CN109976890A - A kind of conversion method minimizing the privately owned cloud computing resources energy consumption of isomery - Google Patents
A kind of conversion method minimizing the privately owned cloud computing resources energy consumption of isomery Download PDFInfo
- Publication number
- CN109976890A CN109976890A CN201910240851.6A CN201910240851A CN109976890A CN 109976890 A CN109976890 A CN 109976890A CN 201910240851 A CN201910240851 A CN 201910240851A CN 109976890 A CN109976890 A CN 109976890A
- Authority
- CN
- China
- Prior art keywords
- task
- physical machine
- energy consumption
- sub
- copy
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/16—Error detection or correction of the data by redundancy in hardware
- G06F11/20—Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements
- G06F11/202—Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements where processing functionality is redundant
-
- 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
- G06F9/5038—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 considering the execution order of a plurality of tasks, e.g. taking priority or time dependency constraints into consideration
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Abstract
The invention discloses a kind of conversion methods for minimizing the privately owned cloud computing resources energy consumption of isomery, carry out task ranking to workflow application first, according to task schedule sequence, are followed successively by each task and divide sub- deadline；The physical machine in private clound is ranked up according to power dissipation ratio preferential principle again；Processor for task distribution adjusts frequency, so that task energy consumption under the premise of meeting the sub- off period is minimum, and determine whether task needs copy, according to copy amount, determines for copy task to be preferentially placed in execution in physical machine or newly open a physical machine execution copy task from remaining cloud service resource；Conversion method according to the present invention, the physical machine for distributing each task obtains the frequency of an optimization, fault-tolerant processing has been done on the transient fault influenced by DVFS, task number of copies is adjusted in the way of Task Duplication for workflow task, and energy optimization scheduling is had also been made to task copy, it is effectively reduced the energy consumption of the workflow application of off period constraint.
Description
Technical field
The invention belongs to cloud computing resources dispatching technique fields, and in particular to a kind of privately owned cloud computing resources of minimum isomery
The conversion method of energy consumption.
Background technique
Cloud computing is a kind of large-scale parallel and distributed computing model, by grid computing, parallel computation and effectiveness
The multiple technologies co-evolutions such as calculating, wherein private clound refers to a kind of special pattern of cloud computing, and IT service passes through special
It is configured with IT infrastructure, for single organisation specific.Private clound is usually managed by internal resource, and uniqueness is related to
And the environment based on cloud of safety, only specified client can be run.
Currently, energy consumption has become the maximum cost place of cloud computing center operation and maintenance.Although the fortune of computer equipment
Line efficiency increases, and the power consumption of global cloud computation data center is estimated will to increase 15%-20% every year.In view of current
Still based on thermal power generation, the atmosphere that the greenhouse gases that thermal power generation generates can live to us is brought greatly for the generation of electric energy
Destruction.It is estimated that the CO2 emissions at 2011 annual data centers are 78,700,000 tons, it is equivalent to the 2% of global emissions
, the energy consumption of cloud computation data center how is reduced also therefore as academia and industry urgent problem to be solved.
To the cloud computing center of heterogeneous resource, can be adjusted by improving the scheduling strategy of task, for example, task is preferential
Being assigned on the energy consumption virtual machine that better results and reducing energy consumption is common method.In addition to this, dynamic electric voltage and frequency scaling
(DVFS) technology is maintained at processor under the mode of low-power consumption and is run by dynamically reduction processor frequencies and voltage,
It is one of selectable method, it, can be to extend task execution time although reducing the decline that frequency band carrys out processor performance
Cost exchange lower power consumption for；On the other hand, processor frequencies and voltage, the incidence of calculate node transient fault are reduced
It can greatly increase, the reliability of task execution will reduce.The logic unit that a kind of high energy charged particles hit processor leads to list
Transient fault caused by particle effect, also referred to as soft error, and in cloud computing center, with the increasing of processor scale
Add, the problem of transient fault can become especially prominent, once some calculate node breaks down, if do not used suitable fault-tolerant
Technology, then being executed in the calculate node for task otherwise selection in other calculate nodes from the beginning execute or directly
Fail because mistake is executed, especially for workflow, the execution of task unsuccessfully can also be had an impact the execution of subsequent tasks.
