CN102981952A - Procedure performance analysis method based on target machine - Google Patents
Procedure performance analysis method based on target machine Download PDFInfo
- Publication number
- CN102981952A CN102981952A CN2012104397853A CN201210439785A CN102981952A CN 102981952 A CN102981952 A CN 102981952A CN 2012104397853 A CN2012104397853 A CN 2012104397853A CN 201210439785 A CN201210439785 A CN 201210439785A CN 102981952 A CN102981952 A CN 102981952A
- Authority
- CN
- China
- Prior art keywords
- program
- performance
- analyzed
- target machine
- characteristic
- 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
Abstract
The invention provides a procedure performance analysis method based on a target machine. The method comprises the steps of executing a procedure to be analyzed and acquiring characteristic data relative to performance index of time hot points in the procedure to be analyzed, acquiring a performance description standard and comparing the characteristic data with the performance description standard to obtain a performance bottleneck of the procedure to be analyzed aiming at the target machine, wherein the performance index is selected based on the target machine and a sampling procedure package and the performance description standard is reference data in view of the target machine and the sampling procedure package. The method is capable of finding out the performance bottleneck of the procedure to be analyzed aiming at the target machine.
Description
Technical field
The present invention relates to software testing technology, particularly a kind of performance analysis of program method of based target machine.
Background technology
The performance testing tool of concurrent program is mainly used in monitoring the performance parameter of concurrent program, such as execution time, call number, the utilization of resources etc. is for the further optimization of concurrent program, the raising of parallel algorithm quality and the improvement of system hardware provide foundation.
Yet, in the prior art, by analyzing the performance parameter of concurrent program, and then the method for the qualitative description that the performance of concurrent program is carried out shortcoming very still.In addition, along with the continuous expansion of concurrent program scale, the quantity of performance data is more and more.Thereupon, it is also increasing to analyze the difficulty of these mass performance datas.
In general, a concurrent program comes actual execution by a target machine the most at last, to realize user's appointed function.If before reality is carried out described concurrent program, the performance parameter of described concurrent program is not analyzed, to realize improvement and the optimization to described parallel program performance parameter, follow-up described concurrent program may take too much internal memory, take excessive bandwidth in target machine, cause executive capability low thereby produce excessive communication time-delay, even situation about can't carry out.
Therefore, remain to be proposed a kind of performance analysis of program method of based target machine, treat routine analyzer analysis for target machine, finding out the performance bottleneck that exists in the program to be analyzed, thereby realize treating improvement and the optimization of routine analyzer.
Summary of the invention
The problem that the present invention solves is the performance analysis of program method that proposes a kind of based target machine, treats routine analyzer analysis for target machine, finding out the performance bottleneck that exists in the program to be analyzed, thereby realizes treating improvement and the optimization of routine analyzer.
For addressing the above problem, embodiments of the invention provide a kind of performance analysis of program method of based target machine, described method comprises: carry out program to be analyzed, obtain the characteristic relevant with performance index of time focus in the program to be analyzed, described performance index are based on target machine and sample routine package and are selected; Obtain the performance specification standard corresponding with described performance index, described performance specification standard is for described target machine and the definite reference data of sample routine package; And compare described characteristic and described performance specification standard, to obtain described program pin to be analyzed to the performance bottleneck of described target machine.
Alternatively, the localization method of described time focus is: in described program process to be analyzed, monitor and add up by the time loss to each function of comprising in the described program to be analyzed, find out the function part that elapsed time in the whole program process surpasses preassigned, and be located the time focus into described program to be analyzed.
Alternatively, the method for obtaining the characteristic relevant with performance index of time focus in the program to be analyzed comprises: establish the counter that calculating behavior in Ge and the described time focus, memory access behavior and communication behavior are associated according to described performance index; And analyze the data that described counter is collected and added up, to obtain the described time focus characteristic relevant with described performance index.
Alternatively, the method for obtaining described performance specification standard comprises: filter out the sample program that approaches with described Program Type to be analyzed from described sample routine package; Obtain the characteristic relevant with described performance index of described sample program; And according to the characteristic relevant with described performance index of the described sample program of the characteristic information analysis of target machine, to determine the performance specification standard relevant with described performance index.
