CN103064776A - Performance monitoring method and performance monitoring device - Google Patents
Performance monitoring method and performance monitoring device Download PDFInfo
- Publication number
- CN103064776A CN103064776A CN2011103235912A CN201110323591A CN103064776A CN 103064776 A CN103064776 A CN 103064776A CN 2011103235912 A CN2011103235912 A CN 2011103235912A CN 201110323591 A CN201110323591 A CN 201110323591A CN 103064776 A CN103064776 A CN 103064776A
- Authority
- CN
- China
- Prior art keywords
- program statement
- execution time
- time
- program
- alarm
- 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
Images
Landscapes
- Debugging And Monitoring (AREA)
Abstract
The invention discloses a performance monitoring method and a performance monitoring device and aims at solving the problems that a performance monitoring method in the prior art enables efficiency of removing failure to be low and reduces business quality. The method comprises steps of extracting each program statement contained in programs of business, executing the program statement aiming at each extracted program statement, acquiring first executing time corresponding to the executed program statement, judging whether the acquired executing time is more than the standard executing time according to standard executing time corresponding to the program statement, and sounding alarm of failure instruction of the program statement. By adoption of the method, the operation condition of each program statement contained in the business can be monitored, alarm of the failure instruction of the program sentence can be sounded when a certain program statement fails so that failure statement locating time can be shortened, and therefore failure removing efficiency is improved, and business quality is improved.
Description
Technical field
The present invention relates to communication technical field, relate in particular to a kind of method for monitoring performance and device.
Background technology
Along with the development of business, service logic is increasingly sophisticated, and each professional program influencing each other aspect operational efficiency also constantly increases, therefore in order to improve quality of service, and need to be to each professional monitoring performance.
In the prior art, to the method for the monitoring performance of business be: this professional program ruuning situation is monitored, sent alarm when this professional program breaks down when monitoring.
Yet because present service logic is increasingly sophisticated, the program of single business has often just comprised a large amount of program statements, and the fault that performance issue all can cause whole program appears in any program statement wherein.Therefore, in the prior art, the O﹠M personnel are after receiving alarm, can only know that fault has appearred in this professional program, but can not know that really performance issue has appearred in which bar or which bar program statement in this program, causes this program fault to occur, thereby has expended a large amount of time in the location of fault statement, reduce the efficient of fixing a breakdown, also reduced quality of service.
Summary of the invention
The embodiment of the invention provides a kind of method for monitoring performance and device, makes the efficient of fixing a breakdown low in order to the method that solves the monitoring of prior art performance, has reduced the problem of quality of service.
A kind of method for monitoring performance that the embodiment of the invention provides comprises:
Extract every the program statement that comprises in the professional program; And
For every the program statement that extracts, carry out this program statement, obtain to carry out the first execution time corresponding to this program statement; And
According to benchmark execution time corresponding to this program statement that arranges, judge that whether described the first execution time that obtains is greater than the described benchmark execution time that arranges;
When judged result when being, send the alarm that this program statement of indication breaks down.
A kind of performance monitoring apparatus that the embodiment of the invention provides comprises:
Extraction module is used for extracting every program statement that professional program comprises;
Execution module is used for carrying out this program statement for every the program statement that extracts, and obtains to carry out the first execution time corresponding to this program statement;
Judge module was used for according to benchmark execution time corresponding to this program statement that arranges, and judged that whether described the first execution time that obtains is greater than the described benchmark execution time that arranges;
Alarm module, be used for when judged result when being, send the alarm of indicating this program statement to break down.
The embodiment of the invention provides a kind of method for monitoring performance and device, the method is extracted every the program statement that comprises in the professional program, and for every the program statement that extracts, carry out this program statement, obtain to carry out the first execution time corresponding to this program statement, according to benchmark execution time corresponding to this program statement that arranges, judge that whether the first execution time that obtains is greater than the benchmark execution time, if then send the alarm that this program statement of indication breaks down.Pass through said method, the ruuning situation of every the program statement that comprises in can the program to business is monitored, and when certain bar program statement breaks down, send the alarm that this program statement of indication breaks down, shortened greatly the time of location fault statement, and then improved the efficient of fixing a breakdown, improved quality of service.