So being very important for workflow design fault tolerant mechanism.
Workflow schedule problem itself is a kind of np hard problem, is one in cloud computing task schedule Related Research Domain
Difficult point.In recent years, researcher has carried out extensive research to workflow energy consumption scheduling problem, proposes the scheduling of a few thing stream
Algorithm, as the research hotspot of field of cloud calculation, the energy consumption minimized problem of workflow has done numerous studies by many scholars.?
In existing research, Tang et al. proposes the workflow schedule algorithm DEWTS perceived using the energy consumption of DVFS, and this method is directed to
The problem of workflow of deadline constraint realizes energy optimization, merges the relatively low place of efficiency by recycling free time
Device is managed, and reuses DVFS technology after server merging to utilize available free time.However, the above method but without
Method is in the variable ratio frequency changer computing resource of transient fault, the problem of realizing workflow energy optimization under the constraint of off period；Meanwhile
When carrying out the scheduling of workflow energy consumption perception using DVFS, transient error brought by DVFS is seldom considered, if thus having one kind
Method, which can combine, to be reduced energy consumption and considers transient fault problem, then will be to cloud computing resources dispatching technique field
One quantum jump and improvement.
Summary of the invention
The present invention is exactly directed to the problems of the prior art, provides a kind of privately owned cloud computing resources energy consumption of minimum isomery
Conversion method, comprising the following steps: task ranking is carried out to workflow application first and is followed successively by according to task schedule sequence
Each task divides sub- deadline；The physical machine in private clound is ranked up according to power dissipation ratio preferential principle again；To appoint
The processor of business distribution adjusts frequency, so that task energy consumption under the premise of meeting the sub- off period is minimum, and whether determines task
Copy is needed, according to copy amount, determines for copy task to be preferentially placed in and is executed in physical machine or from remaining cloud service resource
In newly open physical machine and execute copy task；Conversion method according to the present invention adjusts physical machine using DVFS technology
Cpu frequency, the physical machine for distributing each task obtain the frequency of an optimization, do on the transient fault influenced by DVFS
Fault-tolerant processing adjusts task number of copies in the way of Task Duplication for workflow task, and to task copy energy has also been made
Optimized Operation is consumed, the energy consumption of the workflow application of off period constraint is effectively reduced.
To achieve the goals above, the technical solution adopted by the present invention is that: a kind of privately owned cloud computing resources of minimum isomery
The conversion method of energy consumption, comprising the following steps:
S1 carries out task ranking to workflow application, according to task schedule sequence, when being followed successively by each task division son cut-off
Between；
S2 is ranked up the physical machine in private clound, is that the task in step S1 task sequence distributes physical machine, the distribution
Principle is that power dissipation ratio is preferential；
S3, the processor for task distribution adjust frequency, so that task energy consumption under the premise of meeting the sub- off period is minimum, and really
Determine whether task needs copy, if you do not need to copy, execution terminates；Otherwise, continue step S4；
S4, for task copy, the free time block of used physical machine before first search, if meeting placement condition,
Copy task is preferentially placed in physical machine and is executed；Otherwise, a physical machine is newly opened from remaining cloud service resource execute pair
This task.
As an improvement of the present invention, in the step S2, if task execution time is more than the sub- off period, preferential point
With the physical machine that the execution time is short.