Alternatively, the method for obtaining described performance specification standard comprises: set up for the Program Type of described target machine and the data relationship model between the fixed reference feature data by the sample program of analyzing in the described sample routine package; The type of described program to be analyzed is inputted described data relationship model; And according to the fixed reference feature data of the described data relationship model data of the characteristic information analysis of target machine output, to determine the performance specification standard relevant with described performance index.
Alternatively, by comparing described characteristic and described performance specification standard, to obtain described program pin to be analyzed the method for the performance bottleneck of described target machine is comprised: by comparing described characteristic and described performance specification standard, to find out the characteristic that departs from described performance specification standard; And analyze the characteristic that departs from described performance specification standard, to obtain described program pin to be analyzed to the performance bottleneck of described target machine.
Alternatively, the performance analysis of program method of described based target machine also comprises: by analyzing the performance bottleneck information of described program to be analyzed, to obtain the performance specification of described program to be analyzed; And by analyzing the performance specification information of described program to be analyzed, to obtain the Optimizing Suggestions of described program performance bottleneck to be analyzed.
Alternatively, by analyzing the performance specification information of described program to be analyzed, method with the Optimizing Suggestions that obtains described program performance bottleneck to be analyzed comprises: rule of thumb the performance specification of described program to be analyzed is analyzed in the storehouse, to obtain the Optimizing Suggestions of described program performance bottleneck to be analyzed.
Alternatively, described sample routine package comprises the efficient sample program with reference significance.
Compared with prior art, embodiments of the invention have the following advantages: in an embodiment of the present invention,
At first, described performance index are for target machine and selected, fully taken into account the impact of target machine on program feature, therefore, can find out targetedly, for the characteristic parameter that needs in the described program to be analyzed of specific objective machine to be taken seriously, and it is chosen to be performance index.
Secondly, described performance specification standard is to obtain a reference standard of ideal performance at target machine according to the program that can make that target machine and sample data bag are determined, when determining described reference standard, the influence factor of considering target machine is also arranged, determined performance specification standard is had more with reference to meaning.
In addition, by comparing described characteristic and described performance specification standard, to obtain described program pin to be analyzed to the performance bottleneck of described target machine, so that the user treats routine analyzer a more directly perceived and deep understanding is arranged, thereby so that the user can take relevant Optimized Measures for described performance bottleneck.
Description of drawings
Fig. 1 is the process flow diagram of the performance analysis of program method of based target machine in one embodiment of the invention;
Fig. 2 concerns synoptic diagram between each key element in the performance analysis of program method of based target machine in one embodiment of the invention;
Fig. 3 is the method flow diagram that obtains the performance specification standard in one embodiment of the invention;
Fig. 4 is the method flow diagram that obtains the performance specification standard in another embodiment of the present invention;
Fig. 5 is the synoptic diagram of the data relationship model between one embodiment of the invention Program type and the fixed reference feature data.
Embodiment
The inventor finds after deliberation, same concurrent program is after the different target machine is carried out, the performance issue that occurs is different, also just says the different target machine because the difference of its software and hardware Pei Ge, and it may be not identical that the performance parameter of same type program is required.Therefore, if when the performance parameter of concurrent program is analyzed, do not consider the characteristic factor of target machine, follow-up when carrying out described concurrent program, inharmonic situation may appear between described concurrent program and the target machine.Thereby cause executive capability low, even the situation that program can't be carried out occurs.
For the problems referred to above, embodiments of the invention provide a kind of performance analysis of program method of based target machine.Fig. 1 is the process flow diagram of the performance analysis of program method of based target machine in one embodiment of the invention.
For above-mentioned purpose of the present invention, feature and advantage can be become apparent more, below in conjunction with accompanying drawing the specific embodiment of the present invention is described in detail.Set forth in the following description a lot of details so that fully understand the present invention, implement but the present invention can also adopt other to be different from alternate manner described here, so the present invention has not been subjected to the restriction of following public specific embodiment.
At first, please refer to Fig. 1, execution in step S1 carries out program to be analyzed, obtains the characteristic relevant with performance index of time focus in the program to be analyzed.
In an embodiment of the present invention, the localization method of described time focus is: in described program process to be analyzed, monitor and add up by the time loss to each function of comprising in the described program to be analyzed, find out that elapsed time exceeds the function part of preassigned in the whole program process, and be located the time focus into described program to be analyzed.In addition, in an embodiment of the present invention, described time focus can be one or more.
