WO2013128789A1 - キャパシティ管理支援装置、キャパシティ管理方法およびプログラム - Google Patents
キャパシティ管理支援装置、キャパシティ管理方法およびプログラム Download PDFInfo
- Publication number
- WO2013128789A1 WO2013128789A1 PCT/JP2013/000188 JP2013000188W WO2013128789A1 WO 2013128789 A1 WO2013128789 A1 WO 2013128789A1 JP 2013000188 W JP2013000188 W JP 2013000188W WO 2013128789 A1 WO2013128789 A1 WO 2013128789A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- log
- definition
- data
- resource
- load
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5083—Techniques for rebalancing the load in a distributed system
-
- 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
- G06F11/3442—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 planning or managing the needed capacity
-
- 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
- G06F11/3476—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
- G06F2201/875—Monitoring of systems including the internet
Definitions
- the present invention relates to an apparatus for managing system capacity, a system capacity management method, and a program therefor.
- cloud computing such as IaaS (Infrastructure as a Service) and SaaS (Software as a Service) have begun to spread widely.
- IaaS Infrastructure as a Service
- SaaS Software as a Service
- the system provider needs to ensure that the performance required by the user can be achieved in the system after the change. There is. Therefore, the system provider is required to manage the capacity to estimate whether the system has sufficient processing capacity for the expected load. For example, how much specifications should the CPU and memory have for the expected load, or how much the current specs are insufficient to handle the assumed load, etc. Such information must be known by the system provider.
- Patent Document 1 discloses an example of a capacity prediction system.
- resource usage and transaction logs are acquired from a computer, and a resource usage rate for each transaction is calculated using multiple regression analysis.
- the future processing amount is predicted for each transaction based on the transaction log.
- a transition of the computer resource usage rate is predicted.
- load logs and resource usage logs recorded in the past in the monitored system Is used.
- the amount of resources that can handle the assumed load can be obtained by deriving the relationship between the load and the resource usage based on these logs.
- the distribution of the measured log is not necessarily due to the lack of the measured log or the error based on the characteristics of the middleware that measures the log. May not match true distribution based on relationship between load and resource usage.
- Patent Document 1 uses a log acquired from a computer as it is and derives a relationship between a transaction and a resource usage. For this reason, log errors and deficiencies can cause errors in deriving the relationship between transactions and resource usage.
- An object of the present invention is to provide a capacity management support apparatus, a capacity management method, and a capacity management program that calculate a predicted value with high accuracy when predicting the relationship between load and resource usage.
- a type definition that associates a log related to a resource and a log related to a load corresponding to the log related to the resource, and a load definition that defines an assumed load value that is a value of the load assumed for the monitored system
- Storage means for storing;
- Input means for acquiring input information for specifying an association between the log related to the resource and the log related to the load from the type definition;
- a log type to be acquired is determined based on the input information and the type definition, and first log data obtained by extracting data related to the determined log type from a log held by the monitored system Log acquisition means to acquire;
- second log data which is data obtained by extracting a correspondence relationship between the specific resource and the specific load, is obtained from the first log data, and based on the second log data, resource use is acquired.
- a distribution density function indicating a true distribution of quantity and load value, a range satisfying a specific condition is selected from the distribution density function, and third log data that is data belonging to the range among the second log data Log distribution estimation means for obtaining
- a prediction formula related to a resource usage rate is calculated based on data of a certain threshold value or more in the third log data, and a predicted value of the resource usage rate is calculated based on the prediction formula and the load definition.
- a capacity management support apparatus having a resource usage rate predicting means is provided.
- Computer Defines a type definition that associates a log related to a resource and a log related to a load corresponding to the log related to the resource, and an assumed load value that is a value of the load assumed for the monitored system from the storage unit
- Load definition From the type definition, obtain input information for specifying the association between the log related to the resource and the log related to the load, A log type to be acquired is determined based on the input information and the type definition, and first log data obtained by extracting data related to the determined log type from a log held by the monitored system Acquired, Based on the type definition, second log data, which is data obtained by extracting a correspondence relationship between the specific resource and the specific load, is obtained from the first log data, and based on the second log data, resource use is acquired.