It is improved as another kind of the invention, the step S1 further comprises:
S11 carries out task ranking to workflow application using HEFT algorithm, calculates the Upward Rank value of each task, and right
The Upward Rank value of each task carries out task ranking, and the task after sequence is placed in task queue and is saved；
S12, the slack time of calculation workflow application, the slack time of the workflow application are to answer with workflow deadline
The difference of minimum scheduling length；
S13 is calculated the Upward value of each task, is ranked up using the sequence successively decreased as workflow task；
S14 is each task distribution by time, the sub- deadline, according to as predecessor according to the sequence of step S13
Depth capacity of the being engaged in division proportional to workflow application depth capacity.
It is improved as another kind of the invention, the step S2 further comprises:
S21 successively takes out the task in task sequence according to the principle of prerequisite variable；
S22 is ranked up the physical machine in private clound according to power dissipation ratio preferential principle, and the high physical machine of power dissipation ratio is excellent
Task execution is first distributed to, if task execution time is more than the sub- off period, continues step S23；Otherwise, step terminates；
S23 is ranked up the physical machine in private clound, according to period preferential principle is executed, preferentially distributes to task and holds
Row time short physical machine, if it is more than the sub- off period that task execution time, which remains unchanged, return step S1 changes sub- deadline
Division mode, the division principle are the sub- off period of the task is extended to earliest finish time in assignable resource flat
Mean value；Otherwise, step terminates.
As another improvement of the invention, the step S3 further comprises:
S31 judges that the deadline of each task can or can not be beyond the sub- off period, if it is greater, then return step S1；Otherwise,
Continue to execute step S32；
S32 is calculated under the constraint for meeting physical machine execution reliability, and the frequency and task minimum of physical machine execute copy, hold
The corresponding relationship of row time and operation energy consumption, obtains mapping table；
S33 successively takes out the task in task sequence, if not having task in task sequence, side according to the principle of prerequisite variable
Method terminates；Otherwise, continue step S34；
S34 reduces processor frequencies according to mapping table in step S32, so that task is under the premise of meeting the sub- off period
Energy consumption is minimum, while determining task number of copies, if task number of copies is greater than 0, continues step S4；Otherwise, return step S33.
As another improvement of the invention, the step S4 further comprises:
S41, search have distributed in physical machine with the presence or absence of the free time block that can be placed, and if it exists, continue step S42；Otherwise,
Skip to step S43；
Task is preferentially assigned to the interval time shortest free time block placed by S42；
S43 opens a physical machine newly for task, preferentially by copy task be assigned to in the physical machine of main task same type, if
It is not present, is assigned in the high physical machine of power dissipation ratio in available resources in available resources；
S44 distributes physical machine for task.
As a further improvement of the present invention, free time block places condition in the step S4 are as follows: the beginning of task
At the beginning of time is greater than free time block, the sub- deadline of task is less than end time and the physical machine of free time block
The sum of transmission time.
Compared with prior art, the method for the present invention combines DVFS and Task Duplication technology, not only for physical machine isomorphism
Scene can more support isomery scene, in office to minimize workflow energy consumption as target from the angle of cloud service provider
Business level considers the Multi-workflow of hard off period constraint, considers the isomerism of variable ratio frequency changer virtual machine in data center in resource level
And the transient fault that may occur, by four didactic task schedules and resource frequency adjustable strategies, using DVFS's
Technology makes full use of the free time in scheduling process, and having advanced optimized workflow to extend the cost of task execution time exists
Execution energy consumption in private clound；Meanwhile being appointed for the transient fault problem of processor by way of Task Duplication for workflow
Business setting copy, carries out fault-tolerant processing, ensure that the reliable execution of workflow application.
Detailed description of the invention
Fig. 1 is that the present invention realizes the structure chart for minimizing the conversion method of the privately owned cloud computing resources energy consumption of isomery；
Fig. 2 is the step flow chart for the conversion method that the present invention minimizes the privately owned cloud computing resources energy consumption of isomery.
Specific embodiment
Below with reference to drawings and examples, the present invention is described in detail.
