CN106445661A - Dynamic optimization method and system - Google Patents
Dynamic optimization method and system Download PDFInfo
- Publication number
- CN106445661A CN106445661A CN201610821159.9A CN201610821159A CN106445661A CN 106445661 A CN106445661 A CN 106445661A CN 201610821159 A CN201610821159 A CN 201610821159A CN 106445661 A CN106445661 A CN 106445661A
- Authority
- CN
- China
- Prior art keywords
- optimized
- task
- optimization
- memory
- intensive
- 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
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
Abstract
The invention discloses a dynamic optimization method and system. The method comprises the following steps: obtaining running information of all running tasks and selecting the to-be-optimized tasks with performance bottlenecks from all the tasks based on the running information; optimizing the to-be-optimized tasks by utilizing a pre-optimization principle input from the outside; obtaining optimized running information of the optimized to-be-optimized tasks and returning the corresponding to-be-optimized tasks to states before optimization when the optimized running information does not meet the preset requirements. Through the technical scheme disclosed by the invention, the running information of all the running tasks is obtained, and the to-be-optimized tasks are determined based on the running information; the to-be-optimized tasks are optimized, so that dynamic obtaining of the running information of all the running tasks is realized, dynamic optimization of the tasks is further realized, the effectiveness of task optimization is ensured, and efficient running of the various tasks in the operating system is ensured.
Description
Technical field
The present invention relates to operation system technology field, more particularly, it relates to a kind of dynamic optimization method and system.
Background technology
Operating system (Operating System, abbreviation OS) is management and the meter of control computer hardware and software resource
Calculation machine program, is the most basic systems soft ware running directly on " bare machine ", any other software all must be in operating system
Support under could run.
In order to ensure the normally and efficiently operation of operating system it is often necessary to be optimized to operating in task therein.Existing
Having in technology is typically required before task run for the optimization of task, by taking feelings to existing resource in operating system
Condition and treat that the relevant information of operation task determines the need for this task is optimized and how this task is optimized.
But, due to task in running it may happen that some changes, and above-mentioned static optimal way cannot know this
Change, more cannot make corresponding Optimal Decision-making for this change, therefore, lead to have for the optimization of task in operating system
Effect property is poor.
In sum, there is poor the asking of effect of optimization for the optimal way of task in operating system in prior art
Topic.
Content of the invention
It is an object of the invention to provide a kind of dynamic optimization method and system, to solve in prior art for operating system
The poor problem of effect of optimization that the optimal way of middle task exists.
To achieve these goals, the present invention provides following technical scheme：
A kind of dynamic optimization method, including：
Obtain the operation information of each task in being currently running, and selected by each task described based on this operation information
Take out the task to be optimized with performance bottleneck；
Using the optimization principles being inputted by the external world in advance, described task to be optimized is optimized；
Obtain the optimization operation information of the task to be optimized being optimized, and do not meet in this optimization operation information default
During requirement, corresponding task to be optimized is back to the state before being optimized.
Preferably, obtain the operation information of each task in being currently running, including：
Obtain the CPU usage of each task in being currently running, memory usage and IO occupancy；
Corresponding, the task to be optimized with performance bottleneck is selected by each task described based on this operation information,
Including：
Determine described CPU usage be more than CPU threshold value and/or described memory usage be more than corresponding memory threshold and/or
The task that described IO occupancy is more than I/O threshold is described task to be optimized.
Preferably, after selecting described task to be optimized, also include：
Described task to be optimized is divided into computation-intensive, memory-intensive and I/O intensive type；
Corresponding, using the optimization principles being inputted by the external world in advance, described task to be optimized is optimized, including：
According in described optimization principles respectively with described computation-intensive, memory-intensive and the corresponding optimization of I/O intensive type
The principle task intensive to this is optimized.
Preferably, described task to be optimized is divided into computation-intensive, memory-intensive and I/O intensive type, including：
For arbitrary task to be optimized, determine that its CPU usage and the difference of CPU threshold value and the ratio of CPU threshold value are the
One ratio, memory usage and the difference of memory threshold are the second ratio, IO occupancy and I/O threshold with the ratio of memory threshold
The ratio of difference and I/O threshold be the 3rd ratio, if the first ratio is maximum, this task to be optimized is computation-intensive, such as
Really the second ratio is maximum, then this task to be optimized is memory-intensive, if the 3rd ratio is maximum, this task to be optimized is IO
Intensive.