- a distribution density function indicating a true distribution of quantity and load value, a range satisfying a specific condition is selected from the distribution density function, and third log data that is data belonging to the range among the second log data Get
- a prediction formula related to a resource usage rate is calculated based on data of a certain threshold value or more in the third log data, and a predicted value of the resource usage rate is calculated based on the prediction formula and the load definition.
- a capacity management method is provided.
- a log type to be acquired is determined based on the input information and the type definition, and first log data obtained by extracting data related to the determined log type from a log held by the monitored system Means to obtain,
- second log data which is data obtained by extracting a correspondence relationship between the specific resource and the specific load, is obtained from the first log data, and based on the second log data, resource use is acquired.
- a distribution density function indicating a true distribution of quantity and load value, a range satisfying a specific condition is selected from the distribution density function, and third log data that is data belonging to the range among the second log data Means to obtain the A prediction formula related to a resource usage rate is calculated based on data of a certain threshold value or more in the third log data, and a predicted value of the resource usage rate is calculated based on the prediction formula and the load definition.
- a program for functioning as a means is provided.
- the relationship between the load and the resource usage can be predicted with high accuracy.
- FIG. 1 It is a figure which shows the example of a service level definition. It is a figure which shows the flow of a process of the capacity management assistance apparatus which concerns on the 5th Embodiment of this invention. It is a block diagram which shows the structure of the capacity management assistance apparatus which concerns on the 6th Embodiment of this invention. It is a figure which shows the example of a structure definition. It is a flowchart which shows the flow of the process which concerns on the 6th Embodiment of this invention.
- FIG. 1 is a block diagram showing a configuration of a capacity management support apparatus 10 according to the first embodiment of the present invention.
- the capacity management support apparatus 10 includes a storage unit 102, an input unit 104, a log acquisition unit 106, a log distribution estimation unit 108, and a resource usage rate prediction unit 110.
- the storage unit 102 stores a type definition 112 and a load definition 114.
- the type definition 112 defines the correspondence between the log related to the resource and the log related to the load acquired by the capacity management support apparatus 10 from the monitoring target system.
- FIG. 2 is a diagram illustrating an example of the type definition 112.
- the row with the group ID “1” indicates the log “resource_CPU_Usage” recorded by the infrastructure “WEB001” and the load “Web Request” recorded by the infrastructure “LB001” among the logs acquired from the monitoring target system. This corresponds to the corresponding log.
- the type definition 112 may determine not only the correspondence between resources and loads, but also the correspondence between resources.
- the type definition 112 is preset in the storage unit 102, for example. Further, the capacity management support apparatus 10 monitors what log is recorded in which infrastructure according to the processing executed in the monitored system, and the type definition 112 is based on the monitoring result. It may be set dynamically.
- the load definition 114 defines a load value assumed for the monitoring target system (hereinafter referred to as an assumed load value) according to a certain load type.
- FIG. 3 is a diagram illustrating an example of the load definition 114.
- the assumed load value may be, for example, a value determined in advance, or for each load type obtained by calculating the load value statistics for a certain period in the monitored system for each load type. May be the maximum value or average value.
- the input unit 104 acquires input information from another device located outside the capacity management support apparatus 10, a storage area of the capacity management support apparatus 10, or the like.
- the log acquisition unit 106 determines the type of log to be extracted based on the input information and the type definition 112. Then, the log acquisition unit 106 extracts information on the determined log type from the logs of the monitoring target system, and creates first log data.
- the log distribution estimation unit 108 extracts information on a specific resource and load from the first log data based on a set of log types determined by the type definition 112, and creates second log data. Next, the log distribution estimation unit 108 estimates a distribution density function indicating the true distribution of the second log data based on the second log data. Then, the log distribution estimation unit 108 selects a range that satisfies a specific condition in the distribution density function, extracts second log data existing in the range, and creates third log data.
- the resource usage rate prediction unit 110 calculates a prediction formula indicating a set relationship defined by the type definition 112 based on the third log data that is equal to or greater than a certain threshold. Then, the resource usage rate prediction unit 110 substitutes the assumed load value determined by the load definition 114 into the prediction formula, and calculates the predicted value of the resource usage rate.
- each component of the capacity management support apparatus 10 shown in each figure is not a hardware unit configuration but a functional unit block.
- Each component of the capacity management support apparatus 10 includes a CPU of any computer, a memory, a program for realizing the components shown in the figure loaded in the memory, a storage medium such as a hard disk for storing the program, and a network connection interface. It is realized by any combination of hardware and software. There are various modifications of the implementation method and apparatus.