Embodiment 1
Present example realizes the structure chart of workflow application schedules and variable frequency adjustment method under the privately owned cloud environment of isomery, such as schemes
Shown in 1, including isomery private clound variable ratio frequency changer resource and workflow application, Service Source S={ S1,S2,…,SmIt is one comprising m
A function is identical, the different privately owned cloud service physical machine of working ability；Workflow list WL={ W to be processed1,W2,…,WnIt is one
It is a to include n workflow applications to be treated.
A kind of conversion method minimizing the privately owned cloud computing resources energy consumption of isomery, as shown in Figure 2, comprising the following steps:
S1 carries out task ranking to workflow application, according to task schedule sequence, when being followed successively by each task division son cut-off
Between；
S11 carries out task ranking to workflow application using HEFT algorithm, calculates the Upward Rank value of each task, and right
The Upward Rank value of each task carries out task ranking, and the task after sequence is placed in task queue and is saved；
S12, the slack time of calculation workflow application, the slack time of the workflow application are to answer with workflow deadline
The difference of minimum scheduling length；
S13 is calculated the Upward value of each task, is ranked up using the sequence successively decreased as workflow task；
S14 is distributed sub- deadline, the sub- deadline according to the sequence of step S13 for each task, according to as predecessor
Depth capacity of the being engaged in division proportional to workflow application depth capacity.
S2 is ranked up the physical machine in private clound, is that the task in step S1 task sequence distributes physical machine, described
Distribution principle is that power dissipation ratio is preferential；
S21 successively takes out the task in task sequence according to the principle of prerequisite variable；
S22 is ranked up the physical machine in private clound according to power dissipation ratio preferential principle, and the high physical machine of power dissipation ratio is excellent
Task execution is first distributed to, if task execution time is more than the sub- off period, continues step S23；Otherwise, step terminates；
S23 is ranked up the physical machine in private clound, according to period preferential principle is executed, preferentially distributes to task and holds
Row time short physical machine, if it is more than the sub- off period that task execution time, which remains unchanged, return step S1 changes sub- deadline
Division mode, the division principle are the sub- off period of the task is extended to earliest finish time in assignable resource flat
Mean value；Otherwise, step terminates.
S3, the processor for task distribution adjust frequency, so that task energy consumption under the premise of meeting the sub- off period is minimum,
And determine whether task needs copy, if you do not need to copy, execution terminates；Otherwise, continue step S4；
S31 judges that the deadline of each task can or can not be beyond the sub- off period, if it is greater, then return step S1；Otherwise,
Continue to execute step S32；
S32 is calculated under the constraint for meeting physical machine execution reliability, and the frequency and task minimum of physical machine execute copy, hold
The corresponding relationship of row time and operation energy consumption, obtains mapping table；
S33 successively takes out the task in task sequence, if not having task in task sequence, side according to the principle of prerequisite variable
Method terminates；Otherwise, continue step S34；
S34 reduces processor frequencies according to mapping table in step S32, so that task is under the premise of meeting the sub- off period
Energy consumption is minimum, while determining task number of copies, if task number of copies is greater than 0, continues step S4；Otherwise, return step S33.
S4, for task copy, the free time block of used physical machine before first search places item if met
Copy task is preferentially placed in physical machine and executes by part；Otherwise, a physical machine is newly opened from remaining cloud service resource to execute
Copy task；
S41, search have distributed in physical machine with the presence or absence of the free time block that can be placed, and if it exists, continue step S42；Otherwise,
Skip to step S43；
Task is preferentially assigned to the interval time shortest free time block placed by S42；
S43 opens a physical machine newly for task, preferentially by copy task be assigned to in the physical machine of main task same type, if
It is not present, is assigned in the high physical machine of power dissipation ratio in available resources in available resources；
S44 distributes physical machine for task.