In one embodiment of the invention, focus of described time also can obtain by repeatedly carrying out described program to be analyzed.In addition, in the process of described time focus location, can progressively obtain whole program to be analyzed, certain process or thread, certain function, the execution time of certain fundamental block or certain program line.
Need to prove that as shown in Figure 2, described performance index are based on target machine and sample routine package and selected, described sample routine package comprises the efficient sample program that has in a large number reference significance.
At first, described performance index are to extract in the characteristic of the sample program that comprises according to described sample data bag, can reflect the key characterization parameter of the performance of program.
Specifically, for a concrete program example, reflect that the characteristic parameter of its performance is a lot, for example, the ratio that the time that program is carried out, the number of times of function call and communicating by letter takies etc.Although these characteristic parameters can both be on certain meaning the internal feature of reflection program operation, but, in an embodiment of the present invention, only select those performance bottleneck location for the treatment of routine analyzer to produce characteristic parameter key effect and that have the important analysis meaning as performance index.
Secondly, as previously mentioned, target machine is one of key factor that affects the program execution, and the execution result of same code on different target machines may be far from each other, and different target machines requires also may be not quite similar to the performance parameter of identical code.Therefore, the selection of described performance index need to be considered the characteristic factor of target machine, for example, and hardware configuration, system architecture etc.In an embodiment of the present invention, can for different target machines, select different performance index.
Particularly, in an embodiment of the present invention, the method for obtaining the characteristic relevant with performance index of time focus in the described program to be analyzed can comprise: the counter that is associated according to calculating behavior in described performance index setting and the described time focus, memory access behavior and communication behavior; And analyze the data that described counter is collected and added up, to obtain the described time focus characteristic relevant with described performance index.For example, in of the present invention one implemented, described performance index comprised the file read-write bandwidth, and the characteristic relevant with the file read-write bandwidth that obtains by She Ge counter is 300M/s.
Then, please continue with reference to figure 1, execution in step S2 obtains the performance specification standard corresponding with described performance index, and described performance specification standard is for described target machine and the definite reference data of sample routine package.
Please in conjunction with Fig. 2, in some embodiments of the invention, the method for obtaining described performance specification standard comprises: S201a, S202a and S203a.
At first, please refer to Fig. 3, carry out S201a, from described sample routine package, filter out the sample program that approaches with described Program Type to be analyzed.
In an embodiment of the present invention, described sample routine package comprises the efficient sample program that has in a large number reference significance, and the sample program in the described sample routine package is classified according to Program Type, therefore, can find out the efficient sample program with reference significance that approaches with described Program Type to be analyzed in the described sample routine package by screening.
Then, please refer to Fig. 3, carry out S202a, obtain the characteristic relevant with described performance index with the approaching sample program of Program Type to be analyzed.Because the sample program that Program Type described and to be analyzed approaches is according to the efficient program of investigating and research is selected, therefore, its characteristic relevant with performance index has important analogy to the performance of subsequent analysis program to be analyzed and with reference to meaning.
Then, please refer to Fig. 3, carry out S203a, and according to the characteristic relevant with described performance index of the described sample program of the characteristic information analysis of target machine, with definite performance specification standard relevant with described performance index.
In an embodiment of the present invention, the characteristic information of described target machine comprises that target machine affects the relevant software-hardware configuration information that program is carried out, such as memory size, and the network bandwidth, dominant frequency size, OS Type etc.Different target machines may be different to the requirement of the characteristic of same class program as previously mentioned, therefore, need to according to the characteristic information of target machine to above-mentioned have with reference to the characteristic of the sample program of meaning do some adjustment, it is had more with reference to being worth.In an embodiment of the present invention, when determining described performance specification index, can also analyze the universal law whether characteristic relevant with described performance index of described sample program is in order and is write, the universal law of described programming comprises form and the standard of programming, such as standard of function calling relationship etc.
In addition, in an embodiment of the present invention, when determining described performance specification index, all right programming experience with reference to senior programmer, thus give a clue for process analysis quickly.Described senior programmer's experience can be stored in the experience storehouse, because the universal law of each side parameter in senior programmer's understanding program, be used under normal circumstances the data with reference to meaning of having that the description program is carried out various aspects thereby can sum up better.