Description of drawings
The process of the performance monitoring that Fig. 1 provides for the embodiment of the invention;
The detailed process of the performance monitoring that Fig. 2 provides for the embodiment of the invention;
The performance monitoring apparatus structural representation that Fig. 3 provides for the embodiment of the invention;
Fig. 4 specifically uses synoptic diagram for the performance monitoring apparatus that the embodiment of the invention provides;
Every program statement that Fig. 5 provides for the embodiment of the invention represent synoptic diagram;
Fig. 6 for the embodiment of the invention provide based on the monitoring certain bar program statement and break down of Fig. 5 the time the alarm synoptic diagram;
Fig. 7 represents synoptic diagram for the fault statement based on Fig. 6 that the embodiment of the invention provides.
Embodiment
Owing to can only monitor the ruuning situation of the whole program of certain business in the prior art, the granularity of its monitoring is thin not, when therefore certain the bar program statement that comprises in this program breaks down, can only monitor this whole program breaks down, can not providing specifically, which bar program statement breaks down, therefore the O﹠M personnel need to expend a large amount of time and locate the fault statement, thereby cause the Efficiency Decreasing of fixing a breakdown, and have also reduced quality of service.In order to shorten the time of location fault statement, improve the efficient of fixing a breakdown, the embodiment of the invention provides a kind of thinner method for monitoring performance of granularity of monitoring, every program statement in the program of business is monitored, when monitoring certain bar program statement and break down, send the alarm that this program statement of indication breaks down, make the O﹠M personnel can locate fast the fault statement, and carry out the eliminating of fault.
Below in conjunction with Figure of description, the embodiment of the invention is described in detail.
The process of the performance monitoring that Fig. 1 provides for the embodiment of the invention specifically may further comprise the steps:
S101: extract every the program statement that comprises in the professional program.
In embodiments of the present invention, each professional program can be write by various computereses.Take Structured Query Language (SQL) (Structured Query Language, SQL) as example, performance monitoring apparatus can extract every SQL statement that this professional program comprises from the database bottom.
Wherein, can adopt the dbms_application routine package to stamp label for each professional program, be used for distinguishing each professional program.For each professional program, can follow the tracks of by the V $ SESSION in the dbms_application routine package ruuning situation of this professional program, concrete, can in V $ SESSION, insert client_info, module, action field and this professional program be followed the tracks of realizing, and record every the SQL statement that comprises in this program.For example, in the time need to carrying out performance monitoring to the program of fee payment service, write the program name jfkj of this program by dbms_application_info.set_module, performance monitoring apparatus is then found the program jfkj of this fee payment service by scan database, and follow the tracks of the program jfkj of this fee payment service by dbms_application_info.read_module, record every the SQL statement that comprises in this program, as every the program statement that comprises in this program of extracting.
S102: for every the program statement that extracts, carry out this program statement, obtain to carry out the first execution time corresponding to this program statement.
In embodiments of the present invention, performance monitoring apparatus can only be carried out this program statement once, and obtain for the first execution time for every the program statement that extracts.
In order to improve the accuracy of monitoring, performance monitoring apparatus can also be in the monitoring period of setting, according to the time interval of setting, carry out this program statement, obtain each time and carry out each second execution time corresponding to this program statement, and the mean value of each second execution time that will obtain, as the first execution time corresponding to this program statement of execution that obtains.Wherein, the monitoring period of this setting and the time interval of setting can arrange as required, for example, monitoring period is set to 10 minutes, the time interval of this setting is 30 seconds, then performance monitoring apparatus is carried out this program statement 20 times altogether in this monitoring period, obtains 20 the second execution time.
And the method that obtained for the second execution time is specially: carry out this program statement, and execution start time and the complete time of this program statement of record execution, subtracted the difference of execution start time the complete time as the second execution time that obtains.
S103: according to benchmark execution time corresponding to this program statement that arranges, whether judge this first execution time that obtains greater than this benchmark execution time that arranges, if judged result is yes, execution in step S104 then, otherwise execution in step S105.
In embodiments of the present invention, the benchmark execution time that this program statement is corresponding can be set to: when writing this program, this program statement execution time under normal circumstances, be designated as the design execution time.Certainly, also can adjust this benchmark execution time according to actual conditions.
S104: send the alarm that this program statement of indication breaks down.
During the execution time, represent that this program statement breaks down greater than benchmark when the first execution time of judgement, therefore send alarm for this program statement, personnel get rid of the fault that this program statement occurs with the indication O﹠M.
S105: do not send alarm.