- FIG. 4 is a flowchart showing a processing flow of the capacity management support apparatus 10 according to the first embodiment of the present invention.
- the input unit 104 acquires information about a resource whose usage rate is predicted (S102). Information about the resource is acquired, for example, when a user inputs using GUI (Graphical User Interface) or CUI (Character User Interface). Further, information on resources may be input from other software through API (Application Programming Interface). Further, information regarding resources may be acquired by reading a file (not shown) in which necessary information is recorded.
- the input unit 104 transmits the acquired information to the log acquisition unit 106.
- the log acquisition unit 106 extracts information based on the information received from the input unit 104 and the type definition 112 from the log recording the resource usage and load of the monitoring target system as shown in FIG. First log data is created (S104). For example, when the resources “CPU_Usage” and “MEM_Usage” and the infrastructure ID “WEB001” are received as input information from the input unit 104, the log acquisition unit 106 based on the input information and the type definition 112 as shown in FIG. First log data is created. Specifically, in the type definition shown in FIG. 2, the information corresponding to “CPU_Usage” and “WEB001” is “Web Request” and “LB001”. In the type definition shown in FIG.
- the log acquisition unit 106 extracts “Web Request” and “LB001” logs using “CPU_Usage” and “WEB001”, and extracts “Throughput” and “LB001” logs using “MEM_Usage” and “WEB001”. .
- the log acquisition unit 106 also extracts the input information “CPU_Usage” and “WEB001” logs and the “MEM_Usage” and “WEB001” logs.
- the log acquisition unit 106 checks the time when the extraction target log is recorded, and assigns the same log ID to the logs recorded at the same time. The log acquisition unit 106 transmits the created first log data to the log distribution estimation unit 108.
- the log distribution estimation unit 108 extracts information specified by the group defined by the type definition 112 from the first log data received from the log acquisition unit 106.
- the log distribution estimation unit 108 can determine from the type definition shown in FIG. 2 that, for example, the “CPU_Usage” log recorded with “WEB001” corresponds to the “Web Request” log recorded with “LB001”. .
- the log distribution estimation unit 108 extracts information from the first log data by using the correspondence relationship of the log type, and creates second log data as shown in FIG. 7, for example (S106).
- the log distribution estimation unit 108 estimates a distribution density function based on the second log data (S108).
- FIG. 8 is a diagram illustrating an example of a distribution density function estimated by the log distribution estimation unit 108.
- the horizontal axis represents the usage amount of the resource “CPU_Usage”
- the vertical axis represents the value of the load “Web Request”.
- the upper left cell has a probability that the load “Web Request” value is “0 to 1145” when the usage amount of the resource “CPU_Usage” is “0 to 2.0”. 2.61E-09 ".
- the probability that the usage amount of “CPU_Usage” is “0 to 2.0” is “2.61E-09”.
- the log distribution estimation unit 108 can estimate the distribution density function using a non-parametric method represented by, for example, a kernel density estimation method.
- the log distribution estimation unit 108 selects a range considered to be highly reliable from the distribution density function estimated in S108. For example, the log distribution estimation unit 108 selects a 99% confidence interval, a 95% confidence interval, an upper XX%, etc. as a high reliability range from the distribution density function estimated in S108.
- the log distribution estimation unit 108 extracts second log data existing in the estimated range, and creates third log data (S110). The log distribution estimation unit 108 transmits the created third log data to the resource usage rate prediction unit 110.
- the resource usage rate prediction unit 110 calculates a prediction formula related to the resource usage rate from the third log data received from the log distribution estimation unit 108 (S112). First, the resource usage rate prediction unit 110 selects data that is equal to or greater than a certain threshold among the third log data as data used for derivation of the prediction formula. For example, in addition to the value of the third log data itself, the resource usage rate prediction unit 110 is based on the distance between the average value and median value of the third log data, and each data included in the third log data, and its variance. , Select data with a certain value or more.
- the resource usage rate prediction unit 110 selects data having a certain value or more based on the inclination and distance of a straight line connecting the origin and each data when the resource usage amount and the load value are represented by a two-dimensional coordinate axis. May be. For example, the resource usage rate prediction unit 110 selects data corresponding to the upper 50% of “CPU_Usage” from the third log data. Next, the resource usage rate prediction unit 110 derives an equation representing the relationship between the resource usage amount and the load, as shown in FIG. 9, using data in a range selected based on the threshold value.