Preferably, judge whether the optimization operation information of the task to be optimized being optimized meets preset requirement, including：
The optimization operation information of described task to be optimized includes CPU usage, memory usage and the IO of its current time
During occupancy, if the CPU usage of this task current time to be optimized is less than or equal to CPU threshold value and memory usage is less than
Or be equal to memory threshold and IO occupancy and be less than or equal to I/O threshold it is determined that this task to be optimized meets preset requirement；Otherwise,
Then determine that this task to be optimized does not meet preset requirement.
A kind of dynamic optimization system, including：
Choose module, for obtaining the operation information of each task in being currently running, and based on this operation information by institute
State and in each task, select the task to be optimized with performance bottleneck；
Optimization module, for being optimized to described task to be optimized using the optimization principles being inputted by the external world in advance；
Judging module, for obtaining the optimization operation information of the task to be optimized being optimized, and runs in this optimization
Information does not meet, during preset requirement, corresponding task to be optimized is back to the state before being optimized.
Preferably, described selection module includes：
Acquiring unit, CPU usage, memory usage and IO for obtaining each task in being currently running take
Rate；
Choose unit, for determining that described CPU usage is more than CPU threshold value and/or described memory usage is more than correspondence
Memory threshold and/or described IO occupancy are described task to be optimized more than the task of I/O threshold.
Preferably, also include：
Sort module, for being divided into computation-intensive, memory-intensive and I/O intensive type by described task to be optimized；
Corresponding, described optimization module includes：
Optimize unit, for according to close with described computation-intensive, memory-intensive and IO respectively in described optimization principles
The corresponding optimization principles of the collection type task intensive to this is optimized.
Preferably, described sort module includes：
Taxon, for for arbitrary task to be optimized, determining difference and the CPU threshold of its CPU usage and CPU threshold value
The ratio of value is the first ratio, and memory usage and the difference of memory threshold are the second ratio with the ratio of memory threshold, and IO accounts for
It is the 3rd ratio with the difference of rate and I/O threshold and the ratio of I/O threshold, if the first ratio is maximum, this task to be optimized is
Computation-intensive, if the second ratio is maximum, this task to be optimized is memory-intensive, if the 3rd ratio is maximum, should
Task to be optimized is I/O intensive type.
Preferably, described judging module includes：
Decision unit, the operation information that optimizes for described task to be optimized includes the CPU usage of its current time, interior
When depositing occupancy and IO occupancy, if the CPU usage of this task current time to be optimized is less than or equal to CPU threshold value and interior
Deposit occupancy be less than or equal to memory threshold and IO occupancy be less than or equal to I/O threshold it is determined that this task to be optimized meet pre-
If requiring；Otherwise it is determined that this task to be optimized does not meet preset requirement.
The invention provides a kind of dynamic optimization method and system, wherein the method include：Obtain each in being currently running
The operation information of individual task, and there is to be optimized of performance bottleneck based on this operation information by selecting in each task described
Business；Using the optimization principles being inputted by the external world in advance, described task to be optimized is optimized；Treating that acquisition has been optimized is excellent
The optimization operation information of change task, and when this optimization operation information does not meet preset requirement, corresponding task to be optimized is back to
State before being optimized.By technique scheme disclosed in the present application, obtain the operation letter of each task in being currently running
Breath, and task to be optimized is determined based on this operation information, this task to be optimized is optimized, it is achieved thereby that for each
The dynamic access of the operation information of task, and then realize dynamic optimization for task it is ensured that effective for task optimization
Property it is ensured that in operating system each task Effec-tive Function.In addition, when the optimization to task to be optimized is unsuccessful, will treat
Optimization task is back to the state before optimization, it can be avoided that the waste of spare resources is it is achieved that the reasonable utilization of resource.
Brief description
In order to be illustrated more clearly that the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
Have technology description in required use accompanying drawing be briefly described it should be apparent that, drawings in the following description be only this
Inventive embodiment, for those of ordinary skill in the art, on the premise of not paying creative work, can also basis
The accompanying drawing providing obtains other accompanying drawings.
Fig. 1 is a kind of flow chart of dynamic optimization method provided in an embodiment of the present invention；
Fig. 2 is a kind of structural representation of dynamic optimization system provided in an embodiment of the present invention.