Conversion method in the present invention is adjusted the cpu frequency of physical machine using DVFS technology, each task is made to distribute to obtain
Physical machine obtain one optimization frequency, fault-tolerant processing has been done on the transient fault influenced by DVFS, has utilized Task Duplication
Mode is that workflow task adjusts task number of copies, and energy optimization scheduling has also been made to task copy, is effectively reduced cut-off
The energy consumption of the workflow application of phase constraint.
The basic principles, main features and advantages of the present invention have been shown and described above.The technology of the industry
Personnel only illustrate the present invention it should be appreciated that the present invention is not limited by examples detailed above described in examples detailed above and specification
Principle, various changes and improvements may be made to the invention without departing from the spirit and scope of the present invention, these variation and
Improvement all fall within the protetion scope of the claimed invention.The claimed scope of the invention is by appended claims and its is equal
Object defines.
Claims (7)
1. a kind of conversion method for minimizing the privately owned cloud computing resources energy consumption of isomery, which comprises the following steps:
S1 carries out task ranking to workflow application, according to task schedule sequence, when being followed successively by each task division son cut-off
Between；
S2 is ranked up the physical machine in private clound, is that the task in step S1 task sequence distributes physical machine, the distribution
Principle is that power dissipation ratio is preferential；
S3, the processor for task distribution adjust frequency, so that task energy consumption under the premise of meeting the sub- off period is minimum, and really
Determine whether task needs copy, if you do not need to copy, execution terminates；Otherwise, continue step S4；
S4, for task copy, the free time block of used physical machine before first search, if meeting placement condition,
Copy task is preferentially placed in physical machine and is executed；Otherwise, a physical machine is newly opened from remaining cloud service resource execute pair
This task.
2. a kind of conversion method for minimizing the privately owned cloud computing resources energy consumption of isomery as described in claim 1, it is characterised in that
In the step S2, if task execution time is more than the sub- off period, preferential distribution executes time short physical machine.
3. a kind of conversion method for minimizing the privately owned cloud computing resources energy consumption of isomery as described in claim 1, it is characterised in that
The step S1 further comprises:
S11 carries out task ranking to workflow application using HEFT algorithm, calculates the Upward Rank value of each task, and right
The Upward Rank value of each task carries out task ranking, and the task after sequence is placed in task queue and is saved；
S12, the slack time of calculation workflow application, the slack time of the workflow application are to answer with workflow deadline
The difference of minimum scheduling length；
S13 is calculated the Upward value of each task, is ranked up using the sequence successively decreased as workflow task；
S14 is distributed sub- deadline, the sub- deadline according to the sequence of step S13 for each task, according to as predecessor
Depth capacity of the being engaged in division proportional to workflow application depth capacity.
4. a kind of conversion method for minimizing the privately owned cloud computing resources energy consumption of isomery as claimed in claim 2, it is characterised in that
The step S2 further comprises:
S21 successively takes out the task in task sequence according to the principle of prerequisite variable；
S22 is ranked up the physical machine in private clound according to power dissipation ratio preferential principle, and the high physical machine of power dissipation ratio is excellent
Task execution is first distributed to, if task execution time is more than the sub- off period, continues step S23；Otherwise, step terminates；
S23 is ranked up the physical machine in private clound, according to period preferential principle is executed, preferentially distributes to task and holds
Row time short physical machine, if it is more than the sub- off period that task execution time, which remains unchanged, return step S1 changes sub- deadline
Division mode, the division principle are the sub- off period of the task is extended to earliest finish time in assignable resource flat
Mean value；Otherwise, step terminates.
5. a kind of conversion method for minimizing the privately owned cloud computing resources energy consumption of isomery as described in claim 1, it is characterised in that
The step S3 further comprises:
S31 judges that the deadline of each task can or can not be beyond the sub- off period, if it is greater, then return step S1；Otherwise,
Continue to execute step S32；
S32 is calculated under the constraint for meeting physical machine execution reliability, and the frequency and task minimum of physical machine execute copy, hold
The corresponding relationship of row time and operation energy consumption, obtains mapping table；
S33 successively takes out the task in task sequence, if not having task in task sequence, side according to the principle of prerequisite variable
Method terminates；Otherwise, continue step S34；
S34 reduces processor frequencies according to mapping table in step S32, so that task is under the premise of meeting the sub- off period
Energy consumption is minimum, while determining task number of copies, if task number of copies is greater than 0, continues step S4；Otherwise, return step S33.