Be not more than benchmark during the execution time when the first execution time of judgement, represent that this program statement does not break down, therefore do not send alarm, wait for the arrival of next monitoring period, proceed monitoring.
In said process, extract every the program statement that comprises in the professional program, and for every the program statement that extracts, carry out this program statement, obtain to carry out the first execution time corresponding to this program statement, according to benchmark execution time corresponding to this program statement that arranges, judge that whether the first execution time that obtains is greater than the benchmark execution time, if then send the alarm that this program statement of indication breaks down.Pass through said method, can realize the method for monitoring performance that granularity is thinner, the ruuning situation of every the program statement that comprises in the program to business is monitored, and when certain bar program statement breaks down, send the alarm that this program statement of indication breaks down, shorten greatly the time of location fault statement, and then improved the efficient of fixing a breakdown, improved quality of service.
In addition, method for monitoring performance of the prior art is also too passive, when namely the application server side is moved certain professional program, this program is monitored.The method for monitoring performance that the embodiment of the invention provides, after performance monitoring apparatus extracts every the program statement that comprises in this program, regularly carry out every program statement, whether moving this program with application server has nothing to do, even therefore not yet move this program in the application server, perhaps this program of operation does not break down in the current application server, can server regular every the program statement of execution of do as one likes then can this program of discovery initiatively in potential fault, also namely can predict the program statement that may break down in this program, so that the O﹠M personnel can prevent the appearance of fault as early as possible, further improved quality of service.
In embodiments of the present invention, when the first execution time of the execution bar program statement that obtains is the mean value of each the second execution time corresponding to this program statement of each time execution, statistical error may appear in each second execution time, so that the first execution time is greater than reference time.For example certain is carried out the second execution time that this program statement obtains and increases suddenly so that as the mean value of each the second execution time of the first execution time greater than the benchmark execution time, thereby mistake is sent out an alarm.Cause mistake to send out alarm for fear of the statistical error that occurs, improve the accuracy of performance monitoring, in embodiments of the present invention, performance monitoring apparatus sends before the alarm of breaking down of this program statement of indication, also will be according to the principle of statistics test of hypothesis, verify that the first execution time was to cause owing to statistical error that each exists second execution time greater than the benchmark execution time, or fault causes owing to this program statement has occurred, when determining to cause for the first execution time greater than benchmark during the execution time owing to fault has appearred in this program statement, then send alarm, when determining that the statistical error owing to existing caused for the first execution time during the execution time, do not send alarm greater than benchmark.
Concrete, performance monitoring apparatus is according to the level of significance value that arranges, and the number of times of carrying out this program statement, determine critical value, and determine that the first execution time subtracted the difference of benchmark execution time, determine the standard deviation of each the second execution time, determine subduplicate first ratio of the number of times of this standard deviation and this program statement of execution, determine that the first execution time subtracted the difference of benchmark execution time and the second ratio of this first ratio, when this second ratio of determining during greater than this critical value of determining, send the alarm that this program statement of indication breaks down.
The below illustrates said process with a specific embodiment.
Suppose that the monitoring period of setting is 10 minutes, the time interval of setting is 30 seconds, and then performance monitoring apparatus is carried out this program statement 20 times altogether in this monitoring period, and the frequency n of namely carrying out this program statement is 20, has obtained 20 the second execution time.Wherein, 20 of acquisition the second execution time are as shown in table 1:
Table 1
Suppose that the level of significance value α that arranges is 0.05, then according to this level of significance value 0.05, and the number of times 20 of carrying out this program statement, can determine critical value t
α/2Be 2.093.Wherein, critical value t
α/2With level of significance value α, and it is relevant to carry out the frequency n of this program statement, can determine level of significance value 0.05 and carry out the critical value t of number of times 20 correspondences of this program statement by the mode of tabling look-up
α/2, also can determine by the mode of calculating.
By 20 the second execution time in the table 1, can obtain the mean value of 20 execution time
Be 5.5, should
As the first execution time.When supposing coding, this program statement design execution time under normal circumstances is 5 seconds, and should design the execution time as benchmark execution time μ, and this benchmark execution time, μ was 5 seconds, then determines for the first execution time
Subtract the difference of benchmark execution time μ
It is 0.5 second.