Specific embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation description is it is clear that described embodiment is only a part of embodiment of the present invention, rather than whole embodiments.It is based on
Embodiment in the present invention, it is every other that those of ordinary skill in the art are obtained under the premise of not making creative work
Embodiment, broadly falls into the scope of protection of the invention.
Refer to Fig. 1, the flow chart that it illustrates a kind of dynamic optimization method provided in an embodiment of the present invention, can include
Following steps：
S11：Obtain the operation information of each task in being currently running, and selected by each task based on this operation information
Take out the task to be optimized with performance bottleneck.
Wherein, the operation information of each task being currently running can include this task in running to operating system
The occupancy situation of middle items resource；When based on each task to operating system in every resource occupancy situation determine wherein certain
The larger situation of occupation proportion in a little resources and this situation can lead to running using this resource of task at aspects such as speed
When being adversely affected, illustrate that running using this resource of task is the task to be optimized with performance bottleneck, at this time, it may be necessary to right
It is optimized, to avoid it to be subject to above-mentioned harmful effect.
Specifically, above-mentioned realize process can by the instrument installed in operating system (as vmstat, top,
Iostat, perf etc.) obtain each task above-mentioned operation information, will have to be optimized of performance bottleneck based on this operation information
Business
S12：Using the optimization principles being inputted by the external world in advance, task to be optimized is optimized.
Wherein, optimization principles can be write according to certain rule according to its experience by staff and be obtained, and specifically may be used
With optimization principles are inputted and store to corresponding optimization file (can also be database, specifically can be according to actual needs
It is determined) in, thus, can achieve the optimization treating optimization task based on optimization principles, thus ensure that task to be optimized
Effec-tive Function.Specifically, the optimization for task to be optimized can include two suboptimization, is once User space optimization, and one
Kernel state optimization when secondary, naturally it is also possible to only carry out an optimization in above-mentioned two, specifically can be carried out according to actual needs
Determine.
S13：Obtain the optimization operation information of the task to be optimized being optimized, and do not meet in this optimization operation information
During preset requirement, corresponding task to be optimized is back to the state before being optimized.
After task to be optimized is optimised, need to monitor whether this task obtains real optimization.Now only need to obtain
Take the current optimization operation information of optimized task to be optimized, with the difference of above-mentioned operation information, this information is that it is
For the information obtaining of running of optimized task current time to be optimized, excellent by knowing to these information
The occupation condition of the task to be optimized changed, and then determine whether there is some resources and occur that occupation proportion is larger and this leads to
Situation about being adversely affected at aspects such as speed using the task to be optimized that this resource is run, if it is present illustrate that this is treated
Optimization task does not obtain real optimization, only need to return it to the state before not optimizing, if it does not exist, then explanation
This task to be optimized has obtained real optimization, continues to carry out running according to the mode after optimizing.
By technique scheme disclosed in the present application, obtain the operation information of each task in being currently running, and be based on
This operation information determines task to be optimized, and this task to be optimized is optimized, it is achieved thereby that the fortune for each task
The dynamic access of row information, so realize dynamic optimization for task it is ensured that effectiveness for task optimization it is ensured that
The Effec-tive Function of each task in operating system.In addition, when the optimization to task to be optimized is unsuccessful, task to be optimized is returned
It is back to the state before optimization, it can be avoided that the waste of spare resources is it is achieved that the reasonable utilization of resource.
A kind of dynamic optimization method provided in an embodiment of the present invention, obtains the operation letter of each task in being currently running
Breath, can include：
Obtain the CPU usage of each task in being currently running, memory usage and IO occupancy；
Corresponding, based on this operation information by selecting the task to be optimized with performance bottleneck in each task, including：
Determine that CPU usage is more than CPU threshold value and/or memory usage is more than corresponding memory threshold and/or IO occupancy
It is task to be optimized more than the task of I/O threshold.
Wherein, CPU usage, memory usage and IO occupancy are consistent with corresponding concept of the prior art, and here is not
Repeat again；And CPU threshold value, memory threshold and I/O threshold can be determined according to actual needs.When the determination CPU of task takies
Rate be more than CPU threshold value and/or memory usage be more than corresponding memory threshold and/or IO occupancy be more than during I/O threshold it is easy to
The problems such as rate reduction, at this time, it may be necessary to be optimized to this task.
A kind of dynamic optimization method provided in an embodiment of the present invention, after selecting task to be optimized, can also include：
Task to be optimized is divided into computation-intensive, memory-intensive and I/O intensive type.
