CN108052425A - A kind of method and device of the lifting system performance based on kernel tuning - Google Patents
A kind of method and device of the lifting system performance based on kernel tuning Download PDFInfo
- Publication number
- CN108052425A CN108052425A CN201711286959.6A CN201711286959A CN108052425A CN 108052425 A CN108052425 A CN 108052425A CN 201711286959 A CN201711286959 A CN 201711286959A CN 108052425 A CN108052425 A CN 108052425A
- Authority
- CN
- China
- Prior art keywords
- tuning
- computer
- kernel
- mysql
- scheme
- 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
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/22—Detection or location of defective computer hardware by testing during standby operation or during idle time, e.g. start-up testing
- G06F11/2205—Detection or location of defective computer hardware by testing during standby operation or during idle time, e.g. start-up testing using arrangements specific to the hardware being tested
- G06F11/2236—Detection or location of defective computer hardware by testing during standby operation or during idle time, e.g. start-up testing using arrangements specific to the hardware being tested to test CPU or processors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Hardware Design (AREA)
- Quality & Reliability (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Debugging And Monitoring (AREA)
Abstract
The embodiment of the invention discloses a kind of method and device of the lifting system performance based on kernel tuning, method includes the monitoring item data of acquisition system；According to rule base, mapreduce operations are carried out to monitoring item data, filter out computer of the system performance there are bottleneck；The kernel parameter of bottleneck computer is obtained, corresponding solution is chosen in the scheme base stored from mysql.Device includes data collecting system, tuning platform and mysql scheme bases, data collecting system is used for the monitoring item data of collecting computer, and give the monitored item data sending to tuning platform, tuning platform carries out mapreduce operations to monitoring item data, computer of the system performance there are bottleneck is filtered out, kernel parameter and corresponding tuning scheme are stored in mysql scheme bases.The present invention ensures that system for computer performance is continuously in optimum state by the way that kernel parameter tuning, the performance of lifting system, whole process need not think to participate in, and provide work efficiency, and evolutionary process constantly carries out.
Description
Technical field
The present invention relates to field of computer technology, the side of specifically a kind of lifting system performance based on kernel tuning
Method and device.
Background technology
Company's major part machine is all using unified acquiescence kernel setup parameter, as the kernel of the bottom, some cooperations
System performance can be substantially improved in the parameter modification of runtime；Can by kernel parameter tuning, the processing capacity of lifting system,
And where analyzing system performance bottleneck.
Kernel parameter tuning is intended to, by observing system state, the unreasonable place of systematic parameter be found, by adjusting in system
The partial parameters of core, to improve the performance of application.The kernel parameter of system has vital in the operational process of program
Effect, and kernel also provides various means for user to change relevant kernel parameter in itself, to support different applied fields
Scape.
Our kernels carry out independently each product line in application aspect, the default configuration typically used now
Artificial tuning, work efficiency are low.
The content of the invention
A kind of method and device of the lifting system performance based on kernel tuning is provided in the embodiment of the present invention, to solve
The problem of artificial tuning work efficiency is low in the prior art
In order to solve the above-mentioned technical problem, the embodiment of the invention discloses following technical solutions：
The first invention of the present invention provides a kind of method of the lifting system performance based on kernel tuning, this method include with
Lower step：
The monitoring item data of acquisition system；
According to rule base, mapreduce operations are carried out to monitoring item data, filter out calculating of the system performance there are bottleneck
Machine；
The kernel parameter of bottleneck computer is obtained, corresponding solution is chosen in the scheme base stored from mysql.
With reference to first aspect, in first aspect in the first possible realization method, the monitored item data include cpu_
Idle and time_wait.
With reference to first aspect, in second of possible realization method of first aspect, per rule pair in the rule base
The tuning operator of a mapreduce changeable parameters is answered, each tuning operator forms a directed acyclic topological diagram, by a plurality of rule
Then corresponding topological diagram is combined into new topological diagram.
With reference to first aspect, in second of possible realization method of first aspect, when carrying out mapreduce operations,
Cutting is carried out to the new topological diagram, forms mapreduce operations several times.
With reference to first aspect, in first aspect in the third possible realization method, the solution is from mysql
Related command is called, tuning is carried out to system performance.
Second invention of the invention provides a kind of device of the lifting system performance based on kernel tuning, which includes number
According to acquisition system, tuning platform and mysql scheme bases, the data collecting system is used for the monitoring item data of collecting computer,
And give the monitored item data sending to tuning platform, the tuning platform carries out mapreduce operations, sieve to monitoring item data
Computer of the system performance there are bottleneck is selected, kernel parameter and corresponding tuning scheme are stored in the mysql scheme bases.