By 20 the second execution time in the table 1, can determine the standard deviation S of each the second execution time again, wherein, poor S can settle the standard by the stdev function among the excel.Determine the square root of the frequency n of this program statement of execution
And settle the standard poor S and this square root
The first ratio
Determine difference
With the first ratio
The second ratio, the second ratio is t, then have
Be 4.378.Because the critical value t that determines
α/2Be 2.093, therefore the second ratio t is greater than critical value t
α/2, then determined for the first execution time
(5.5 seconds) cause owing to fault has appearred in this program statement greater than benchmark execution time μ (5 seconds), therefore send the alarm that this program statement of indication breaks down.
For another example, suppose that the monitoring period of setting still is 10 minutes, the time interval of setting still is 30 seconds, and 20 the second execution time that performance monitoring apparatus obtains are as shown in table 2:
Table 2
Supposing that the level of significance value α that arranges still is 0.05, also do not have change owing to carry out the frequency n of this program statement, still is 20, so critical value t
α/2Still be 2.093.Suppose that benchmark execution time μ still is 5 seconds also, can be got for the first execution time by table 2
Be 5.01, still greater than benchmark execution time μ.But be based on table 2, by formula
The the second ratio t that calculates is 0.863, is not more than critical value t
α/2, therefore determined for the first execution time
(5.01 seconds) are to cause owing to statistical error that each second execution time that obtains exists greater than benchmark execution time μ (5 seconds), and breaking down not due to this program statement causes, and does not then send alarm.
Because in actual applications, for the same program statement, it must be discrepant carrying out the execution time that obtains in peak period of business and off-peak period, therefore in the embodiment of the invention in order further to improve the accuracy of performance monitoring, benchmark corresponding to this program statement is being set during the execution time, can be for different time ranges, for this program statement arranges the benchmark execution time that correspondence changes time range.For example, in certain month No. 1 to No. 10 time range, for benchmark execution time of this program statement setting is μ 1, in No. 11 to No. 20 the time range of this month, the benchmark execution time of setting is μ 2, in No. 21 to No. 30 the time range of this month, the benchmark execution time that arranges is μ 3, wherein, arranging of μ 1, μ 2, μ 3 can on the basis of design execution time of this program statement, be adjusted according to the service conditions in each time range.
Accordingly, when adopting said method that execution time corresponding to this program statement is set, in step S103 shown in Figure 1, performance monitoring apparatus is according to benchmark execution time corresponding to this program statement that arranges, whether the first execution time that judge to obtain is specially greater than the method for the benchmark execution time that arranges: the time range of determining to carry out the time place of this program statement, time range according to the time place of this program statement of execution of determining, be defined as that this program statement arranges to benchmark execution time that should time range, and judge that whether this first execution time that obtains is greater than the benchmark execution time of determining.
Continue to continue to use example, suppose that performance monitoring apparatus carries out this program statement No. 12 of this month, the time range of then determining the time place of this program statement of execution is No. 11 to No. 20 the time range of this month, according to this time range, the benchmark execution time that the correspondence that is defined as this program statement setting changes time range is μ 2, and then performance monitoring apparatus judges that whether the first execution time that obtains is greater than this benchmark execution time μ 2.
In step S104 shown in Figure 1, in order to make the alarm of sending more accurate, further improve the accuracy of performance monitoring, different alarm grades can be set, and exceed benchmark execution time number percent according to the first execution time, send the alarm of corresponding alarm grade.Concrete, performance monitoring apparatus determines that the first execution time subtracted the difference of benchmark execution time, determine that this difference accounts for the number percent of this benchmark execution time, according to the percentage range of dividing in advance, and this number percent of determining, determine the percentage range at this number percent place, according to the alarm grade of setting and the corresponding relation of percentage range, determine the alarm grade corresponding to percentage range at this number percent place, and send the alarm of the corresponding alarm grade that breaks down of this program statement of indication.
For example, the alarm grade that arranges for remind, general, serious Three Estate, corresponding percentage range is: less than the 10% corresponding alarm of reminding grade, be not less than 10% and less than the alarm of 25% corresponding general grade, be not less than the alarm of 25% corresponding menace level.Suppose the first definite execution time
Be 5.5, benchmark execution time μ is 5, the number percent of then determining
Be 10%, then corresponding percentage range is for being not less than 10% and less than 25%, corresponding alarm grade is general grade, therefore sending the alarm of indicating the general grade that this program statement breaks down.
The detailed process of the performance monitoring that Fig. 2 provides for the embodiment of the invention specifically may further comprise the steps:
S201: extract every the program statement that comprises in the professional program.