Wherein, the division for task to be optimized can take according to its corresponding CPU usage, memory usage and IO
Rate is realized, and specifically can arrange corresponding rule according to actual needs by staff, here is not specifically limited.Simply come
Say, computation-intensive corresponds to CPU usage, memory-intensive corresponds to memory usage, I/O intensive type corresponds to IO occupancy.Specifically
For, task to be optimized is divided into computation-intensive, memory-intensive and I/O intensive type, can include：For arbitrary treat excellent
Change task, determines that the difference of its CPU usage and CPU threshold value is the first ratio with the ratio of CPU threshold value, memory usage with interior
The ratio of the difference and memory threshold of depositing threshold value is the second ratio, and IO occupancy with the difference of I/O threshold with the ratio of I/O threshold is
3rd ratio, if the first ratio is maximum, this task to be optimized is computation-intensive, if the second ratio is maximum, this is treated
Optimization task is memory-intensive, if the 3rd ratio is maximum, this task to be optimized is I/O intensive type.Hereby it is achieved that treating excellent
Change task corresponds to different types of division.
Corresponding, using the optimization principles being inputted by the external world in advance, task to be optimized is optimized, can include：
According in optimization principles respectively with computation-intensive, memory-intensive and the corresponding optimization principles of I/O intensive type to this
Intensive task is optimized.
Wherein, can there are different principles corresponding to different types in optimization principles, can carry out according to actual needs
Determine, specifically, for computation-intensive, following optimization principles can be included：
1st, binding node (CPU can be refine to)：By on task immigration to be optimized to single node (CPU), it is somebody's turn to do with utilizing
Individually node (CPU) realizes the operation of this task to be optimized；
2nd, resource is specified by cgroup：Distribute corresponding cpu resource for task to be optimized；
3rd, pass through the parameter regulation below proc：Task distribution to be optimized is embodied as by the regulation of these parameters rational
Cpu resource；
4th, adjust the priority of process；Distribute corresponding priority for different tasks to be optimized, treated with high to priority
Optimize priority of task to process；
5th, balance is interrupted：Interruption on current CPU is migrated to other CPU, to reduce this CPU usage.
Thus improve the operating rate of corresponding task by such scheme.Certainly can also other be set according to actual needs
Concrete optimization principles, and adopt which kind of optimization principles can be by staff according to the work of this personnel to be optimized in the specific implementation
Carry out actual determination as scene, and write in optimization principles.
For I/O intensive type, following optimization principles can be included：
1st, on the clump that task immigration to be optimized is located to corresponding I/O device；
2nd, pass through to adjust related with I/O transfer parameter below proc, such as block device queue etc. is task distribution to be optimized
Corresponding I/O resource；
3rd, pass through to adjust disk subsystem (I/O dispatches, and file system selects etc.) or (network interface card is tied up to adjust network subsystem
Fixed, increase meshwork buffering area etc.) distribute corresponding I/O resource for task to be optimized.
Thus improve the operating rate of corresponding task by such scheme.Certainly can also other be set according to actual needs
Concrete optimization principles, and adopt which kind of optimization principles can be by staff according to the work of this personnel to be optimized in the specific implementation
Carry out actual determination as scene, and write in optimization principles.
For memory-intensive, following optimization principles can be included：
1st, big page mode：Memory setting is comprised data volume for every page and is more than given threshold according to actual needs；
2nd, Memory recycle parameter：Occupied internal memory is reclaimed according to the rule setting according to actual needs；
3rd, adjust swap：Partial data in internal memory is written in swap.
Thus improve the operating rate of corresponding task by such scheme.Certainly can also other be set according to actual needs
Concrete optimization principles, and adopt which kind of optimization principles can be by staff according to the work of this personnel to be optimized in the specific implementation
Carry out actual determination as scene, and write in optimization principles.
A kind of dynamic optimization method provided in an embodiment of the present invention, judges that the optimization of the task to be optimized being optimized is transported
Whether row information meets preset requirement, can include：
CPU usage, memory usage and IO that the optimization operation information of task to be optimized includes its current time take
During rate, if the CPU usage of this task current time to be optimized is less than or equal to CPU threshold value and memory usage is less than or waits
In memory threshold and IO occupancy be less than or equal to I/O threshold it is determined that this task to be optimized meets preset requirement；Otherwise, then really
This task to be optimized fixed does not meet preset requirement.
