiency priority cloud resource allocation and scheduling method
CN112148475B (en) Loongson big data integrated machine task scheduling method and system integrating load and power consumption
CN109144693A (en) A kind of power adaptive method for scheduling task and system
CN107370783A (en) A kind of dispatching method and device of cloud computing cluster resource

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
TR01 Transfer of patent right

Effective date of registration: 20220113

Address after: 200335 room 2547, floor 1, building 8, No. 33, Guangshun Road, Changning District, Shanghai

Patentee after: Shanghai Agriculture Information Co.,Ltd.

Address before: 200000 No. 999, Hucheng Ring Road, Lingang New Town, Pudong New Area, Shanghai

Patentee before: SHANGHAI OCEAN University

TR01 Transfer of patent right