S202: for every the program statement that extracts, in the monitoring period of setting, according to the time interval of setting, carry out this program statement.
S203: obtain each time and carry out each second execution time corresponding to this program statement, with the mean value of each the second execution time of obtaining as the first execution time corresponding to this program statement of execution that obtains.
S204: determine to carry out the time range at the time place of this program statement, according to the time range at the time place of this program statement of execution of determining, be defined as that this program statement arranges to benchmark execution time that should time range.
S205: judge that whether the first execution time that obtains is greater than the benchmark execution time of determining, if then carry out step S206, otherwise carry out step S212.
S206: according to the level of significance value that arranges, and the number of times of carrying out this program statement, determine critical value.
S207: determine that the first execution time subtracted the difference of benchmark execution time, determine the standard deviation of each the second execution time, determine subduplicate first ratio of the number of times of this standard deviation and this program statement of execution, determine that the first execution time subtracted the difference of benchmark execution time and the second ratio of this first ratio.
S208: judge that whether the second ratio is greater than the critical value of determining, if then carry out step S209, otherwise carry out step S212.
S209: determine that the first execution time subtracted the difference of benchmark execution time and account for the number percent of this benchmark execution time, according to the percentage range of dividing in advance, and this number percent of determining, determine the percentage range at this number percent place.
S210: according to the alarm grade of setting and the corresponding relation of percentage range, determine the alarm grade corresponding to percentage range at this number percent place.
S211: the alarm of sending the corresponding alarm grade that breaks down of this program statement of indication.
S212: do not send alarm.
Fig. 3 is the performance monitoring apparatus structural representation that the embodiment of the invention provides, and specifically comprises:
Described execution module 302 specifically is used for, in the monitoring period of setting, according to the time interval of setting, carry out this program statement, obtain each time and carry out each second execution time corresponding to this program statement, and the mean value of each second execution time that will obtain, as the first execution time corresponding to this program statement of execution that obtains.
Described alarm module 304 specifically is used for, before sending the alarm of breaking down of this program statement of indication, and according to the level of significance value that arranges, and the number of times of carrying out this program statement, determine critical value; Determine that described the first execution time subtracts the difference of described benchmark execution time; And determine the standard deviation of each the second execution time, determine subduplicate first ratio of described standard deviation and the number of times of carrying out this program statement; Determine the second ratio of described difference and described the first ratio; When described the second ratio of determining during greater than the described critical value determined, send the alarm that this program statement of indication breaks down.
Described alarm module 304 specifically is used for, according to formula
Determine described the second ratio, wherein, t is described the second ratio,
Be described the first execution time, μ is the described benchmark execution time, and S is the standard deviation of described each the second execution time, and n is for carrying out the number of times of this program statement.
Described judge module 303 specifically is used for, determine the time range at the time place of this program statement of execution, time range according to the time place of this program statement of execution of determining, be defined as that this program statement arranges to benchmark execution time that should time range, and judge that whether described the first execution time that obtains is greater than the benchmark execution time of determining; Wherein, benchmark corresponding to this program statement being set during the execution time, for different time ranges, for this program statement arranged benchmark execution time that should time range.
Described alarm module 304 specifically is used for, determine that described the first execution time subtracts the difference of described benchmark execution time, determine that described difference accounts for the number percent of described benchmark execution time, according to the percentage range of dividing in advance, and the described number percent of determining, determine the percentage range at described number percent place, and according to the alarm grade of setting and the corresponding relation of percentage range, determine the alarm grade corresponding to percentage range at described number percent place, send the alarm of the corresponding alarm grade that breaks down of this program statement of indication.
Fig. 4 specifically uses synoptic diagram for the performance monitoring apparatus that the embodiment of the invention provides, and Fig. 4 is written as example with each professional program by SQL to describe.As shown in Figure 4, performance monitoring apparatus comprises that program statement extracts server, supervisory system database, performance evaluation server, alarm foreground workstation.Wherein, program statement extraction server is equivalent to the extraction module 301 among Fig. 3, the supervisory system database is equivalent to the execution module 302 among Fig. 3, and the performance evaluation server is equivalent to the judge module 303 among Fig. 3, and alarm foreground workstation is equivalent to the alarm module 304 among Fig. 3.This performance monitoring apparatus can also comprise the back-stage management workstation, manages for the task of the supervisory system database being carried out each bar program statement.