When above-mentioned items are respectively less than or are equal to corresponding threshold value, illustrate that occupation condition has reached a comparison reasonable
State, therefore, the result whether task to be optimized is really optimized can accurately be drawn by above-mentioned judgement.
In addition, in technique scheme provided in an embodiment of the present invention, staff can be according to actual needs any
Time adjustment or increase or minimizing optimization principles are such that it is able to enable optimization principles to meet actual feelings to the full extent
Condition is it is ensured that effective optimization for task to be optimized.
The embodiment of the present invention additionally provides a kind of dynamic optimization system, as shown in Fig. 2 can include：
Choose module 11, for obtain be currently running in each task operation information, and based on this operation information by
The task to be optimized with performance bottleneck is selected in each task；
Optimization module 12, for being optimized to task to be optimized using the optimization principles being inputted by the external world in advance；
Judging module 13, for obtaining the optimization operation information of the task to be optimized being optimized, and in this optimization fortune
Row information does not meet, during preset requirement, corresponding task to be optimized is back to the state before being optimized.
A kind of dynamic optimization system provided in an embodiment of the present invention, choosing module can include：
Acquiring unit, CPU usage, memory usage and IO for obtaining each task in being currently running take
Rate；
Choose unit, for determining that CPU usage is more than CPU threshold value and/or memory usage is more than corresponding memory threshold
And/or IO occupancy is task to be optimized more than the task of I/O threshold.
A kind of dynamic optimization system provided in an embodiment of the present invention, can also include：
Sort module, for being divided into computation-intensive, memory-intensive and I/O intensive type by task to be optimized；
Corresponding, optimization module includes：
Optimize unit, for according to corresponding with computation-intensive, memory-intensive and I/O intensive type respectively in optimization principles
The optimization principles task intensive to this be optimized.
A kind of dynamic optimization system provided in an embodiment of the present invention, sort module can include：
Taxon, for for arbitrary task to be optimized, determining difference and the CPU threshold of its CPU usage and CPU threshold value
The ratio of value is the first ratio, and memory usage and the difference of memory threshold are the second ratio with the ratio of memory threshold, and IO accounts for
It is the 3rd ratio with the difference of rate and I/O threshold and the ratio of I/O threshold, if the first ratio is maximum, this task to be optimized is
Computation-intensive, if the second ratio is maximum, this task to be optimized is memory-intensive, if the 3rd ratio is maximum, should
Task to be optimized is I/O intensive type.
A kind of dynamic optimization system provided in an embodiment of the present invention, judging module can include：
Decision unit, for task to be optimized optimize operation information includes the CPU usage of its current time, internal memory accounts for
During with rate and IO occupancy, if the CPU usage of this task current time to be optimized is less than or equal to CPU threshold value and internal memory accounts for
It is less than or equal to memory threshold with rate and IO occupancy is less than or equal to I/O threshold it is determined that this task to be optimized meets default wanting
Ask；Otherwise it is determined that this task to be optimized does not meet preset requirement.
In a kind of dynamic optimization system provided in an embodiment of the present invention, the explanation of relevant portion refers to the embodiment of the present invention
In a kind of dynamic optimization method providing, the detailed description of corresponding part, will not be described here.
Described above to the disclosed embodiments, makes those skilled in the art be capable of or uses the present invention.To this
Multiple modifications of a little embodiments will be apparent from for a person skilled in the art, and generic principles defined herein can
Without departing from the spirit or scope of the present invention, to realize in other embodiments.Therefore, the present invention will not be limited
It is formed on the embodiments shown herein, and be to fit to consistent with principles disclosed herein and features of novelty the widest
Scope.
Claims (10)
1. a kind of dynamic optimization method is it is characterised in that include：
Obtain the operation information of each task in being currently running, and selected by each task described based on this operation information
There is the task to be optimized of performance bottleneck；
Using the optimization principles being inputted by the external world in advance, described task to be optimized is optimized；
Obtain the optimization operation information of the task to be optimized being optimized, and do not meet preset requirement in this optimization operation information
When corresponding task to be optimized is back to the state before being optimized.
2. method according to claim 1 it is characterised in that obtain be currently running in each task operation information,
Including：
Obtain the CPU usage of each task in being currently running, memory usage and IO occupancy；
Corresponding, the task to be optimized with performance bottleneck is selected by each task described based on this operation information, including：
Determine that described CPU usage is more than CPU threshold value and/or described memory usage and is more than corresponding memory threshold and/or described
The task that IO occupancy is more than I/O threshold is described task to be optimized.