With reference to second aspect, in second aspect in the first possible realization method, the data collecting system is additionally operable to
The kernel parameter data of bottleneck computer system are gathered, and tuning platform is given the kernel parameter data sending of acquisition, tuning is put down
Platform is called corresponding tuning scheme from mysql scheme bases, kernel parameter is adjusted by the kernel parameter data of acquisition
It is excellent.
A kind of device of lifting system performance based on kernel tuning of second aspect of the present invention can realize first aspect
And the method in each realization method of first aspect, and obtain identical effect.
The effect provided in the content of the invention is only the effect of embodiment rather than invents all whole effects, above-mentioned
A technical solution in technical solution has the following advantages that or advantageous effect：
1st, by obtaining the monitoring data of computer system, mapreduce operations is carried out to monitoring data, filter out system
Performance and obtains the kernel parameter of bottleneck computer, is chosen in the scheme base stored from mysql corresponding there are the computer of bottleneck
Solution, so as to by the way that kernel parameter tuning, the performance of lifting system, whole process need not think to participate in, and provide work
Make efficiency, and the above process constantly carries out, and ensures that system for computer performance is continuously in optimum state.
2nd, every rule in rule base corresponds to a tuning operator, and each tuning operator one directed acyclic of composition is opened up
Figure is flutterred, when adding in new rule, it is only necessary to the corresponding topological diagram of new rule is combined into new topological diagram, without to newly adding
The rule entered writes complete logical code, easy to operate, and to topological diagram carry out mapreduce operations when, tuning platform pair
Topological diagram carries out cutting, is carried out at the same time mapreduce operations to each subtopology figure point, improves the efficiency of mapreduce operations, soon
Speed filters out computer of the system performance there are bottleneck.
3rd, when carrying out system performance tuning, solution directly is called from mysql scheme bases, is improving tuning efficiency
While, mistake caused by artificial tuning is avoided, enhances the reliability of tuning result.
Description of the drawings
It in order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
There is attached drawing needed in technology description to be briefly described, it should be apparent that, for those of ordinary skill in the art
Speech, without creative efforts, can also be obtained according to these attached drawings other attached drawings.
Fig. 1 is a kind of flow diagram of the method for the lifting system performance based on kernel tuning；
Fig. 2 is a kind of topological schematic diagram of tuning operator composition；
Fig. 3 is a kind of structural representation for lifting system performance system based on kernel tuning that the embodiment of the present invention is applied
Figure.
Specific embodiment
In order to which those skilled in the art is made to more fully understand the technical solution in the present invention, below in conjunction with of the invention real
The attached drawing in example is applied, the technical solution in the embodiment of the present invention is clearly and completely described, it is clear that described implementation
Example is only part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, this field is common
Technical staff's all other embodiments obtained without making creative work, should all belong to protection of the present invention
Scope.
As shown in Figure 1, a kind of method of lifting system performance based on kernel tuning of the present invention comprises the following steps：
S1 obtains the monitoring item data of system；
S2 according to rule base, carries out mapreduce operations, filtering out system performance, there are bottlenecks to monitoring item data
Computer；
S3 obtains the kernel parameter of bottleneck computer, and corresponding solution is chosen in the scheme base stored from mysql.
In step S1, the monitoring data of acquisition includes cpu_idle and time_wait etc..
Rule base in step S2 refers to the rule base that the corresponding screening rule of each monitoring data is formed, such as monitors number
It is cpu_idle_low according to the corresponding rules of cpu_idle, when carrying out mapreduce operations, that is, filters out cpu_idle and be less than
The computer of given threshold, threshold value can be according to actual conditions sets itself herein.
In order to ensure the scalability of rule base and versatility, a mapreduce will be corresponded in rule base per rule
The tuning operator of changeable parameters, each tuning operator form a directed acyclic topological diagram.So, when a plurality of rule of progress
Matching when, different rules can be combined into one and new big open up benefit structure again.Such as when adding in new rule, it is only necessary to
The corresponding topological diagram of new rule is combined into new topological diagram.When carrying out mapreduce operations, complement can be opened up greatly to this
Cutting is carried out, several wheel mapreduce operations is organized into and is calculated.Alloing a plurality of rule, utmostly Shangdi is parallel
The matching that a rule carries out a rule again is calculated rather than matched, improves operating efficiency.
As shown in Fig. 2, the topological diagram of tuning operator composition.Wherein tuning operator is divided into operation and aggregation
Two major classes.Operation classes are that wall scroll record is operated, while also only export wall scroll record；Aggregation classes are then
A plurality of record is polymerize, output one to a plurality of record.Common tuning operator has following several：
Initialize (initializer)：Operation types, the entrance operator of strictly all rules, for by text lattice
The data conversion of formula is into corresponding data structure to facilitate subsequent processing.
