CN101158916A - Data-base performance monitoring method - Google Patents
Data-base performance monitoring method Download PDFInfo
- Publication number
- CN101158916A CN101158916A CNA2007101568869A CN200710156886A CN101158916A CN 101158916 A CN101158916 A CN 101158916A CN A2007101568869 A CNA2007101568869 A CN A2007101568869A CN 200710156886 A CN200710156886 A CN 200710156886A CN 101158916 A CN101158916 A CN 101158916A
- Authority
- CN
- China
- Prior art keywords
- performance
- database
- data
- monitoring method
- period
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Abstract
The invention discloses a database performance monitoring method, which belongs to the field of database performance monitoring technology. The monitoring method obtains the relevant performance parameters from the database, (CPU waiting time plus I/O waiting time)divided by total waiting time of the database, then multiplied by 100 percent shall be considered as a comprehensive index of the database performance, and the indicator is compared with preset normal index value/performance poor index value, thereby generating an alarm in accordance with the comparison. The monitoring method can promptly and effectively monitor the database performance.
Description
Technical field
The invention belongs to the data-base performance monitoring technical field, be specifically related to a kind of data-base performance monitoring method.
Background technology
For a long time, can only find thoroughly the delay situation of machine of database, then can't in time find for the database performance deterioration problem to the monitoring of core database.In the industry cycle, do not have a kind of data-base performance monitoring scheme of maturation yet, to the monitoring of database can only be qualitative can not be quantitative.At present, the difficult point of database monitoring is that the performance index of database inside are hundreds and thousands of, and single index can collect, but the deterioration of single index can not say something, and needs the combination property that a comprehensive index can reflect entire database.
Summary of the invention
For solving problems of the prior art, the present invention aims to provide a kind of technical scheme that can in time find the effective monitoring method that database performance worsens.
Described a kind of data-base performance monitoring method may further comprise the steps: obtain performance data CPU stand-by period, I/O stand-by period and total stand-by period of database from monitored database; Above-mentioned performance data is calculated as follows the current integrated performance index value T of monitored database
c, T
c=(CPU stand-by period+I/O stand-by period)/total stand-by period * 100% of database; With above-mentioned current integrated performance index value T
cWith predefined normal value T
aRelatively big or small, if T
cLess than T
a, then point out monitored database performance existing problems.
Described data-base performance monitoring method also can be with above-mentioned current integrated performance index value T
cBadly be worth T with predefined performance
bRelatively big or small, if T
cLess than T
b, then point out monitored database performance abominable.
Described data-base performance monitoring method, wherein, described normal value T
aSpan is 40%～60%, is preferably 50%.
Described data-base performance monitoring method, wherein, described performance badly is worth T
bSpan is 10%～20%, is preferably 15%.
The importance that all kinds of stand-by period parameters in the database are database performance parameters mainly comprises:
The CPU stand-by period: the time of waiting for CPU computing;
The I/O stand-by period: the time of waiting system I/O read-write;
The internal memory stand-by period: the time of waiting for memory read-write;
The lock stand-by period: wait for the time that lock discharges;
Other various stand-by period.
According to long-term database O﹠M experience, CPU stand-by period and I/O stand-by period are the main embodiments of the whole traffic pressure of database, generally all account for the significant proportion of overall stand-by period of database, and under normal circumstances can not become the main bottleneck of traffic pressure such as internal memory wait, lock wait etc.But when database performance took place to worsen, abnormal conditions often can show all that and corresponding, the CPU of this moment waits for that the ratio that I/O waits for will descend rapidly on the indexs such as internal memory is waited for, lock wait.
Based on above-mentioned analysis, will (CPU stand-by period+I/O stand-by period)/total stand-by period * 100% of database as an overall target of database performance, and with this index and the abominable desired value comparison of predefined normal index value/performance, generate alarm according to comparative result, thereby effective monitor database performance in time,, alarm rate of accuracy reached 95% efficiently reaches 80%.
Certainly, at different applicable cases, the abominable desired value scope of above-mentioned normal index value/performance can be adjusted to some extent according to actual conditions.
Description of drawings
Fig. 1 is the schematic flow sheet of a preferred embodiment of data-base performance monitoring method of the present invention.
Fig. 2 is the monitoring alarm statistical graph in the specific embodiment of the invention.
Embodiment
Now, describe a preferred embodiment of the present invention and effect thereof in detail in conjunction with Figure of description.
Referring to Fig. 1, move the oracle database of present stage at Zhejiang, adopt described data-base performance monitoring method to realize the monitoring of this database combination property: can generate a oracle database performance data in per 15 minutes by existing P ricise monitoring tools, wherein comprise much more very oracle database performance index data, from above-mentioned performance index extracting data: CPU stand-by period, I/O stand-by period and total stand-by period of database; Then according to our combination property operational formula T
c=(CPU stand-by period+I/O the stand-by period)/total stand-by period * 100% of database obtains the current integrated performance index value T of this database
cWith T
cWith predefined normal threshold values T
aRelatively big or small, T
aSpan be 40%～60%, be preferably 50%, if T
cLess than T
a, then point out monitored database performance existing problems, generate corresponding monitoring alarm, the notice associated maintenance personnel handle; Simultaneously, also can be with T
cBadly be worth T with predefined performance
bRelatively big or small, T
bSpan is 10%～20%, is preferably 15%, if T
cLess than T
b, then point out monitored database performance abominable, generate corresponding monitoring alarm, the notice associated maintenance personnel handle.On the other hand, also can be for future use with above-mentioned performance index data archiving.