3., after method according to claim 2 is it is characterised in that select described task to be optimized, also include：
Described task to be optimized is divided into computation-intensive, memory-intensive and I/O intensive type；
Corresponding, using the optimization principles being inputted by the external world in advance, described task to be optimized is optimized, including：
According in described optimization principles respectively with described computation-intensive, memory-intensive and the corresponding optimization principles of I/O intensive type
To this, intensive task is optimized.
4. method according to claim 3 is it is characterised in that be divided into computation-intensive, interior by described task to be optimized
Deposit intensity and I/O intensive type, including：
For arbitrary task to be optimized, determine that its CPU usage and the difference of CPU threshold value and the ratio of CPU threshold value are the first ratio
Value, memory usage and the difference of memory threshold are the second ratio with the ratio of memory threshold, the difference of IO occupancy and I/O threshold
Value is the 3rd ratio with the ratio of I/O threshold, if the first ratio is maximum, this task to be optimized is computation-intensive, if the
Two ratios are maximum, then this task to be optimized is memory-intensive, if the 3rd ratio is maximum, this task to be optimized is I/O intensive
Type.
5. method according to claim 2 is it is characterised in that judge that the optimization of task to be optimized being optimized runs
Whether information meets preset requirement, including：
CPU usage, memory usage and IO that the optimization operation information of described task to be optimized includes its current time take
During rate, if the CPU usage of this task current time to be optimized is less than or equal to CPU threshold value and memory usage is less than or waits
In memory threshold and IO occupancy be less than or equal to I/O threshold it is determined that this task to be optimized meets preset requirement；Otherwise, then really
This task to be optimized fixed does not meet preset requirement.
6. a kind of dynamic optimization system is it is characterised in that include：
Choose module, for obtaining the operation information of each task in being currently running, and based on this operation information by described each
The task to be optimized with performance bottleneck is selected in individual task；
Optimization module, for being optimized to described task to be optimized using the optimization principles being inputted by the external world in advance；
Judging module, for obtaining the optimization operation information of the task to be optimized being optimized, and in this optimization operation information
Do not meet, during preset requirement, corresponding task to be optimized is back to the state before being optimized.
7. system according to claim 6 is it is characterised in that described selection module includes：
Acquiring unit, for obtaining CPU usage, memory usage and the IO occupancy of each task in being currently running；
Choose unit, for determining that described CPU usage is more than CPU threshold value and/or described memory usage is more than corresponding internal memory
Threshold value and/or described IO occupancy are described task to be optimized more than the task of I/O threshold.
8. system according to claim 7 is it is characterised in that also include：
Sort module, for being divided into computation-intensive, memory-intensive and I/O intensive type by described task to be optimized；
Corresponding, described optimization module includes：
Optimize unit, for according in described optimization principles respectively with described computation-intensive, memory-intensive and I/O intensive type
The corresponding optimization principles task intensive to this is optimized.
9. system according to claim 8 is it is characterised in that described sort module includes：
Taxon, for for arbitrary task to be optimized, determining difference and the CPU threshold value of its CPU usage and CPU threshold value
Ratio is the first ratio, and memory usage and the difference of memory threshold are the second ratio with the ratio of memory threshold, IO occupancy
It is the 3rd ratio with the difference of I/O threshold and the ratio of I/O threshold, if the first ratio is maximum, this task to be optimized is to calculate
Intensity, if the second ratio is maximum, this task to be optimized is memory-intensive, if the 3rd ratio is maximum, this treats excellent
Change task is I/O intensive type.