6. a kind of conversion method for minimizing the privately owned cloud computing resources energy consumption of isomery as described in claim 1, it is characterised in that
The step S4 further comprises:
S41, search have distributed in physical machine with the presence or absence of the free time block that can be placed, and if it exists, continue step S42；Otherwise,
Skip to step S43；
Task is preferentially assigned to the interval time shortest free time block placed by S42；
S43 opens a physical machine newly for task, preferentially by copy task be assigned to in the physical machine of main task same type, if
It is not present, is assigned in the high physical machine of power dissipation ratio in available resources in available resources；
S44 distributes physical machine for task.
7. a kind of conversion method for minimizing the privately owned cloud computing resources energy consumption of isomery as described in claim 1 or 6, feature exist
Free time block places condition in the step S4 are as follows: at the beginning of being greater than free time block at the beginning of task, appoints
The sub- deadline of business is less than the sum of end time and transmission time of physical machine of free time block.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910240851.6A CN109976890B (en) | 2019-03-28 | 2019-03-28 | Variable frequency method for minimizing heterogeneous private cloud computing resource energy consumption |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910240851.6A CN109976890B (en) | 2019-03-28 | 2019-03-28 | Variable frequency method for minimizing heterogeneous private cloud computing resource energy consumption |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109976890A true CN109976890A (en) | 2019-07-05 |
| CN109976890B CN109976890B (en) | 2023-05-30 |
Family
ID=67081076
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910240851.6A Active CN109976890B (en) | 2019-03-28 | 2019-03-28 | Variable frequency method for minimizing heterogeneous private cloud computing resource energy consumption |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109976890B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110780974A (en) * | 2019-09-10 | 2020-02-11 | 杭州电子科技大学 | Workflow-oriented fault-tolerant scheduling method in mobile edge computing environment |
| CN117349029A (en) * | 2023-12-04 | 2024-01-05 | 浪潮电子信息产业股份有限公司 | Heterogeneous computing system, energy consumption determining method and device, electronic equipment and medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8583467B1 (en) * | 2012-08-23 | 2013-11-12 | Fmr Llc | Method and system for optimized scheduling of workflows |
| CN103970613A (en) * | 2014-05-21 | 2014-08-06 | 哈尔滨工程大学 | Multi-copy task fault tolerance scheduling method of heterogeneous distributed system |
| CN104102544A (en) * | 2014-06-30 | 2014-10-15 | 武汉理工大学 | Multi QoS (quality of service)-constrained parallel task scheduling cost optimizing method under mixed cloud environment |
| CN106934537A (en) * | 2017-03-02 | 2017-07-07 | 北京工业大学 | The sub- time limit based on the scheduling of reverse operation stream obtains optimization method |
| CN107908464A (en) * | 2017-10-12 | 2018-04-13 | 华东师范大学 | A kind of cloud computing workflow energy-saving scheduling method for considering reliability |
| CN108108233A (en) * | 2017-11-29 | 2018-06-01 | 上海交通大学 | The cluster job scheduling method and system that the more copies of task perform |
-
2019
- 2019-03-28 CN CN201910240851.6A patent/CN109976890B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8583467B1 (en) * | 2012-08-23 | 2013-11-12 | Fmr Llc | Method and system for optimized scheduling of workflows |
| CN103970613A (en) * | 2014-05-21 | 2014-08-06 | 哈尔滨工程大学 | Multi-copy task fault tolerance scheduling method of heterogeneous distributed system |
| CN104102544A (en) * | 2014-06-30 | 2014-10-15 | 武汉理工大学 | Multi QoS (quality of service)-constrained parallel task scheduling cost optimizing method under mixed cloud environment |