And program statement extracts server and links to each other with each professional database, in order to extract every SQL statement that each professional program comprises at the database bottom.
In embodiments of the present invention, performance monitoring apparatus can also be every the program statement that extracts, and according to the logical relation of every program statement, shows with the form of view.Every program statement that Fig. 5 provides for the embodiment of the invention represent synoptic diagram, in Fig. 5, after every the program statement that comprises in the program of performance monitoring apparatus with fee payment service extracts, for every program statement arranges sign, according to the logical relation of every program statement, the sign of every program statement is represented.
Fig. 6 for the embodiment of the invention provide based on the monitoring certain bar program statement and break down of Fig. 5 the time the alarm synoptic diagram, in Fig. 6, when monitoring certain bar program statement and break down, sign with this program statement in view changes color, as shown in Figure 6, when the program statement that monitors ZK.cm_busi_provision_wait_all sign breaks down, should identify the change color, to send the alarm of indicating this program statement to break down to the O﹠M personnel.The O﹠M personnel can click this sign, click sign after, view will change to the program statement of this sign correspondence, and the first execution time and the benchmark execution time of this program statement of monitoring of each monitoring period.
Fig. 7 represents synoptic diagram for the fault statement based on Fig. 6 that the embodiment of the invention provides, as shown in Figure 7, this view has represented the program name AccBill_PTK@server of the program at this program statement place, and represented this program statement that breaks down, and the first execution time and the benchmark execution time of this program statement of monitoring of each time.
The embodiment of the invention provides a kind of method for monitoring performance and device, the method is extracted every the program statement that comprises in the professional program, and for every the program statement that extracts, carry out this program statement, obtain to carry out the first execution time corresponding to this program statement, according to benchmark execution time corresponding to this program statement that arranges, judge that whether the first execution time that obtains is greater than the benchmark execution time, if then send the alarm that this program statement of indication breaks down.Pass through said method, the ruuning situation of every the program statement that comprises in can the program to business is monitored, and when certain bar program statement breaks down, send the alarm that this program statement of indication breaks down, shortened greatly the time of location fault statement, and then improved the efficient of fixing a breakdown, improved quality of service.
Obviously, those skilled in the art can carry out various changes and modification to the present invention and not break away from the spirit and scope of the present invention.Like this, if of the present invention these are revised and modification belongs within the scope of claim of the present invention and equivalent technologies thereof, then the present invention also is intended to comprise these changes and modification interior.
Claims (12)
1. a method for monitoring performance is characterized in that, comprising:
Extract every the program statement that comprises in the professional program; And
For every the program statement that extracts, carry out this program statement, obtain to carry out the first execution time corresponding to this program statement; And
According to benchmark execution time corresponding to this program statement that arranges, judge that whether described the first execution time that obtains is greater than the described benchmark execution time that arranges;
When judged result when being, send the alarm that this program statement of indication breaks down.
2. the method for claim 1 is characterized in that, carries out this program statement, obtains to carry out execution time corresponding to this program statement, specifically comprises:
In the monitoring period of setting, according to the time interval of setting, carry out this program statement; And
Obtain each time and carry out each second execution time corresponding to this program statement; And
With the mean value of each the second execution time of obtaining, as the first execution time corresponding to this program statement of execution that obtains.
3. method as claimed in claim 2 is characterized in that, sends before the alarm of breaking down of this program statement of indication, and described method also comprises:
According to the level of significance value that arranges, and the number of times of carrying out this program statement, determine critical value;
Determine that described the first execution time subtracts the difference of described benchmark execution time; And
Determine the standard deviation of each the second execution time, determine subduplicate first ratio of the number of times of described standard deviation and this program statement of execution;
Determine the second ratio of described difference and described the first ratio;
When described the second ratio of determining during greater than the described critical value determined, send the alarm that this program statement of indication breaks down.
4. method as claimed in claim 3 is characterized in that, determines described the second ratio according to following formula:
Wherein, t is described the second ratio,
Be described the first execution time, μ is the described benchmark execution time, and S is the standard deviation of described each the second execution time, and n is for carrying out the number of times of this program statement.
5. the method for claim 1 is characterized in that, benchmark execution time corresponding to this program statement is set, and specifically comprises:
For different time ranges, for this program statement arranged benchmark execution time that should time range;
According to benchmark execution time corresponding to this program statement that arranges, whether described the first execution time that judgement obtains specifically comprises greater than the described benchmark execution time that arranges:
Determine the time range at the time place of this program statement of execution; And
According to the time range at the time place of this program statement of execution of determining, be defined as that this program statement arranges to benchmark execution time that should time range; And
Judge that whether described the first execution time that obtains is greater than the benchmark execution time of determining.