- the resource usage rate prediction unit 110 derives an approximate function from the selected data using, for example, the least square method, polynomial approximation, and fitting to a plurality of equations that are separately defined, and the approximate function is used as a prediction formula. .
- the resource usage rate prediction unit 110 converts the log type into a corresponding resource type. For example, when the log type is “CPU_Usage”, the resource usage rate prediction unit 110 converts the log type into “CPU” that is a corresponding resource type.
- the resource usage rate prediction unit 110 stores, for example, information defining the correspondence between the log type and the resource type in the storage unit 102, and converts the log type and the resource type based on the definition. can do.
- the resource usage rate prediction unit 110 may use a linear function such as a linear function, a higher-order polynomial of second or higher order, a logarithmic function, a power function, or an exponential function as an approximation function.
- the approximate function to be used as the prediction formula may be stored based on the definition of the approximate function used for each log type in the storage unit 102 and determined based on the definition.
- the determination coefficient shown may be calculated for each approximate function and selected based on the threshold value.
- Equation 1 is an equation for obtaining a coefficient of determination indicating the degree to which the approximate function matches the selected data.
- R is a coefficient of determination
- yi is a data value
- fi is a solution of an approximation function
- ya is an average value of data.
- data corresponding to yi is data indicating a resource usage amount such as “CPU_Usage (WEB001)”.
- the approximate function solution fi is obtained by substituting data indicating a load such as “Web Request (LB001)” into the approximate function among the data input to the resource usage rate prediction unit 110.
- the resource usage rate prediction unit 110 selects an approximation function having the largest determination coefficient as a prediction formula.
- the resource usage rate prediction unit 110 calculates a predicted value of the resource usage rate as shown in FIG. 10 based on the prediction formula calculated in S112 and the assumed load value of the load definition 114 (S114). For example, in FIG. 9, attention is paid to the relational expression between the load type “Web Request” and the resource type “CPU” in the infrastructure ID “WEB001”.
- the resource usage rate prediction unit 110 substitutes the assumed load value “300,000” of “Web Request” into the prediction formula shown in FIG. ”Is calculated as“ 67 ”. This indicates that when the assumed load value “Web Request” exists in the monitoring target system, the usage rate of the resource “CPU” of the infrastructure “WEB001” is 67%.
- the capacity management support apparatus 10 can also present the prediction formula and the prediction value calculated by the resource usage rate prediction unit 110 to the user using a display unit (not shown). For example, the capacity management support apparatus 10 may display a prediction formula and a predicted value on a display. Further, the capacity management support apparatus 10 may print a form on which the prediction formula and the predicted value are printed using a printer or the like.
- the true distribution of the logs of the monitoring target system is obtained. Then, the resource usage is predicted based on the true distribution. Thereby, it is possible to correct an error or a deficiency in the actual measurement value of the log in the monitoring target system. Therefore, according to this configuration, the resource usage prediction accuracy is improved as compared with the method of using the actually measured log as it is.
- FIG. 11 is a block diagram showing the configuration of the capacity management support apparatus 10 according to the second embodiment of the present invention.
- the capacity management support apparatus 10 further includes a log classification unit 202.
- the storage unit 102 further stores a classification definition 204.
- the classification definition 204 defines a method for classifying data included in the third log data.
- FIG. 12 is a diagram illustrating an example of the classification definition 204.
- a classification method set in the classification definition 204 for example, a Ward method, a K-average method, a shortest distance method, a longest distance method, a group average method, or the like can be used.
- the index is a threshold for classifying data included in the third log data.
- the calculation formula is a formula for calculating the index.
- the condition indicates a threshold related to the index, the number of data to be classified, and the like.
- the log classification unit 202 receives the third log data output from the log distribution estimation unit 108 as an input, and classifies the data included in the third log data into a plurality of sets based on the classification definition 204.
- the log classification unit 202 receives the third log data from the log distribution estimation unit 108. Then, based on the classification method defined in the classification definition 204, the data included in the received third log data is classified into a plurality of fourth log data (S202).
- the fourth log data is obtained by clustering data similar to each other in the relationship between the resource usage and the load among the data included in the third log data.