Above-mentioned data-base performance monitoring method is applied to Zhejiang move database combination property alarm, Alarm Classification statistical graph from April, 2007 to July, referring to Fig. 2, wherein, effective 1/ effective 2 represents that all alarm is effectively handled, effective 1 expression can judge clearly that by daily record of work this alarm effectively avoided follow-up fortuitous event, effective 2 expressions can't clearly judge whether effectively to have avoided fortuitous event, database performance had returned to normal level when burst value was represented this alarming processing, but the effect of reminding has been played in this part alarm, set of systems can be paid close attention to database performance targetedly, database performance had returned to normal level when burst value was represented this alarming processing, but the effect of reminding has been played in this part alarm, set of systems can be paid close attention to database performance targetedly, and wrong report returns from the Xueyuan Road mainly due to the business storehouse that alarm configuration does not have time update to cause when cutting to the hinge building, existing adjusted, can not occur wrong report again, repeat alarm on a small quantity.As can be seen from Figure 2, during this period monitoring alarm has occurred 42 times altogether, and alarm rate of accuracy reached to 95% efficiently reaches 80%; If remove the wrong report that allocation problem produces, the alarm rate of accuracy reached to 100% of this method for supervising efficiently reaches 85%; As seen, this data-base performance monitoring method has been brought into play good effect in actual production.
Claims (6)
1. a data-base performance monitoring method is characterized in that, may further comprise the steps:
Obtain performance data CPU stand-by period, I/O stand-by period and total stand-by period of database from monitored database;
Above-mentioned performance data is calculated as follows the current integrated performance index value T of monitored database
c:
T
c=(CPU stand-by period+I/O stand-by period)/total stand-by period * 100% of database;
With above-mentioned current integrated performance index value T
cWith predefined normal value T
aRelatively big or small, if T
cLess than T
a, then point out monitored database performance existing problems.
2. data-base performance monitoring method as claimed in claim 1 is characterized in that:
Described current integrated performance index value T
cBadly be worth T with predefined performance
bRelatively big or small, if T
cLess than T
b, then point out monitored database performance abominable.
3. data-base performance monitoring method as claimed in claim 1 is characterized in that: described normal value T
aSpan is 40%～60%.
4. data-base performance monitoring method as claimed in claim 3 is characterized in that: described normal value T
aBe 50%.
5. data-base performance monitoring method as claimed in claim 2 is characterized in that: described performance badly is worth T
bSpan is 10%～20%.
6. data-base performance monitoring method as claimed in claim 5 is characterized in that: described performance badly is worth T
bBe 15%.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2007101568869A CN101158916A (en) | 2007-11-19 | 2007-11-19 | Data-base performance monitoring method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2007101568869A CN101158916A (en) | 2007-11-19 | 2007-11-19 | Data-base performance monitoring method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101158916A true CN101158916A (en) | 2008-04-09 |
Family
ID=39307024
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2007101568869A Pending CN101158916A (en) | 2007-11-19 | 2007-11-19 | Data-base performance monitoring method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101158916A (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101989283A (en) * | 2009-08-04 | 2011-03-23 | 中兴通讯股份有限公司 | Monitoring method and device of performance of database |
| CN102081623B (en) * | 2009-11-30 | 2012-10-03 | 中国移动通信集团浙江有限公司 | Method and system for detecting database abnormality |
| CN102790695A (en) * | 2012-07-23 | 2012-11-21 | 华为技术有限公司 | Diagnostic system and method for performance bottleneck of server I/O (input/output) subsystem |
| CN102982037A (en) * | 2011-09-05 | 2013-03-20 | 中国移动通信集团浙江有限公司 | Database node health condition detection method and detection device |
| CN101770419B (en) * | 2008-12-31 | 2013-03-20 | 中国银联股份有限公司 | System robustness analyzer and analysis method |
| CN103412911A (en) * | 2013-08-02 | 2013-11-27 | 中国工商银行股份有限公司 | Method and device for monitoring performance of database system |
| CN103885995A (en) * | 2012-12-21 | 2014-06-25 | 中国移动通信集团河北有限公司 | List-based database monitoring method and list-based database monitoring device |
| CN104980318A (en) * | 2015-06-19 | 2015-10-14 | 上海卓悠网络科技有限公司 | Visualized health monitoring method and apparatus for network and server of IDC |