6. the method for claim 1 is characterized in that, sends the alarm that this program statement of indication breaks down, and specifically comprises:
Determine that described the first execution time subtracts the difference of described benchmark execution time, determine that described difference accounts for the number percent of described benchmark execution time; And
According to the percentage range of dividing in advance, and the described number percent of determining, determine the percentage range at described number percent place; And
According to the alarm grade of setting and the corresponding relation of percentage range, determine the alarm grade corresponding to percentage range at described number percent place, and send the alarm of indicating the corresponding alarm grade that this program statement breaks down.
7. a performance monitoring apparatus is characterized in that, comprising:
Extraction module is used for extracting every program statement that professional program comprises;
Execution module is used for carrying out this program statement for every the program statement that extracts, and obtains to carry out the first execution time corresponding to this program statement;
Judge module was used for according to benchmark execution time corresponding to this program statement that arranges, and judged that whether described the first execution time that obtains is greater than the described benchmark execution time that arranges;
Alarm module, be used for when judged result when being, send the alarm of indicating this program statement to break down.
8. device as claimed in claim 7, it is characterized in that, described execution module specifically is used for, in the monitoring period of setting, according to the time interval of setting, carry out this program statement, obtain each time and carry out each second execution time corresponding to this program statement, and the mean value of each second execution time that will obtain, as the first execution time corresponding to this program statement of execution that obtains.
9. device as claimed in claim 8 is characterized in that, described alarm module specifically is used for, before sending the alarm of breaking down of this program statement of indication, and according to the level of significance value that arranges, and the number of times of carrying out this program statement, determine critical value; Determine that described the first execution time subtracts the difference of described benchmark execution time; And determine the standard deviation of each the second execution time, determine subduplicate first ratio of described standard deviation and the number of times of carrying out this program statement; Determine the second ratio of described difference and described the first ratio; When described the second ratio of determining during greater than the described critical value determined, send the alarm that this program statement of indication breaks down.
10. device as claimed in claim 9 is characterized in that, described alarm module specifically is used for, according to formula
Determine described the second ratio, wherein, t is described the second ratio,
Be described the first execution time, μ is the described benchmark execution time, and S is the standard deviation of described each the second execution time, and n is for carrying out the number of times of this program statement.
11. device as claimed in claim 7, it is characterized in that, described judge module specifically is used for, determine the time range at the time place of this program statement of execution, time range according to the time place of this program statement of execution of determining, be defined as that this program statement arranges to benchmark execution time that should time range, and judge that whether described the first execution time that obtains is greater than the benchmark execution time of determining; Wherein, benchmark corresponding to this program statement being set during the execution time, for different time ranges, for this program statement arranged benchmark execution time that should time range.
12. device as claimed in claim 7, it is characterized in that, described alarm module specifically is used for, determine that described the first execution time subtracts the difference of described benchmark execution time, determine that described difference accounts for the number percent of described benchmark execution time, according to the percentage range of dividing in advance, and the described number percent of determining, determine the percentage range at described number percent place, and according to the alarm grade of setting and the corresponding relation of percentage range, determine the alarm grade corresponding to percentage range at described number percent place, send the alarm of the corresponding alarm grade that breaks down of this program statement of indication.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110323591.2A CN103064776B (en) | 2011-10-21 | 2011-10-21 | Performance monitoring method and performance monitoring device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110323591.2A CN103064776B (en) | 2011-10-21 | 2011-10-21 | Performance monitoring method and performance monitoring device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103064776A true CN103064776A (en) | 2013-04-24 |