Described performance specification standard is to obtain a reference standard of ideal performance at target machine according to the program that can make that above-mentioned steps is determined.
Please in conjunction with Fig. 2, in other embodiment of the present invention, the method for obtaining described performance specification standard comprises: S201b, S202b and S203b.
At first, please refer to Fig. 4 and Fig. 5, carry out S201b, by analyzing sample program creation facilities program (CFP) type in the described sample routine package and the data relationship model between the fixed reference feature data.In embodiments of the present invention, described sample routine package comprises the efficient sample program that has in a large number reference significance, different with S201a is, need not the sample program in the described sample routine package is classified, can collect a large amount of described sample programs, the Program Type of the described sample program of statistics and analysis and the relation between the characteristic, thereby the data relationship model between creation facilities program (CFP) type and the fixed reference feature data.
Then, please refer to Fig. 4 and Fig. 5, carry out S202b, the type of described program to be analyzed is inputted described data relationship model, so that follow-up described data relationship model data output fixed reference feature data.
Then, please refer to Fig. 4 and Fig. 5, carry out S203b, according to the fixed reference feature data relevant with performance index of the described data relationship model data output of the characteristic information analysis of target machine, to determine the performance specification standard relevant with performance index.
As previously mentioned, except the characteristic information of target machine, when determining described performance specification standard, it is also conceivable that with reference to the universal law of programming and senior programmer's experience, further the fixed reference feature data relevant with performance index of described data relationship model data output are adjusted, thereby so that determined performance standard has reference value.
Described performance specification standard is to obtain a reference standard of ideal performance at target machine according to the program that can make that above-mentioned steps is determined.
Then, please continue with reference to figure 1, execution in step S3 compares described characteristic and described performance specification standard, to obtain described program pin to be analyzed to the performance bottleneck of described target machine.
Because described performance specification standard has with reference to being worth, if the characteristic relevant with performance index of program to be analyzed departed from described performance specification standard, can reflect that so described program to be analyzed can not obtain desirable performance at target machine, and the described characteristic that departs from the performance specification standard will be to cause program to be analyzed can not obtain at target machine the very crux of desirable performance, therefore, by the described characteristic that departs from described performance specification standard is analyzed, can obtain described program pin to be analyzed to the performance bottleneck of described target machine.
Please refer to Fig. 2, in an embodiment of the present invention, by comparing described characteristic and described performance specification standard, to obtain described program pin to be analyzed the method for the performance bottleneck of described target machine is comprised: by comparing described characteristic and described performance specification standard, to find out the characteristic that departs from described performance specification standard; And analyze the characteristic that departs from described performance specification standard, to obtain described program pin to be analyzed to the performance bottleneck of described target machine.For example, performance specification standard-required file read-write bandwidth is 400M/s, and the characteristic relevant with the file read-write bandwidth of program to be analyzed shows that its file read-write bandwidth is 300M/s, so, the file read-write bandwidth may be one of factor that causes the performance bottleneck that program to be analyzed carries out in target machine.
In embodiments of the present invention, described performance bottleneck may comprise the performance bottleneck of memory access, the performance bottleneck of read-write operation and performance bottleneck of communication etc.
Then, please refer to Fig. 1, execution in step S4 is by analyzing the performance bottleneck information of described program to be analyzed, to obtain the performance specification of described program to be analyzed.
Described performance specification is exactly to the location of performance of user program bottleneck and analysis, and it is the basis of making correct Optimizing Suggestions for program.In enforcement of the present invention, described performance specification is to treat the conclusion that the performance bottleneck of routine analyzer is analyzed, the performance bottleneck for the treatment of routine analyzer is classified, describe the induced factor of each type performance bottleneck in detail, explain the performance deficiency of program to be analyzed.
Treating after routine analyzer made performance specification, for the veteran programmer of minority, obtained the information of enough their improvement program codes.But, for the junior programmer of most of codings, still be necessary for the shortcut that they indicate update routine,
Therefore, then, please refer to Fig. 1, execution in step S5, by analyzing the performance specification information of described program to be analyzed, to obtain the Optimizing Suggestions of the performance bottleneck in the described program to be analyzed, described solution, be to be described as the suggestion that the improvement of program to be analyzed provides for performance, help the user optimization code.