Filter (filter operator)：Operation types, for being filtered to wall scroll record.
Refactor (reconstruct operator)：Operation types, between key/value fields between being recorded to wall scroll
Conversion delete or order adjust.
Simple Join (combination operators)：Aggregation types, for record join identical key is new into one
Record.
Time Filter (time filtration operator)：Aggregation types, for by the period of eligible rule
It screens, it is incongruent to filter out.
In step S3, the kernel parameter with performance bottleneck computer that is filtered out in obtaining step S2, these corresponding ginsengs
Number, tuning platform choose corresponding solution from the solution storehouse that mysql is stored again.Such as when cpu_idle is too low
It is ./bioscfg.py-s cstate=off that the kernel parameter of corresponding mysql storages, which repaiies solution, for example works as socket
It is calling/proc/ that time_wait states, which excessively occupy a large amount of port and correspond to the kernel parameter that mysql is stored to repair solution,
Sys/net/ipv4/tcp_tw_reuse=1；/ proc/sys/net/ipv4/tcp_timestamps=1；/proc/sys/
Net/ipv4/tcp_tw_recycle=1, so as to the continuous tuning of the performance of these machines.
As shown in figure 3, a kind of device of lifting system performance based on kernel tuning of the present invention, which includes data
Acquisition system 1, tuning platform 2 and mysql scheme bases 3, data collecting system are used for the monitoring item data of collecting computer, and will
Monitored item data sending carries out mapreduce operations to monitoring item data, filters out systematicness to tuning platform 2, tuning platform 2
Kernel parameter and corresponding tuning scheme can be stored in mysql scheme bases 3 there are the computer of bottleneck.
Data collecting system 1 is additionally operable to the kernel parameter data of acquisition bottleneck computer system, and the kernel of acquisition is joined
Number data sending puts down 2 to tuning, and tuning platform 2 calls phase by the kernel parameter data of acquisition from 3 storehouse of mysql schemes
The tuning scheme answered carries out tuning to kernel parameter.
The above is only the specific embodiment of the present invention, is made skilled artisans appreciate that or realizing this hair
It is bright.A variety of modifications of these embodiments will be apparent to one skilled in the art, it is as defined herein
General Principle can be realized in other embodiments without departing from the spirit or scope of the present invention.Therefore, it is of the invention
The embodiments shown herein is not intended to be limited to, and is to fit to and the principles and novel features disclosed herein phase one
The most wide scope caused.
Claims (7)
1. a kind of method of the lifting system performance based on kernel tuning, it is characterized in that：Comprise the following steps：
The monitoring item data of acquisition system；
According to rule base, mapreduce operations are carried out to monitoring item data, filter out computer of the system performance there are bottleneck；
The kernel parameter of bottleneck computer is obtained, corresponding solution is chosen in the scheme base stored from mysql.
2. according to the method described in claim 1, it is characterized in that：The monitored item data include cpu_idle and time_wait.
3. according to the method described in claim 1, it is characterized in that：A mapreduce is corresponded in the rule base per rule
The tuning operator of changeable parameters, each tuning operator forms a directed acyclic topological diagram, by the corresponding topological diagram of a plurality of rule
It is combined into new topological diagram.
4. according to the method described in claim 3, it is characterized in that：When carrying out mapreduce operations, to the new topological diagram into
Row cutting forms mapreduce operations several times.
5. according to the method described in claim 1, it is characterized in that：The solution is right to call related command from mysql
System performance carries out tuning.
6. a kind of device of the lifting system performance based on kernel tuning, it is characterized in that：Including data collecting system, tuning platform
With mysql scheme bases, the data collecting system is used for the monitoring item data of collecting computer, and the monitoring item data is sent out
Tuning platform is given, the tuning platform carries out mapreduce operations to monitoring item data, and filtering out system performance, there are bottlenecks
Computer, store kernel parameter and corresponding tuning scheme in the mysql scheme bases.