10. system according to claim 7 is it is characterised in that described judging module includes：
Decision unit, for described task to be optimized optimize operation information includes the CPU usage of its current time, internal memory accounts for
During with rate and IO occupancy, if the CPU usage of this task current time to be optimized is less than or equal to CPU threshold value and internal memory accounts for
It is less than or equal to memory threshold with rate and IO occupancy is less than or equal to I/O threshold it is determined that this task to be optimized meets default wanting
Ask；Otherwise it is determined that this task to be optimized does not meet preset requirement.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610821159.9A CN106445661A (en) | 2016-09-13 | 2016-09-13 | Dynamic optimization method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610821159.9A CN106445661A (en) | 2016-09-13 | 2016-09-13 | Dynamic optimization method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106445661A true CN106445661A (en) | 2017-02-22 |
Family
ID=58168084
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610821159.9A Pending CN106445661A (en) | 2016-09-13 | 2016-09-13 | Dynamic optimization method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106445661A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110750346A (en) * | 2019-10-17 | 2020-02-04 | Oppo（重庆）智能科技有限公司 | Task operation optimization method, device, terminal and storage medium |
| CN116991592A (en) * | 2023-09-26 | 2023-11-03 | 中汽信息科技(天津)有限公司 | Optimization method for IO intensive task memory utilization rate based on neural network |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103164268A (en) * | 2013-04-02 | 2013-06-19 | 北京奇虎科技有限公司 | System optimization method and system optimization device |
| CN103793265A (en) * | 2012-10-30 | 2014-05-14 | 腾讯科技（深圳）有限公司 | Processing method and device for process optimization |
| CN104375898A (en) * | 2014-11-20 | 2015-02-25 | 无锡悟莘科技有限公司 | Mobile terminal CPU occupancy rate optimization method |
-
2016
- 2016-09-13 CN CN201610821159.9A patent/CN106445661A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103793265A (en) * | 2012-10-30 | 2014-05-14 | 腾讯科技（深圳）有限公司 | Processing method and device for process optimization |
| CN103164268A (en) * | 2013-04-02 | 2013-06-19 | 北京奇虎科技有限公司 | System optimization method and system optimization device |
| CN104375898A (en) * | 2014-11-20 | 2015-02-25 | 无锡悟莘科技有限公司 | Mobile terminal CPU occupancy rate optimization method |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110750346A (en) * | 2019-10-17 | 2020-02-04 | Oppo（重庆）智能科技有限公司 | Task operation optimization method, device, terminal and storage medium |
| CN110750346B (en) * | 2019-10-17 | 2022-11-15 | Oppo（重庆）智能科技有限公司 | Task operation optimization method, device, terminal and storage medium |
| CN116991592A (en) * | 2023-09-26 | 2023-11-03 | 中汽信息科技(天津)有限公司 | Optimization method for IO intensive task memory utilization rate based on neural network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110727512B (en) | Cluster resource scheduling method, device, equipment and storage medium | |
| CN100377113C (en) | Methods and apparatus to process cache allocation requests based on priority | |
| US6587938B1 (en) | Method, system and program products for managing central processing unit resources of a computing environment | |
| US7051188B1 (en) | Dynamically redistributing shareable resources of a computing environment to manage the workload of that environment | |
| CN103729248B (en) | A kind of method and apparatus of determination based on cache perception task to be migrated | |
| US6519660B1 (en) | Method, system and program products for determining I/O configuration entropy | |
| US6651125B2 (en) | Processing channel subsystem pending I/O work queues based on priorities | |
| US8996811B2 (en) | Scheduler, multi-core processor system, and scheduling method | |
| US7007276B1 (en) | Method, system and program products for managing groups of partitions of a computing environment | |
| CN110333937A (en) | Task distribution method, device, computer equipment and storage medium | |
| CN103646006B (en) | The dispatching method of a kind of processor, device and system | |
| US7698529B2 (en) | Method for trading resources between partitions of a data processing system | |
| US20060020944A1 (en) | Method, system and program products for managing logical processors of a computing environment | |
| CN101366012A (en) | Methods and system for interrupt distribution in a multiprocessor system | |
| CN109684074A (en) | Physical machine resource allocation methods and terminal device | |
| CN103959275B (en) | The method and system of the memory access in optimization multiple processor computation system | |
| CN103365726A (en) | Resource management method and system facing GPU (Graphic Processing Unit) cluster | |
| US20220195434A1 (en) | Oversubscription scheduling | |
| CN115033340A (en) | Host selection method and related device | |
| CN113342477A (en) | Container group deployment method, device, equipment and storage medium | |
| CN111352731A (en) | Method, system, apparatus and medium for distributing tasks in edge computing network | |
| CN106164888A (en) | The sequencing schemes of network and storage I/O request for minimizing interference between live load free time and live load | |
| CN115237586A (en) | GPU resource configuration method for deep learning inference performance interference perception | |
| Zhang et al. | Zeus: Improving resource efficiency via workload colocation for massive kubernetes clusters | |
| CN106059940A (en) | Flow control method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170222 |
|
| WD01 | Invention patent application deemed withdrawn after publication |