| CN106293941A (en) * | 2016-08-09 | 2017-01-04 | 上海新炬网络信息技术有限公司 | A kind of distribution method of Database Systems resource |
| CN106407085A (en) * | 2016-11-24 | 2017-02-15 | 中国银行股份有限公司 | Performance monitoring method and apparatus |
-
2007
- 2007-11-19 CN CNA2007101568869A patent/CN101158916A/en active Pending
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101770419B (en) * | 2008-12-31 | 2013-03-20 | 中国银联股份有限公司 | System robustness analyzer and analysis method |
| CN101989283B (en) * | 2009-08-04 | 2014-06-11 | 中兴通讯股份有限公司 | Monitoring method and device of performance of database |
| CN101989283A (en) * | 2009-08-04 | 2011-03-23 | 中兴通讯股份有限公司 | Monitoring method and device of performance of database |
| CN102081623B (en) * | 2009-11-30 | 2012-10-03 | 中国移动通信集团浙江有限公司 | Method and system for detecting database abnormality |
| CN102982037B (en) * | 2011-09-05 | 2016-05-25 | 中国移动通信集团浙江有限公司 | Method and the device of Test database node health status |
| CN102982037A (en) * | 2011-09-05 | 2013-03-20 | 中国移动通信集团浙江有限公司 | Database node health condition detection method and detection device |
| CN102790695B (en) * | 2012-07-23 | 2015-03-25 | 华为技术有限公司 | Diagnostic system and method for performance bottleneck of server I/O (input/output) subsystem |
| CN102790695A (en) * | 2012-07-23 | 2012-11-21 | 华为技术有限公司 | Diagnostic system and method for performance bottleneck of server I/O (input/output) subsystem |
| CN103885995A (en) * | 2012-12-21 | 2014-06-25 | 中国移动通信集团河北有限公司 | List-based database monitoring method and list-based database monitoring device |
| CN103885995B (en) * | 2012-12-21 | 2017-05-03 | 中国移动通信集团河北有限公司 | List-based database monitoring method and list-based database monitoring device |
| CN103412911A (en) * | 2013-08-02 | 2013-11-27 | 中国工商银行股份有限公司 | Method and device for monitoring performance of database system |
| CN103412911B (en) * | 2013-08-02 | 2016-08-10 | 中国工商银行股份有限公司 | The method for monitoring performance of Database Systems and device |
| CN104980318A (en) * | 2015-06-19 | 2015-10-14 | 上海卓悠网络科技有限公司 | Visualized health monitoring method and apparatus for network and server of IDC |
| CN106293941A (en) * | 2016-08-09 | 2017-01-04 | 上海新炬网络信息技术有限公司 | A kind of distribution method of Database Systems resource |
| CN106293941B (en) * | 2016-08-09 | 2019-12-31 | 上海新炬网络信息技术股份有限公司 | Method for distributing database system resources |
| CN106407085A (en) * | 2016-11-24 | 2017-02-15 | 中国银行股份有限公司 | Performance monitoring method and apparatus |
| CN106407085B (en) * | 2016-11-24 | 2019-03-15 | 中国银行股份有限公司 | A kind of method for monitoring performance and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101158916A (en) | Data-base performance monitoring method | |
| Qiu et al. | Wind turbine SCADA alarm analysis for improving reliability | |
| CN101286897B (en) | Network flow rate abnormality detecting method based on super stochastic theory | |
| CN109981328B (en) | Fault early warning method and device | |
| CN107888441B (en) | Network traffic baseline self-learning self-adaption method | |
| US10489711B1 (en) | Method and apparatus for predictive behavioral analytics for IT operations | |
| CN103971024A (en) | Method for evaluating reliability of relaying protection systems under small sample failure data | |
| US7289935B1 (en) | Statistical quality control of alarm occurrences | |
| CN107733737A (en) | A kind of abnormal method of monitoring traffic in network | |
| CN113762604B (en) | Industrial Internet big data service system | |
| Zhu et al. | Dynamic alarm prediction for critical alarms using a probabilistic model | |
| CN104991939A (en) | Transaction data monitoring method and system | |
| JPH11296208A (en) | Production management information output device | |
| CN103986240A (en) | Power distribution real-time data validity analytical processing system and method | |
| US11887465B2 (en) | Methods, systems, and computer programs for alarm handling | |
| CN110232792A (en) | Intelligent alarm technology and auxiliary operation method based on SCADA system | |
| CN106199251B (en) | A kind of distribution network failure early warning system and method based on adaptive modeling analysis | |
| Sebuk-gu et al. | Assessing software reliability based on NHPP using SPC | |
| WO2024011745A1 (en) | Fault warning method based on oil field grid power device | |
| CN107453882B (en) | Alarm information aggregation system and method in cluster | |
| CN113110350A (en) | Cement production equipment halt analysis management system | |
| CN113361904B (en) | Monitoring and alarming method, device, equipment and readable storage medium | |
| CN112580837A (en) | Equipment maintenance method, device, equipment and storage medium | |
| CN113064942A (en) | Environment-friendly monitoring system and method based on electric power big data analysis | |
| Nochur et al. | Alarm performance metrics |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Open date: 20080409 |