7. device according to claim 5, it is characterized in that：The data collecting system is additionally operable to acquisition bottleneck department of computer science
The kernel parameter data of system, and give tuning platform, the kernel that tuning platform passes through acquisition the kernel parameter data sending of acquisition
Supplemental characteristic calls corresponding tuning scheme from mysql scheme bases, and tuning is carried out to kernel parameter.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711286959.6A CN108052425A (en) | 2017-12-07 | 2017-12-07 | A kind of method and device of the lifting system performance based on kernel tuning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711286959.6A CN108052425A (en) | 2017-12-07 | 2017-12-07 | A kind of method and device of the lifting system performance based on kernel tuning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108052425A true CN108052425A (en) | 2018-05-18 |
Family
ID=62123313
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711286959.6A Pending CN108052425A (en) | 2017-12-07 | 2017-12-07 | A kind of method and device of the lifting system performance based on kernel tuning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108052425A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113626073A (en) * | 2021-08-06 | 2021-11-09 | 航天中认软件测评科技（北京）有限责任公司 | Software adaptation optimization method based on knowledge base |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102929613A (en) * | 2012-10-16 | 2013-02-13 | 无锡江南计算技术研究所 | Adjusting and optimizing device and method for operating system |
| CN103116538A (en) * | 2013-01-25 | 2013-05-22 | 浪潮电子信息产业股份有限公司 | Design for computer performance self-adjusting system |
| CN104933175A (en) * | 2015-06-30 | 2015-09-23 | 深圳市金证科技股份有限公司 | Performance data dependency analyzing method and performance monitoring system |
| CN106202431A (en) * | 2016-07-13 | 2016-12-07 | 华中科技大学 | A kind of Hadoop parameter automated tuning method and system based on machine learning |
| CN106506235A (en) * | 2016-12-06 | 2017-03-15 | 郑州云海信息技术有限公司 | A kind of network tuning method and apparatus |
-
2017
- 2017-12-07 CN CN201711286959.6A patent/CN108052425A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102929613A (en) * | 2012-10-16 | 2013-02-13 | 无锡江南计算技术研究所 | Adjusting and optimizing device and method for operating system |
| CN103116538A (en) * | 2013-01-25 | 2013-05-22 | 浪潮电子信息产业股份有限公司 | Design for computer performance self-adjusting system |
| CN104933175A (en) * | 2015-06-30 | 2015-09-23 | 深圳市金证科技股份有限公司 | Performance data dependency analyzing method and performance monitoring system |
| CN106202431A (en) * | 2016-07-13 | 2016-12-07 | 华中科技大学 | A kind of Hadoop parameter automated tuning method and system based on machine learning |
| CN106506235A (en) * | 2016-12-06 | 2017-03-15 | 郑州云海信息技术有限公司 | A kind of network tuning method and apparatus |
Non-Patent Citations (1)
| Title |
|---|
| 俞俊生: "《从云到终端提升移动流媒体性能技术研究》", 30 September 2015 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113626073A (en) * | 2021-08-06 | 2021-11-09 | 航天中认软件测评科技（北京）有限责任公司 | Software adaptation optimization method based on knowledge base |
| CN113626073B (en) * | 2021-08-06 | 2022-06-10 | 航天中认软件测评科技（北京）有限责任公司 | Software adaptation optimization method based on knowledge base |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105677918B (en) | A kind of distributed reptile framework and its implementation based on Kafka and Quartz | |
| CN105389402B (en) | A kind of ETL method and apparatus towards big data | |
| CN103207903B (en) | A kind of visualized graphs program topological sorting method | |
| CN106302632A (en) | The method for down loading of a kind of foundation image and management node | |
| CN109271321A (en) | A kind of contribution code number statistical method and device | |
| CN106874482A (en) | A kind of device and method of the patterned data prediction based on big data technology | |
| CN105095425B (en) | A kind of inter-library method and device of carrying down of database | |
| CN105630797B (en) | Data processing method and system | |
| CN108052425A (en) | A kind of method and device of the lifting system performance based on kernel tuning | |
| CN107729063A (en) | Call method, device and the terminal of SDK | |
| CN109582837A (en) | A kind of visualized data processing method based on cloud and system | |
| CN102624544A (en) | Method and device for creating monitoring tasks | |
| CN104408263A (en) | Graphic element grouping and ranking method for automatically processing closed loop | |
| CN102622447B (en) | Hadoop-based frequent closed itemset mining method | |
| CN109002885A (en) | A kind of convolutional neural networks pond unit and pond calculation method | |
| CN103838574A (en) | General method for grouped summarizing of data tables | |
| CN106027032A (en) | RM logic circuit delay optimization method in unit delay model | |
| CN103647694B (en) | E-mail cleaning method and device | |
| CN108171336B (en) | The automatic sensing method and device of machine data model evolution | |
| CN106557519A (en) | File delet method and device | |
| CN104994172A (en) | Monitoring and management system and method for cloud storage system | |
| CN107729507A (en) | A kind of data cleaning method calculated based on terminal storage, nearby | |
| CN110377474A (en) | A kind of loop test method and system of Mellanox network interface card | |
| CN207680201U (en) | A kind of filter material cleaning device of filtration system | |
| CN218729047U (en) | A device for financial affairs fund is settled accounts in concentration |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180518 |
|
| RJ01 | Rejection of invention patent application after publication |