In an embodiment of the present invention, by analyzing the performance specification information of described program to be analyzed, method with the Optimizing Suggestions that obtains described program performance bottleneck to be analyzed comprises: rule of thumb the performance specification of described program to be analyzed is analyzed in the storehouse, to obtain the Optimizing Suggestions of described program performance bottleneck to be analyzed.The working experience that comprises senior order person in the described experience storehouse can according to the programme performance bottleneck of problem of difference, provide corresponding innovative approach, thereby form described Optimizing Suggestions.
The above only is preferred embodiment of the present invention, is not the present invention is done any pro forma restriction.Any those of ordinary skill in the art, do not breaking away from the technical solution of the present invention scope situation, all can utilize method and the technology contents of above-mentioned announcement that technical solution of the present invention is made many possible changes and modification, or be revised as the equivalent embodiment of equivalent variations.Therefore, every content that does not break away from technical solution of the present invention according to any simple modification, equivalent variations and the modification that technical spirit of the present invention is done above embodiment, all still belongs in the scope of technical solution of the present invention protection.
Claims (9)
1. the performance analysis of program method of a based target machine is characterized in that, comprising:
Carry out program to be analyzed, obtain the characteristic relevant with performance index of time focus in the program to be analyzed, described performance index are based on target machine and sample routine package and are selected;
Obtain the performance specification standard corresponding with described performance index, described performance specification standard is for described target machine and the definite reference data of sample routine package; And
Compare described characteristic and described performance specification standard, to obtain described program pin to be analyzed to the performance bottleneck of described target machine.
2. the method for claim 1, it is characterized in that, the localization method of described time focus is: in described program process to be analyzed, monitor and add up by the time loss to each function of comprising in the described program to be analyzed, find out the function part that elapsed time in the whole program process surpasses preassigned, and be located the time focus into described program to be analyzed.
3. the method for claim 1, it is characterized in that the method for obtaining the characteristic relevant with performance index of time focus in the program to be analyzed comprises: establish the counter that calculating behavior in Ge and the described time focus, memory access behavior and communication behavior are associated according to described performance index; And analyze the data that described counter is collected and added up, to obtain the described time focus characteristic relevant with described performance index.
4. the method for claim 1 is characterized in that, the method for obtaining described performance specification standard comprises: filter out the sample program that approaches with described Program Type to be analyzed from described sample routine package; Obtain the characteristic relevant with described performance index of described sample program; And according to the characteristic relevant with described performance index of the described sample program of the characteristic information analysis of target machine, to determine the performance specification standard relevant with described performance index.
5. the method for claim 1, it is characterized in that the method for obtaining described performance specification standard comprises: set up for the Program Type of described target machine and the data relationship model between the fixed reference feature data by the sample program of analyzing in the described sample routine package; The type of described program to be analyzed is inputted described data relationship model; And according to the fixed reference feature data of the described data relationship model data of the characteristic information analysis of target machine output, to determine the performance specification standard relevant with described performance index.
6. the method for claim 1, it is characterized in that, by comparing described characteristic and described performance specification standard, to obtain described program pin to be analyzed the method for the performance bottleneck of described target machine is comprised: by comparing described characteristic and described performance specification standard, to find out the characteristic that departs from described performance specification standard; And analyze the characteristic that departs from described performance specification standard, to obtain described program pin to be analyzed to the performance bottleneck of described target machine.
7. the method for claim 1 is characterized in that, also comprises: by analyzing the performance bottleneck information of described program to be analyzed, to obtain the performance specification of described program to be analyzed; And by analyzing the performance specification information of described program to be analyzed, to obtain the Optimizing Suggestions of described program performance bottleneck to be analyzed.
8. method as claimed in claim 7, it is characterized in that, by analyzing the performance specification information of described program to be analyzed, method with the Optimizing Suggestions that obtains described program performance bottleneck to be analyzed comprises: rule of thumb the performance specification of described program to be analyzed is analyzed in the storehouse, to obtain the Optimizing Suggestions of described program performance bottleneck to be analyzed.
9. the method for claim 1 is characterized in that, described sample routine package comprises the efficient sample program with reference significance.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210439785.3A CN102981952B (en) | 2012-11-06 | 2012-11-06 | Procedure performance analysis method based on target machine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210439785.3A CN102981952B (en) | 2012-11-06 | 2012-11-06 | Procedure performance analysis method based on target machine |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102981952A true CN102981952A (en) | 2013-03-20 |
| CN102981952B CN102981952B (en) | 2015-05-20 |
Family
ID=47856006
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210439785.3A Active CN102981952B (en) | 2012-11-06 | 2012-11-06 | Procedure performance analysis method based on target machine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102981952B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112463569A (en) * | 2020-12-11 | 2021-03-09 | 苏州浪潮智能科技有限公司 | Internal performance evaluation method and system of concurrent system |
| 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 |
| CN113656292A (en) * | 2021-08-04 | 2021-11-16 | 航天中认软件测评科技（北京）有限责任公司 | Multi-dimensional cross-space-time basic software performance bottleneck detection method |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1564136A (en) * | 2004-04-02 | 2005-01-12 | 清华大学 | Realizing method of cross regulator based on EJTAG components of targeting machine |
-
2012
- 2012-11-06 CN CN201210439785.3A patent/CN102981952B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1564136A (en) * | 2004-04-02 | 2005-01-12 | 清华大学 | Realizing method of cross regulator based on EJTAG components of targeting machine |
Non-Patent Citations (2)
| Title |
|---|
| 程克非等: "基于硬件性能计数器的软件性能数据采集与分析研究", 《计算机应用》, vol. 25, no. 10, 31 October 2005 (2005-10-31) * |
| 陈岚: "基于硬件性能监视的性能测试误差分析技术", 《中国优秀硕士学位论文全文数据库•信息科技辑》, no. 11, 15 November 2006 (2006-11-15) * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112463569A (en) * | 2020-12-11 | 2021-03-09 | 苏州浪潮智能科技有限公司 | Internal performance evaluation method and system of concurrent system |
| CN112463569B (en) * | 2020-12-11 | 2023-01-06 | 苏州浪潮智能科技有限公司 | Internal performance evaluation method and system of concurrent system |
| CN113656292A (en) * | 2021-08-04 | 2021-11-16 | 航天中认软件测评科技（北京）有限责任公司 | Multi-dimensional cross-space-time basic software performance bottleneck detection method |
| 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 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102981952B (en) | 2015-05-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107220121B (en) | Sandbox environment testing method and system under NUMA architecture | |
| CN103136471B (en) | A kind of malice Android application program detection method and system | |
| TWI684151B (en) | Method and device for detecting illegal transaction | |
| US9864615B2 (en) | User driven emulation of applications | |
| CN109933984B (en) | Optimal clustering result screening method and device and electronic equipment | |
| CN106022505A (en) | Method and device of predicting user off-grid | |
| CN110532168B (en) | Full-link interface pressure measurement method and device, computer equipment and storage medium | |
| WO2016008398A1 (en) | Program performance test method and device | |
| CN106407768B (en) | A kind of determination of device-fingerprint, the method and apparatus that target device is identified | |
| CN108292346A (en) | The extracts physical access control policy from static rights and Access Events | |
| CN110688168A (en) | Method, device and equipment for improving starting speed of application program and storage medium | |
| CN105721406A (en) | Method and device for obtaining IP black list | |
| CN104778185A (en) | Determination method for abnormal SQL (structured query language) statement and server | |
| CN102981952B (en) | Procedure performance analysis method based on target machine | |
| CN103559330B (en) | Method and system for detecting data consistency | |
| CN106909454A (en) | A kind of rules process method and equipment | |
| CN115130737A (en) | Machine learning method based on working condition, related device and medium program product | |
| CN111091863A (en) | Storage equipment fault detection method and related device | |
| CN112346962A (en) | Comparison data testing method and device applied to comparison testing system | |
| CN111949548B (en) | Automatic unauthorized penetration testing method and storage device | |
| CN110059083A (en) | A kind of data evaluation method, apparatus and electronic equipment | |
| JP7367196B2 (en) | Methods and systems for identification and analysis of regime shifts | |
| CN110826583A (en) | Fault determination method and device, storage medium and electronic device | |
| CN107168778B (en) | Task processing method and task processing device | |
| CN103761432B (en) | Method and device for analyzing board-level test data |
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 |