| CN106934537A (en) * | 2017-03-02 | 2017-07-07 | 北京工业大学 | The sub- time limit based on the scheduling of reverse operation stream obtains optimization method |
| CN107908464A (en) * | 2017-10-12 | 2018-04-13 | 华东师范大学 | A kind of cloud computing workflow energy-saving scheduling method for considering reliability |
| CN108108233A (en) * | 2017-11-29 | 2018-06-01 | 上海交通大学 | The cluster job scheduling method and system that the more copies of task perform |
Non-Patent Citations (1)
| Title |
|---|
| 李廷元等: "QoS约束云环境下的工作流能效调度算法", 《计算机科学》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110780974A (en) * | 2019-09-10 | 2020-02-11 | 杭州电子科技大学 | Workflow-oriented fault-tolerant scheduling method in mobile edge computing environment |
| CN110780974B (en) * | 2019-09-10 | 2022-03-01 | 杭州电子科技大学 | Workflow-oriented fault-tolerant scheduling method in mobile edge computing environment |
| CN117349029A (en) * | 2023-12-04 | 2024-01-05 | 浪潮电子信息产业股份有限公司 | Heterogeneous computing system, energy consumption determining method and device, electronic equipment and medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109976890B (en) | 2023-05-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Katal et al. | Energy efficiency in cloud computing data centers: a survey on software technologies | |
| Calheiros et al. | Energy-efficient scheduling of urgent bag-of-tasks applications in clouds through DVFS | |
| Yang et al. | Stackelberg game approach for energy-aware resource allocation in data centers | |
| Jansen et al. | Energy efficient virtual machine allocation in the cloud | |
| Garg et al. | Energy-efficient scheduling of HPC applications in cloud computing environments | |
| CN108170530B (en) | Hadoop load balancing task scheduling method based on mixed element heuristic algorithm | |
| Mehdi et al. | Minimum completion time for power-aware scheduling in cloud computing | |
| Liu et al. | A survey on virtual machine scheduling in cloud computing | |
| Kaur et al. | Optimization techniques for resource provisioning and load balancing in cloud environment: a review | |
| Liang et al. | A low-power task scheduling algorithm for heterogeneous cloud computing | |
| Cao et al. | Energy efficient workflow job scheduling for green cloud | |
| Barzegar et al. | EATSDCD: A green energy-aware scheduling algorithm for parallel task-based application using clustering, duplication and DVFS technique in cloud datacenters | |
| CN109976890A (en) | A kind of conversion method minimizing the privately owned cloud computing resources energy consumption of isomery | |
| Kang et al. | A SLA driven VM auto-scaling method in hybrid cloud environment | |
| He et al. | Energy-efficient framework for virtual machine consolidation in cloud data centers | |
| Feng et al. | Towards heat-recirculation-aware virtual machine placement in data centers | |
| Xu et al. | Energy-Aware Resource Management in Cloud Computing Considering Load Balance. | |
| Kulseitova et al. | A survey of energy-efficient techniques in cloud data centers | |
| Sukprasert et al. | Quantifying the benefits of carbon-aware temporal and spatial workload shifting in the cloud | |
| Abba et al. | Deadline based performance evaluation of job scheduling algorithms | |
| Garg et al. | Exploiting heterogeneity in grid computing for energy-efficient resource allocation | |
| Ghoreyshi | Energy-efficient resource management of cloud datacenters under fault tolerance constraints | |
| Zhang et al. | Reliability enhancement strategies for workflow scheduling under energy consumption constraints in clouds | |
| Shah et al. | Modified least cost method for grid resource allocation | |
| Amoon | A multi criteria-based approach for virtual machines consolidation to save electrical power in cloud data centers |
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 |