| CN103064776B CN103064776B (en) | 2015-07-01 |
Family
ID=48107409
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110323591.2A Active CN103064776B (en) | 2011-10-21 | 2011-10-21 | Performance monitoring method and performance monitoring device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103064776B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106407085A (en) * | 2016-11-24 | 2017-02-15 | 中国银行股份有限公司 | Performance monitoring method and apparatus |
| CN106649067A (en) * | 2016-12-28 | 2017-05-10 | 深圳先进技术研究院 | Performance and energy consumption prediction method and device |
| CN106921539A (en) * | 2017-02-06 | 2017-07-04 | 上海斐讯数据通信技术有限公司 | A kind of key business module monitor method and system based on cloud AC |
| CN107179975A (en) * | 2016-03-09 | 2017-09-19 | 北京京东尚科信息技术有限公司 | monitoring method and device |
| CN108388499A (en) * | 2018-02-24 | 2018-08-10 | 维沃移动通信有限公司 | A kind of memory monitoring method and terminal |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101226497A (en) * | 2007-12-10 | 2008-07-23 | 金蝶软件（中国）有限公司 | Application program monitoring system and monitoring method |
| CN101763389A (en) * | 2008-12-23 | 2010-06-30 | 中兴通讯股份有限公司 | Device and method for controlling database resource |
| 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 |
| CN102144222A (en) * | 2008-07-02 | 2011-08-03 | 国立大学法人东京工业大学 | Execution time estimation method, execution time estimation program, and execution time estimation device |
-
2011
- 2011-10-21 CN CN201110323591.2A patent/CN103064776B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101226497A (en) * | 2007-12-10 | 2008-07-23 | 金蝶软件（中国）有限公司 | Application program monitoring system and monitoring method |
| CN102144222A (en) * | 2008-07-02 | 2011-08-03 | 国立大学法人东京工业大学 | Execution time estimation method, execution time estimation program, and execution time estimation device |
| CN101763389A (en) * | 2008-12-23 | 2010-06-30 | 中兴通讯股份有限公司 | Device and method for controlling database resource |
| 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 |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107179975A (en) * | 2016-03-09 | 2017-09-19 | 北京京东尚科信息技术有限公司 | monitoring method and device |
| 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 |
| CN106649067A (en) * | 2016-12-28 | 2017-05-10 | 深圳先进技术研究院 | Performance and energy consumption prediction method and device |
| CN106649067B (en) * | 2016-12-28 | 2019-07-02 | 深圳先进技术研究院 | A kind of performance and energy consumption prediction technique and device |
| CN106921539A (en) * | 2017-02-06 | 2017-07-04 | 上海斐讯数据通信技术有限公司 | A kind of key business module monitor method and system based on cloud AC |
| CN108388499A (en) * | 2018-02-24 | 2018-08-10 | 维沃移动通信有限公司 | A kind of memory monitoring method and terminal |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103064776B (en) | 2015-07-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101088072B (en) | A method and system for monitoring transaction based systems | |
| CN103064776B (en) | Performance monitoring method and performance monitoring device | |
| EP2911060B1 (en) | Method and device for determining resource leakage and for predicting resource usage state | |
| EP2759938A1 (en) | Operations management device, operations management method, and program | |
| Carrozza et al. | Memory leak analysis of mission-critical middleware | |
| CN110888783A (en) | Monitoring method and device of micro-service system and electronic equipment | |
| GB2434670A (en) | Monitoring and management of distributed information systems | |
| CN106815125A (en) | A kind of log audit method and platform | |
| CN104182320A (en) | Method and device for monitoring leakage of memory | |
| CN103139007A (en) | Method and system for detecting application server performance | |
| CN106789251B (en) | System and method for monitoring online bank running state | |
| US11221903B2 (en) | Maintenance management system and maintenance management confirmation device used for the same | |
| CN107678908B (en) | Log recording method and device, computer equipment and storage medium | |
| CN104917836A (en) | Method and device for monitoring and analyzing availability of computing equipment based on cluster | |
| CN109002384A (en) | A kind of alarm method of server failure, device, equipment and storage medium | |
| US20140143625A1 (en) | Computer-readable recording medium, failure prediction device and applicability determination method | |
| CN107436414B (en) | Method and device for testing effectiveness of BBU in memory system | |
| CN2894106Y (en) | Computer network credibility estimating device based on event implanting | |
| US20180095819A1 (en) | Incident analysis program, incident analysis method, information processing device, service identification program, service identification method, and service identification device | |
| CN104917639A (en) | Cluster monitoring-based method and device for distributing data business | |
| CN102056200A (en) | Service processing flow monitoring method and system | |
| CN106468560B (en) | Data output method and device based on pedometer message | |
| CN112235262A (en) | Message analysis method and device, electronic equipment and computer readable storage medium | |
| CN105429792A (en) | User behavior flow obtaining method and device and user behavior analysis method and system | |
| CN105183627A (en) | Server performance prediction method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |