WO2012086444A1 - 監視データ分析装置、監視データ分析方法および監視データ分析プログラム - Google Patents
監視データ分析装置、監視データ分析方法および監視データ分析プログラム Download PDFInfo
- Publication number
- WO2012086444A1 WO2012086444A1 PCT/JP2011/078714 JP2011078714W WO2012086444A1 WO 2012086444 A1 WO2012086444 A1 WO 2012086444A1 JP 2011078714 W JP2011078714 W JP 2011078714W WO 2012086444 A1 WO2012086444 A1 WO 2012086444A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- data
- regression model
- monitoring
- group
- data item
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3409—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment for performance assessment
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3447—Performance evaluation by modeling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3452—Performance evaluation by statistical analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3466—Performance evaluation by tracing or monitoring
- G06F11/3485—Performance evaluation by tracing or monitoring for I/O devices
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3466—Performance evaluation by tracing or monitoring
- G06F11/3476—Data logging
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2201/00—Indexing scheme relating to error detection, to error correction, and to monitoring
- G06F2201/81—Threshold
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2201/00—Indexing scheme relating to error detection, to error correction, and to monitoring
- G06F2201/88—Monitoring involving counting
Definitions
- the present invention relates to a monitoring data analysis device, a monitoring data analysis method, and a monitoring data analysis program.
- Patent Document 1 discloses a technique related to a performance monitoring apparatus for a WWW site.
- some systems have a plurality of usage patterns, such as being frequently used by data retrieval such as browsing processing in the daytime and frequently used by data updating such as batch processing at night.
- the conventional performance management method when predicting the disk usage rate for the number of operation requests such as reading and writing as performance values, the conventional performance management method generates a regression model using monitoring data in all usage patterns. And predict.
- the disk usage rate varies greatly depending on the usage pattern, in the conventional method, an error between the predicted value based on the regression model and the actual value increases, and the prediction accuracy of the performance value decreases. .
- An object of the present invention is to provide a data analysis device, a monitoring data analysis method, and a monitoring data analysis program.
- the monitoring data analysis apparatus includes a regression model generated by classifying log data including monitoring data in a monitoring target system that is a target of performance management into a plurality of groups.
- a regression model recalculation unit that recalculates the regression model by combining the log data belonging to the group corresponding to and the inspection target log data that is the log data to be subjected to performance inspection. .
- log data including monitoring data in a monitoring target system that is a target of performance management is classified into a plurality of groups, and regression models generated for the respective groups are converted into the regression models.
- the monitoring data analysis program of the present invention causes a computer to execute the steps included in the monitoring data analysis method.
- the prediction accuracy of the performance value can be improved.
- FIG. 1st embodiment It is the figure which illustrated the composition of the monitoring data analysis device in a 1st embodiment. It is a figure which illustrates the data structure of a log data file. It is a graph which illustrates a characteristic data item. 3 is a diagram illustrating elements of a matrix A. FIG. It is a figure which shows an example of test object log data. It is a flowchart for demonstrating the operation
- a monitoring data analysis apparatus is an apparatus that analyzes monitoring data in a monitoring target system that is a target of performance management.
- the monitoring data analysis device physically includes, for example, a CPU (Central Processing Unit), a storage device, and an input / output interface.
- the storage device includes, for example, a ROM (Read Only Memory) and HDD (Hard Disk Drive) for storing programs and data processed by the CPU, and a RAM (Random Access Memory) mainly used as various work areas for control processing. ) Etc. are included. These elements are connected to each other via a bus.
- the CPU executes the program stored in the ROM and processes the message received via the input / output interface and the data expanded in the RAM, thereby realizing the functions of the respective units in the monitoring data analysis device described later. be able to.
- the monitoring data analysis device 1 functionally includes, for example, a feature data item extraction unit 13, a regression model recalculation unit 14, and a performance value calculation unit 15.
- the log data file 21 for accumulating log data and the regression model file 22 for storing the regression model are provided as files referred to by each unit.
- the log data is data in which monitoring data in the monitoring target system is recorded in time series.
- the regression model is generated by performing regression analysis on each log data classified into groups for each usage pattern.
- the usage pattern for example, a data search type (reading type) or a data update type (writing type) is applicable.
- the regression analysis is performed in advance, and a plurality of regression models are stored in the regression model file 22.
- the data structure of the log data file 21 will be described with reference to FIG.
- the log data file 21 includes, for example, a time division item, a disk usage rate item, a request count item, a write count item, a read count item, and an average data length item as data items.
- the log data file 21 generates one record (hereinafter referred to as “log record”) for each time division item.
- the time classification item stores classification information for identifying the time zone to which each data of the log record belongs.
- the disk usage rate item stores the usage rate of a disk (for example, HDD) in the monitored system.
- the disk usage rate is used as the resource usage rate of the monitoring target system.
- the resource usage rate is not limited to the disk usage rate.
- data indicating the utilization status of the components of the monitoring target system can be used.
- a CPU utilization rate for example, a memory (for example, RAM) utilization rate, a network utilization rate, and the like are applicable.
- the number of requests field stores the number of requests received by the monitored system.
- the number of requests is not limited, and for example, the arrival rate of requests may be used.
- the write count item stores the number of times data has been written to the monitored system disk.
- the read count item stores the number of times data is read from the disk of the monitoring target system.
- the average data length item stores the average value of the data length of each request received in the target time zone in the target time zone.
- the feature data item extraction unit 13 calculates, for each group classified at the time of regression analysis, the dependence of each data item on the explanatory variable for each data item included in the log data forming the group.
- the feature data item extraction unit 13 extracts data items whose dependency is higher than a predetermined threshold as feature data items.
- the feature data item extraction unit 13 targets all groups X classified at the time of regression analysis, and for each data item a j of the log data, the value x of the data item a j .
- a j is the value of the explanatory variable of the regression model (number of requests) x.
- a variance d j of the value divided by ⁇ (hereinafter referred to as “variance d j in the entire group”) is calculated.
- Feature data item extraction section 13 as shown in the following formula (2), and each group X i classified during regression analysis target, for each data item a j log data belonging to each group X i, data Value of item a j x.
- a j is the value of the explanatory variable of the regression model (number of requests) x.
- the variance d ji of the value divided by ⁇ is calculated for each group X i .
- Feature data item extraction section 13 as shown in the following formula (3), the dispersion d ji of each data item a j calculated for each group X i by aggregated in all groups X, for each group X i A total d j ′ of the variance d ji in the entire group (hereinafter referred to as “total variance d j ′ in each group”) is calculated.
- Feature data item extraction section 13 the 'case where the value obtained by dividing the dispersion d j of the entire group is smaller than the predetermined threshold value k (d j' sum d j of the dispersion / d j ⁇ k) in each group,
- the data item a j that satisfies the condition is determined to have a high dependency on the explanatory variable, and the data item a j is extracted as a feature data item.
- FIG. 3 shows a bar graph representing the value of (d j ′ / d j ) for each data item a j .
- each data item a j is arranged on the horizontal axis, and the value of (d j ′ / d j ) is shown on the vertical axis.
- the threshold value k is set to 0.3.
- the data items whose (d j ′ / d j ) value is smaller than 0.3 are “write count item” and “read count item”. Therefore, the “write count item” and the “read count item” are extracted as feature data items.
- the regression model recalculation unit 14 shown in FIG. 1 synthesizes a plurality of regression models stored in the regression model file 22 with reference to the feature data items extracted by the feature data item extraction unit 13, thereby Calculate (recalculate) a regression model to be used when checking system performance.
- the known data item can be used instead of referring to the feature data item extracted by the feature data item extraction unit 13.
- the “write count item” and the “read count item” are characteristic data items.
- the regression model recalculation unit 14 calculates the ratio of the value of each feature data item with respect to the explanatory variable for each group, and synthesizes the regression model of each group using the calculated ratio.
- a regression model for each feature data item which is a regression model for the data item, is calculated.
- the regression model recalculation unit 14 uses each calculated feature data item regression model for each feature data included in log data (hereinafter referred to as “test target log data”) to be subjected to performance inspection in the monitoring target system.
- a regression model is calculated by combining them according to the appearance rate of the item values.
- the characteristic data item regression model is calculated as follows, for example.
- the regression model recalculation unit 14 when the number of classified groups is m and the number of extracted feature data items is n, the regression model recalculation unit 14 generates a matrix A having m rows and n columns. In each element (i, j) of the matrix A, an average value in the group X i of values obtained by dividing the value of the feature data item a j by the explanatory variable (number of requests ⁇ ) is arranged.
- the classified groups are two groups, “group 1” and “group 2”, and the extracted feature data items are “write count item” and “read count item”.
- the matrix A is a 2 ⁇ 2 matrix.
- Each data column shown in FIG. 4 stores an average value in each group of values obtained by dividing the value of each feature data item by the number of requests ⁇ . That is, the value of each data column becomes each element (i, j) of the matrix A.
- the matrix A in this case is expressed as follows.
- the regression model recalculation unit 14 calculates a pseudo inverse matrix A + of the matrix A.
- a technique for calculating the pseudo inverse matrix for example, a technique described in Reference Document 1 below can be used.
- the regression model recalculation unit 14 sets the regression model h aj ( ⁇ ) when the value of a certain feature data item a j is “1” and the value of another feature data item is “0” as follows:
- F i ( ⁇ ) in the following formula (4) is a regression model of the group X i generated by the regression model generation unit 12.
- the regression model recalculation unit 14 calculates the average value b j in the inspection target log data of the value obtained by dividing the value of each feature data item a j included in the inspection target log data by the explanatory variable (number of requests ⁇ ). calculate.
- the regression model recalculation unit 14 generates a regression model f ( ⁇ ) using the following equation (5). That is, the regression model used when performing a performance test based on the inspection target log data is recalculated using a plurality of existing regression models.
- the inspection target log data is the log data shown in FIG.
- the feature data item extraction unit 13 executes a feature data item extraction process to be described later (step S101).
- the regression model recalculation unit 14 executes a regression model recalculation process described later (step S102).
- the performance value calculation unit 15 calculates the disk usage rate corresponding to the assumed number of requests using the regression model f ( ⁇ ) recalculated by the regression model recalculation unit 14 (step S103).
- the assumed number of requests can be arbitrarily set by the administrator in the monitoring data analyzer.
- step S101 With reference to FIG. 7, the feature data item extraction process executed in step S101 will be described.
- step S201 to step S204 described below is executed in a loop for each data item aj in order for all data items aj included in the log data.
- the feature data item extraction unit 13 uses the above formula (1) to divide the value of the target data item a j by the number of requests ⁇ for the log data of all groups X classified at the time of regression analysis.
- a value variance dj is calculated (step S201).
- the feature data item extraction unit 13 using the above equation (2), the log data of each group X i in the subject, the dispersion d ji of a value obtained by dividing the value of the target data item a j in the request number ⁇ Is calculated for each group X i , and the total d j ′ for the entire group of variance d ji is calculated using the above equation (3) (step S202).
- the feature data item extraction unit 13 determines whether or not a value obtained by dividing the sum d j ′ calculated in step S202 by the variance d j calculated in step S201 is smaller than a threshold value k ( Step S203). If this determination is NO (step S203; NO), the process proceeds to the subsequent stage of step S204.
- step S203 determines that the value of (d j ′ / d j ) is smaller than the threshold value k (step S203; YES).
- step S102 the regression model recalculation process executed in step S102 will be described.
- the regression model recalculation unit 14 calculates the average value in the group X i of the values obtained by dividing the value of each feature data item a j extracted in step S101 by the explanatory variable (number of requests ⁇ ). Then, a matrix A having m rows ⁇ groups of feature data items is generated (step S301).
- the regression model recalculation unit 14 calculates a pseudo inverse matrix A + of the matrix A (step S302).
- the regression model recalculation unit 14 calculates a regression model h aj ( ⁇ ) for each feature data item using the above equation (4) (step S303).
- the regression model recalculation unit 14 calculates an average value b j in the inspection target log data of a value obtained by dividing the value of each feature data item a j included in the inspection target log data by the number of requests ⁇ (step S1). S304).
- the regression model recalculation unit 14 substitutes the feature data item-specific regression model h aj ( ⁇ ) calculated in step S303 and the average value b j calculated in step S304 into the equation (5).
- the regression model f ( ⁇ ) for the inspection target log data is generated (recalculated) (step S305).
- a plurality of existing regression models are synthesized according to the ratio of the value of each feature data item included in the inspection target log data, and a new regression is performed.
- a model can be generated (recalculated).
- a regression model adapted to the assumed usage pattern can be generated, and the resource usage rate such as the disk usage rate can be predicted using this regression model. Therefore, even when the resource usage rate such as the disk usage rate is greatly different for each usage pattern, the prediction accuracy of the resource usage rate can be improved.
- the feature data item extraction unit 13 in the first embodiment described above extracts feature data items using the dependence of each data item on the explanatory variable. It is not limited to. For example, the following extraction method can be used.
- the feature data item extraction unit 13 For each group X i classified at the time of regression analysis, the feature data item extraction unit 13 for each data item a j included in the log data forming the group, each data item a j and explanatory variable (number of requests ⁇ ) It calculates an average value of a group X i of the correlation coefficient between. The feature data item extraction unit 13 extracts a data item a j whose calculated average value is equal to or greater than a predetermined threshold k ′ as a feature data item.
- the average value of the correlation coefficient between each data item a j and the explanatory variable can be calculated by the following procedure, for example.
- the feature data item extraction unit 13 calculates the correlation coefficient r ji between each data item a j and the explanatory variable (number of requests ⁇ ) for each group X i using the following equation (6). .
- a j is the value of the data item a j of the k-th log record belonging to the group X i
- ⁇ is the value of the request number item of the k-th log record belonging to the group X i
- m is the number of groups.
- the feature data item extraction unit 13 divides the correlation coefficient r ji for each group X i by all the groups X by the number of groups m, so that each data item a j and the explanatory variable are divided.
- the average value r j of the correlation coefficient is calculated.
- the feature data item extraction unit 13 extracts a data item a j in which
- the feature data item extraction unit 13 extracts the feature data item, but the present invention is not limited to this.
- the administrator may set the feature data item in advance in the memory. In this case, the feature data item extraction unit 13 can be omitted.
- the regression model recalculation unit 14 may acquire the value of the feature data item with reference to the memory.
- the performance value calculation unit 15 is provided in the same device as the feature data item extraction unit 13 and the regression model recalculation unit 14, but the performance value calculation unit 15 is provided in an external device. It is good also as providing in.
- the monitoring data analysis device of the second embodiment includes a data classification unit 11 and a regression model generation unit in addition to the various functions of the monitoring data analysis device (see FIG. 1) of the first embodiment described above. 12 and the abnormality determination unit 16 are different from the monitoring data analyzer of the first embodiment. Since other configurations are the same as those of the monitoring data analysis apparatus of the first embodiment, the same reference numerals are given to the respective components, and the description thereof is omitted. In the following, differences from the first embodiment will be mainly described.
- the data classification unit 11 classifies the log data stored in the log data file 21 into groups for each usage pattern (for example, data search type or data update type).
- a known clustering method such as a shortest distance method, a longest distance method, a group average method, or a ward method can be used.
- the characteristic of the disk usage rate ⁇ with respect to the number of requests ⁇ is used as a characteristic for classification into groups.
- log data is placed on the coordinate plane shown in FIG. 10 in units of log records.
- the horizontal axis in FIG. 10 indicates the number of requests ⁇ , and the vertical axis indicates the disk usage rate ⁇ .
- the data arranged on the coordinate plane are sequentially collected based on the characteristics of the disk usage rate ⁇ with respect to the number of requests ⁇ .
- it is finally classified into a first group G1 and a second group G2 shown in FIG.
- log data belonging to the data update type usage pattern is classified into the first group G1
- log data belonging to the data search type usage pattern is classified into the second group G2.
- the regression model generation unit 12 shown in FIG. 9 performs a regression analysis of log data for each classified group to generate a regression model.
- a method for generating a regression model for example, the method described in Reference Document 2 below can be used.
- the regression model generation unit 12 performs a regression analysis using the disk usage rate ⁇ as an objective variable for regression analysis and the number of requests ⁇ as an explanatory variable for regression analysis.
- a regression model represented by the following equation (7) is generated as the regression model M1 of the log data belonging to the first group G1, and the log belonging to the second group G2
- a regression model represented by the following equation (8) is generated.
- the regression model generation unit 12 stores the generated regression model M1 and regression model M2 in the regression model file 22.
- the abnormality determination unit 16 illustrated in FIG. 9 uses the value f ( ⁇ ′) obtained by substituting the value ⁇ ′ of the request number item into the regression model f ( ⁇ ) for each log record of the inspection target log data, and the disk usage. It is determined whether or not the difference (
- the presence or absence of an abnormality in the monitored system can be determined as follows.
- the abnormality determination unit 16 obtains the variance V ( ⁇ ) with respect to the objective variable (disk utilization rate ⁇ ) of the regression model f ( ⁇ ) recalculated by the regression model recalculation unit 14, and calculates the standard deviation ( ⁇ V). calculate.
- the abnormality determination unit 16 calculates
- the value of the threshold value a can be set arbitrarily. In other words, the abnormality determination unit 16 determines the presence or absence of abnormality based on how far the value of
- the variance V ( ⁇ ) can be obtained, for example, by calculating the following formula (10) derived from the following formula (9).
- the regression model f ( ⁇ ) in the following equation (9) can be derived from the above equations (4) and (5), and the variance V ( ⁇ ) in the following equation (10) can be derived from the following equation (9). Can be derived.
- V ( ⁇ j ) in the above equation (10) is a variance related to the objective variable (disk utilization rate ⁇ ) of the regression model calculated for each group X i .
- ⁇ j is a random variable indicating the distribution of the objective variable in the group X i .
- the operation of the monitoring data analyzer in the second embodiment will be described.
- the operation when calculating the disk usage rate using the regression model is the same as the operation of the monitoring data analyzer in the first embodiment described above (see FIGS. 6 to 8). Omitted.
- the data classification unit 11 clusters the log data stored in the log data file 21 and classifies the log data into a plurality of groups (step S401).
- the regression model generation unit 12 performs a regression analysis of the log data for each classified group, and generates a regression model (step S402).
- the generated regression model can be used, for example, when calculating a resource usage rate such as a disk usage rate or determining an abnormality of the monitoring target system, as will be described later.
- the abnormality determination process is executed after step S102 of FIG. 6 described above is executed.
- the abnormality determination process may be executed in parallel with step S103 of FIG. 6, or may be executed before or after step S103.
- the abnormality determining unit 16 refers to the recalculated regression model f ( ⁇ ) recalculated by the regression model recalculating unit 14, and calculates the standard deviation ( ⁇ V) regarding the objective variable (disk utilization rate ⁇ ). (Step S501).
- step S502 to step S503 described below is executed by sequentially looping for each log record for all log records included in the inspection target log data.
- the abnormality determination unit 16 substitutes the value ⁇ ′ of the request count item of the inspection target log data into the recalculation regression model f ( ⁇ ) and the value f ( ⁇ ′) of the inspection target log data.
- a difference from the value ⁇ ′ of the disk usage rate item is obtained, and a value obtained by dividing the difference by the standard deviation ( ⁇ V) calculated in step S501 is calculated (step S502).
- the abnormality determination unit 16 determines whether or not the value calculated in step S502 is greater than the threshold value a (step S503). If this determination is NO (step S503; NO), the process proceeds to the subsequent stage.
- step S503 when it is determined in step S503 that the value calculated in step S502 is greater than the threshold value a (step S503; YES), the abnormality determination unit 16 determines that there is an abnormality in the monitored system. Then, the administrator is notified of the abnormality (step S504).
- a method for notifying an administrator of abnormality for example, a method for outputting a message to that effect, a method for outputting a warning sound, or the like is applicable.
- the log data is in accordance with the usage pattern such as the data update type and the data search type. It can be classified into groups, and a regression model can be generated for each classified group. As a result, even when the resource usage rate such as the disk usage rate is greatly different for each usage pattern, it is possible to generate regression models corresponding to the usage patterns. Therefore, even when the monitoring target system has a plurality of usage patterns, errors in prediction using the regression model can be reduced.
- the abnormality of the monitored system can be found based on the difference from the regression model adapted to the assumed usage pattern, it is possible to improve the accuracy of finding the abnormality.
- the present invention is implemented by a single monitoring data analysis apparatus, but the present invention can also be implemented by a plurality of apparatuses.
- the functions of the monitoring data analysis device in each of the above-described embodiments are distributed to a plurality of devices, and the plurality of device groups function in the same manner as the monitoring data analysis device in each of the above-described embodiments. Good.
- Log data including monitoring data in the monitoring target system that is the target of performance management is classified into a plurality of groups, and regression models generated for each group are represented by the groups corresponding to the regression models. And a regression model recalculation unit that recalculates the regression model by combining the log data belonging to and the inspection target log data that is the log data to be subjected to performance inspection.
- Monitoring data analysis device Log data including monitoring data in the monitoring target system that is the target of performance management is classified into a plurality of groups, and regression models generated for each group are represented by the groups corresponding to the regression models.
- a regression model recalculation unit that recalculates the regression model by combining the log data belonging to and the inspection target log data that is the log data to be subjected to performance inspection. Monitoring data analysis device.
- the log data is classified into the groups based on the characteristics of the data storage unit that stores the log data and the usage data indicating the usage status of the components of the monitored system included in the log data.
- the monitoring data analysis apparatus further comprising: a data classification unit that performs a regression analysis of the log data for each group to generate the regression model.
- the said feature data extraction part calculates the dependence to the said explanatory variable in each data item of the data which forms the said group for every said group, and the said calculated said dependence is higher than a predetermined threshold value
- the monitoring data analysis apparatus according to appendix 3, wherein a data item is extracted as the feature data item.
- the said feature data extraction part calculates the correlation coefficient between the said explanatory variables in each data item of the data which forms the said group for every said group, and each said calculated correlation coefficient of each said correlation coefficient
- the monitoring data analysis apparatus according to supplementary note 3, wherein a data item having an average value larger than a predetermined threshold is extracted as the feature data item.
- the said regression model recalculation part calculates the ratio of the value of each said characteristic data item with respect to the said explanatory variable for every said group, and synthesize
- a regression model for each feature data item which is the regression model for each of the feature data items, and each of the calculated regression models for each feature data item is included in the inspection target log data. 6.
- the monitoring data analysis apparatus according to any one of appendices 3 to 5, wherein the regression model is recalculated by synthesizing according to the appearance rate of the value of.
- the performance value calculation part which calculates the performance value which is the objective variable of the said regression model using the said regression model recalculated by the said regression model recalculation part is further provided, The additional remark 1 characterized by the above-mentioned. 7.
- the monitoring data analyzer according to any one of items 6 to 6.
- the apparatus further comprises an abnormality determination unit that determines whether there is an abnormality in the monitoring target system based on a difference from a value of a data item corresponding to the objective variable of the regression model included in the target log data.
- the monitoring data analyzer according to any one of appendices 1 to 6.
- Log data including monitoring data in the monitoring target system that is the target of performance management is classified into a plurality of groups, and the regression models generated for each group are classified into the groups corresponding to the regression models.
- a regression model recalculation step of recalculating the regression model by synthesizing using the log data belonging to and the inspection target log data which is the log data to be performance-tested Monitoring data analysis method.
- Appendix 10 A monitoring data analysis program for causing a computer to execute the steps described in Appendix 9.
- the monitoring data analysis apparatus, the monitoring data analysis method, and the monitoring data analysis program according to the present invention improve the prediction accuracy of the performance value even when the performance value of the monitoring target system is greatly different for each usage pattern. Is suitable.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Hardware Design (AREA)
- General Physics & Mathematics (AREA)
- Quality & Reliability (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Debugging And Monitoring (AREA)
Abstract
Description
まず、第1実施形態における監視データ分析装置の構成について説明する。
D.A.ハーヴィル著、「統計のための行列代数 下 (2)」、20章ムーア_ベンローズ形逆行列、シュプリンガー・ジャパン(株)
本発明の第2実施形態について説明する。まず、第2実施形態における監視データ分析装置の構成について説明する。
長畑秀和著、「多変量解析へのステップ」、2章回帰分析、共立出版
回帰モデルM2:f2(λ)=ρ=0.0019λ+0.1188 … 式(8)
上述した各実施形態は、単なる例示に過ぎず、各実施形態に明示していない種々の変形や技術の適用を排除するものではない。すなわち、本発明は、その趣旨を逸脱しない範囲で様々な形態に変形して実施することができる。
Claims (10)
- 性能管理の対象となる監視対象システムにおける監視データを含むログデータが複数のグループに分類されて、当該グループごとに生成された回帰モデルを、それぞれの前記回帰モデルに対応する前記グループに属する前記ログデータ、および性能検査の対象となる前記ログデータである検査対象ログデータを用いて合成することで、前記回帰モデルを再計算する回帰モデル再計算部を、備えることを特徴とする監視データ分析装置。
- 前記ログデータを蓄積するデータ蓄積部と、
前記ログデータに含まれる前記監視対象システムの構成要素の利用状況を示す利用状況データの特性に基づいて、前記ログデータを前記グループに分類するデータ分類部と、
前記グループごとに前記ログデータの回帰分析を実行して前記回帰モデルを生成する回帰モデル生成部と、
さらに備えることを特徴とする請求項1記載の監視データ分析装置。 - 前記ログデータに含まれるデータ項目の中から、前記回帰モデルの説明変数と関連がある前記データ項目を特徴データ項目として抽出する特徴データ項目抽出部を、
さらに備えることを特徴とする請求項1記載の監視データ分析装置。 - 前記特徴データ抽出部は、前記グループごとに、当該グループを形成するデータの各データ項目における前記説明変数への依存度を算出し、当該算出した前記依存度が所定閾値よりも高いデータ項目を、前記特徴データ項目として抽出する、
ことを特徴とする請求項3記載の監視データ分析装置。 - 前記特徴データ抽出部は、前記グループごとに、当該グループを形成するデータの各データ項目における前記説明変数との間の相関係数を算出し、当該算出した各前記相関係数の平均値が所定閾値よりも大きいデータ項目を、前記特徴データ項目として抽出する、
ことを特徴とする請求項3記載の監視データ分析装置。 - 前記回帰モデル再計算部は、前記説明変数に対する各前記特徴データ項目の値の割合を前記グループごとに算出し、当該算出した前記割合を用いて各前記グループの前記回帰モデルを合成することで、各前記特徴データ項目に関する前記回帰モデルである特徴データ項目別回帰モデルを算出し、当該算出した各前記特徴データ項目別回帰モデルを、前記検査対象ログデータに含まれる前記特徴データ項目の値の出現率に応じて合成することで、前記回帰モデルを再計算する、
ことを特徴とする請求項3~5のいずれか1項に記載の監視データ分析装置。 - 前記回帰モデル再計算部により再計算された前記回帰モデルを用いて、当該回帰モデルの目的変数である性能値を算出する性能値算出部を、
さらに備えることを特徴とする請求項1~6のいずれか1項に記載の監視データ分析装置。 - 前記検査対象ログデータに含まれる前記回帰モデルの説明変数に対応するデータ項目の値を前記回帰モデル再計算部により再計算された回帰モデルに代入して得られる値と、前記検査対象ログデータに含まれる前記回帰モデルの目的変数に対応するデータ項目の値との差に基づいて、前記監視対象システムにおける異常の有無を判定する異常判定部を、
さらに備えることを特徴とする請求項1~6のいずれか1項に記載の監視データ分析装置。 - 性能管理の対象となる監視対象システムにおける監視データを含むログデータが複数のグループに分類されて、当該グループごとに生成された回帰モデルを、それぞれの前記回帰モデルに対応する前記グループに属する前記ログデータ、および性能検査の対象となる前記ログデータである検査対象ログデータを用いて合成することで、前記回帰モデルを再計算する回帰モデル再計算ステップを、含むことを特徴とする監視データ分析方法。
- 請求項9に記載のステップをコンピュータに実行させるための監視データ分析プログラム。
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US13/824,539 US9465713B2 (en) | 2010-12-24 | 2011-12-12 | Monitoring data analyzing apparatus, monitoring data analyzing method, and monitoring data analyzing program |
JP2012549724A JP5871193B2 (ja) | 2010-12-24 | 2011-12-12 | 監視データ分析装置、監視データ分析方法および監視データ分析プログラム |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2010288529 | 2010-12-24 | ||
JP2010-288529 | 2010-12-24 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2012086444A1 true WO2012086444A1 (ja) | 2012-06-28 |
Family
ID=46313723
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2011/078714 WO2012086444A1 (ja) | 2010-12-24 | 2011-12-12 | 監視データ分析装置、監視データ分析方法および監視データ分析プログラム |
Country Status (3)
Country | Link |
---|---|
US (1) | US9465713B2 (ja) |
JP (1) | JP5871193B2 (ja) |
WO (1) | WO2012086444A1 (ja) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2015146100A1 (ja) * | 2014-03-28 | 2015-10-01 | 日本電気株式会社 | 負荷推定システム、情報処理装置、負荷推定方法、及び、プログラムを記憶する記憶媒体 |
EP3093760A1 (en) | 2015-05-01 | 2016-11-16 | Fujitsu Limited | Device and method |
WO2019138560A1 (ja) * | 2018-01-12 | 2019-07-18 | 株式会社Jtb総合研究所 | 情報処理装置、情報処理プログラムおよび情報処理システム |
Families Citing this family (24)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102831214B (zh) | 2006-10-05 | 2017-05-10 | 斯普兰克公司 | 时间序列搜索引擎 |
JP5848597B2 (ja) * | 2011-12-21 | 2016-01-27 | インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Machines Corporation | Cmdbを利用したソフトウェア構成値の最適化方法、装置、システムおよびプログラム |
US9015716B2 (en) | 2013-04-30 | 2015-04-21 | Splunk Inc. | Proactive monitoring tree with node pinning for concurrent node comparisons |
US8904389B2 (en) | 2013-04-30 | 2014-12-02 | Splunk Inc. | Determining performance states of components in a virtual machine environment based on performance states of related subcomponents |
US10346357B2 (en) | 2013-04-30 | 2019-07-09 | Splunk Inc. | Processing of performance data and structure data from an information technology environment |
US10353957B2 (en) | 2013-04-30 | 2019-07-16 | Splunk Inc. | Processing of performance data and raw log data from an information technology environment |
US8972992B2 (en) | 2013-04-30 | 2015-03-03 | Splunk Inc. | Proactive monitoring tree with state distribution ring |
US10614132B2 (en) | 2013-04-30 | 2020-04-07 | Splunk Inc. | GUI-triggered processing of performance data and log data from an information technology environment |
US9142049B2 (en) | 2013-04-30 | 2015-09-22 | Splunk Inc. | Proactive monitoring tree providing distribution stream chart with branch overlay |
US9185007B2 (en) | 2013-04-30 | 2015-11-10 | Splunk Inc. | Proactive monitoring tree with severity state sorting |
US10997191B2 (en) | 2013-04-30 | 2021-05-04 | Splunk Inc. | Query-triggered processing of performance data and log data from an information technology environment |
US10019496B2 (en) | 2013-04-30 | 2018-07-10 | Splunk Inc. | Processing of performance data and log data from an information technology environment by using diverse data stores |
US9495187B2 (en) | 2013-04-30 | 2016-11-15 | Splunk, Inc. | Interactive, top-down presentation of the architecture and performance of a hypervisor environment |
US10318541B2 (en) * | 2013-04-30 | 2019-06-11 | Splunk Inc. | Correlating log data with performance measurements having a specified relationship to a threshold value |
US10225136B2 (en) | 2013-04-30 | 2019-03-05 | Splunk Inc. | Processing of log data and performance data obtained via an application programming interface (API) |
KR101768438B1 (ko) * | 2013-10-30 | 2017-08-16 | 삼성에스디에스 주식회사 | 데이터 분류 장치 및 방법과 이를 이용한 데이터 수집 시스템 |
WO2015145500A1 (ja) * | 2014-03-27 | 2015-10-01 | 日本電気株式会社 | システム分析装置、分析モデル生成方法、システム分析方法およびシステム分析プログラム |
US9921930B2 (en) * | 2015-03-04 | 2018-03-20 | International Business Machines Corporation | Using values of multiple metadata parameters for a target data record set population to generate a corresponding test data record set population |
US10361943B2 (en) * | 2017-02-06 | 2019-07-23 | Ca, Inc. | Methods providing performance management using a proxy baseline and related systems and computer program products |
DE112018007597B4 (de) | 2018-06-18 | 2022-06-09 | Mitsubishi Electric Corporation | Diagnosevorrichtung, Diagnoseverfahren und Programm |
WO2020173023A1 (en) * | 2019-02-27 | 2020-09-03 | Huawei Technologies Co., Ltd. | Artificial intelligent enhanced data sampling |
JP7207537B2 (ja) * | 2019-06-17 | 2023-01-18 | 日本電信電話株式会社 | 分類装置、分類方法及び分類プログラム |
CN112711739B (zh) * | 2019-10-25 | 2024-05-28 | 腾讯科技(深圳)有限公司 | 一种数据处理方法、装置及服务器、存储介质 |
CN110851324B (zh) * | 2019-10-25 | 2023-03-21 | 泰康保险集团股份有限公司 | 基于日志的巡检处理方法、装置以及电子设备、存储介质 |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0765168A (ja) * | 1993-08-31 | 1995-03-10 | Hitachi Ltd | 関数近似装置及び方法 |
JP2003263342A (ja) * | 2002-03-07 | 2003-09-19 | Telecommunication Advancement Organization Of Japan | 情報処理装置の監視装置および監視方法並びにそのプログラム |
JP2004086897A (ja) * | 2002-08-06 | 2004-03-18 | Fuji Electric Holdings Co Ltd | モデル構築方法およびモデル構築システム |
WO2006046297A1 (ja) * | 2004-10-28 | 2006-05-04 | Fujitsu Limited | 分析方法及び装置 |
US20090287744A1 (en) * | 2008-05-15 | 2009-11-19 | International Business Machines Corporation | Apparatus, system and method for healthcheck of information technology infrastructure based on log data |
Family Cites Families (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2002268922A (ja) | 2001-03-09 | 2002-09-20 | Ntt Data Corp | Wwwサイトの性能監視装置 |
US8014880B2 (en) * | 2006-09-29 | 2011-09-06 | Fisher-Rosemount Systems, Inc. | On-line multivariate analysis in a distributed process control system |
US8032341B2 (en) * | 2007-01-04 | 2011-10-04 | Fisher-Rosemount Systems, Inc. | Modeling a process using a composite model comprising a plurality of regression models |
US8903757B2 (en) * | 2008-12-12 | 2014-12-02 | Appnomic Systems Private Limited | Proactive information technology infrastructure management |
WO2010082322A1 (ja) * | 2009-01-14 | 2010-07-22 | 株式会社日立製作所 | 装置異常監視方法及びシステム |
JP5297272B2 (ja) * | 2009-06-11 | 2013-09-25 | 株式会社日立製作所 | 装置異常監視方法及びシステム |
JP5431235B2 (ja) | 2009-08-28 | 2014-03-05 | 株式会社日立製作所 | 設備状態監視方法およびその装置 |
-
2011
- 2011-12-12 US US13/824,539 patent/US9465713B2/en active Active
- 2011-12-12 JP JP2012549724A patent/JP5871193B2/ja active Active
- 2011-12-12 WO PCT/JP2011/078714 patent/WO2012086444A1/ja active Application Filing
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0765168A (ja) * | 1993-08-31 | 1995-03-10 | Hitachi Ltd | 関数近似装置及び方法 |
JP2003263342A (ja) * | 2002-03-07 | 2003-09-19 | Telecommunication Advancement Organization Of Japan | 情報処理装置の監視装置および監視方法並びにそのプログラム |
JP2004086897A (ja) * | 2002-08-06 | 2004-03-18 | Fuji Electric Holdings Co Ltd | モデル構築方法およびモデル構築システム |
WO2006046297A1 (ja) * | 2004-10-28 | 2006-05-04 | Fujitsu Limited | 分析方法及び装置 |
US20090287744A1 (en) * | 2008-05-15 | 2009-11-19 | International Business Machines Corporation | Apparatus, system and method for healthcheck of information technology infrastructure based on log data |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2015146100A1 (ja) * | 2014-03-28 | 2015-10-01 | 日本電気株式会社 | 負荷推定システム、情報処理装置、負荷推定方法、及び、プログラムを記憶する記憶媒体 |
JPWO2015146100A1 (ja) * | 2014-03-28 | 2017-04-13 | 日本電気株式会社 | 負荷推定システム、情報処理装置、負荷推定方法、及び、コンピュータ・プログラム |
US10248462B2 (en) | 2014-03-28 | 2019-04-02 | Nec Corporation | Management server which constructs a request load model for an object system, load estimation method thereof and storage medium for storing program |
EP3093760A1 (en) | 2015-05-01 | 2016-11-16 | Fujitsu Limited | Device and method |
WO2019138560A1 (ja) * | 2018-01-12 | 2019-07-18 | 株式会社Jtb総合研究所 | 情報処理装置、情報処理プログラムおよび情報処理システム |
JPWO2019138560A1 (ja) * | 2018-01-12 | 2020-12-17 | 株式会社Jtb総合研究所 | 情報処理装置、情報処理プログラムおよび情報処理システム |
Also Published As
Publication number | Publication date |
---|---|
JP5871193B2 (ja) | 2016-03-01 |
JPWO2012086444A1 (ja) | 2014-05-22 |
US9465713B2 (en) | 2016-10-11 |
US20130191107A1 (en) | 2013-07-25 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP5871193B2 (ja) | 監視データ分析装置、監視データ分析方法および監視データ分析プログラム | |
JP5871192B2 (ja) | 監視データ分析装置、監視データ分析方法および監視データ分析プログラム | |
JP5418408B2 (ja) | シミュレーションパラメータ校正方法、装置及びプログラム | |
CN107003928B (zh) | 性能异常诊断 | |
CN104021264A (zh) | 一种缺陷预测方法及装置 | |
JP2008165352A (ja) | 定期点検データ分析装置およびその方法 | |
JP6079243B2 (ja) | 障害分析支援装置、障害分析支援方法、及びプログラム | |
US20200143904A1 (en) | Predicting molecular properties of molecular variants using residue-specific molecular structural features | |
US20140350900A1 (en) | Simulation system and method thereof and computing system including the simulation system | |
JP5758335B2 (ja) | システム疲労度評価装置およびその方法 | |
WO2015146100A1 (ja) | 負荷推定システム、情報処理装置、負荷推定方法、及び、プログラムを記憶する記憶媒体 | |
JP6697082B2 (ja) | 需要予測方法、需要予測システム及びそのプログラム | |
JP5600693B2 (ja) | クラスタリング装置及び方法及びプログラム | |
JP5668425B2 (ja) | 障害検知装置、情報処理方法、およびプログラム | |
JP5135803B2 (ja) | 最適パラメータ探索プログラム、最適パラメータ探索装置および最適パラメータ探索方法 | |
US9922109B1 (en) | Adaptive column set composition | |
JP5600694B2 (ja) | クラスタリング装置及び方法及びプログラム | |
CN114974462A (zh) | 缓蚀效率预测模型的训练方法、装置、设备及存储介质 | |
JP2009157655A (ja) | データ解析装置、データ解析方法、およびプログラム | |
WO2019060314A1 (en) | APPARATUS AND METHOD FOR INTRODUCING PROBABILITY AND UNCERTAINTY IN CLASSIFICATION OF DATA NOT SUPERVISED BY GROUPING, THROUGH CLASSIFICATION STATISTICS | |
CN116541252B (zh) | 一种机房故障日志数据处理方法及装置 | |
CN116738216B (zh) | 预警大数据的关联处理方法和装置 | |
Li et al. | An Improved SSI Approach for Structural Modal Identification | |
US20240143666A1 (en) | Smart metric clustering | |
JP5582017B2 (ja) | 相図生成プログラム,相図生成システムおよび相図生成方法 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 11851079 Country of ref document: EP Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 13824539 Country of ref document: US |
|
ENP | Entry into the national phase |
Ref document number: 2012549724 Country of ref document: JP Kind code of ref document: A |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 11851079 Country of ref document: EP Kind code of ref document: A1 |