- the index for classifying the log includes, for example, the distance between the origin and each data included in the third log data, and each data of the origin and the third log data when the resource usage and the load are expressed by a two-dimensional coordinate axis.
- the method by which the log classification unit 202 classifies the third log data may be determined in advance for each log type, for example.
- the log classification unit 202 transmits all the classified fourth log data to the resource usage rate prediction unit 110.
- the resource usage rate prediction unit 110 uses, for example, the fourth log data having the maximum median value among the plurality of fourth log data classified by the log classification unit 202 in order to derive a prediction formula. Since the subsequent processing is the same as that of the first embodiment except that the fourth log data is used instead of the third log data, the description thereof is omitted.
- each data included in the third log data is classified into a plurality of fourth log data based on the classification definition 204 by the log classification unit 202.
- FIG. 14 is a block diagram showing the configuration of the capacity management support apparatus 10 according to the third embodiment of the present invention.
- the storage unit 102 further stores a correlation definition 302.
- the correlation definition 302 defines a log type (hereinafter referred to as a secondary log type) having a correlation with a log type (hereinafter referred to as a main log type) acquired by the input unit 104.
- the correlation definition 302 defines a pattern to which the fourth log data belongs based on the resource usage of the main log type and the resource usage of the secondary log type.
- FIG. 15 is a diagram illustrating an example of the correlation definition 302. For example, in FIG. 15, when “CPU_Usage” of the infrastructure “WEB001” is the main log type, it indicates that “CPU_Usage” of the infrastructure “DB001” is a sub log type indicating a correlation.
- the index is used for patterning the fourth log data classified by the log classification unit 202 according to the processing tendency.
- the index is calculated by a mathematical formula set for each correspondence between the main log type and the sub log type. By comparing the calculated index with the condition set as the threshold, it is possible to determine what tendency each fourth log data classified by the log classification unit 202 shows.
- the log acquisition unit 106 acquires a log of the sub log type corresponding to the main log type acquired by the input unit 104, and assigns it to the first log data acquired in the first embodiment. (S302).
- logs of “Web Request” and “LB001” are extracted from “CPU_Usage” and “WEB001”, and logs of “Throughput” and “LB001” are extracted from “MEM_Usage” and “WEB001”.
- the logs of “CPU_Usage” and “WEB001” which are input information, and the logs of “MEM_Usage” and “WEB001” are also extracted.
- the log acquisition unit 106 sets the logs of “CPU_Usage” and “DB001” as sub log type logs corresponding to the main log types “CPU_Usage” and “WEB001”. Extract further. Further, based on the correlation definition 302, the log acquisition unit 106 further extracts “MEM_Usage” and “AP001” logs as sub log type logs corresponding to the main log types “MEM_Usage” and “WEB001”. .
- the log distribution estimation unit 108 extracts secondary log type data from the first log data together with the main log type and load data, and creates second log data (S304).
- second log data S304.
- DB001 column of “CPU_Usage (DB001)” is further extracted.
- the log distribution estimation unit 108 does not use the sub-log type information, and as in the first and second embodiments, the main log type. And processing based on load information. That is, the third log data transmitted to the log classification unit 202 is obtained by adding a sub log type column to the third log data transmitted in the first and second embodiments.
- the log classification unit 202 applies the classification method of the classification definition 204 to the data related to the primary log type and the data related to the secondary log type included in the third log data. Then, clustering is performed based on the result, and the third log data is classified into the fourth log data (S304).
- the log classification unit 202 determines the processing tendency indicated by the fourth log data classified in S304 based on the index and threshold defined in the correlation definition 302, and uses the pattern information indicating the tendency as the fourth log data. (S306).
- the log classification unit 202 uses the correlation definition shown in FIG. In this case, when the sum of values obtained by dividing “CPU_Usage” of “DB001” by “CPU_Usage” of “WEB001” is larger than 90, the log classification unit 202 determines that the pattern A is used.
- the log classification unit 202 determines that the sum is the value obtained by dividing “CPU_Usage” of “DB001” by “CPU_Usage” of “WEB001” and is 90 or less if it is greater than 70. Further, the log classification unit 202 determines that the sum is a pattern C when the sum of values obtained by dividing “CPU_Usage” of “DB001” by “CPU_Usage” of “WEB001” is 70 or less. Then, the log classification unit 202 assigns the determined pattern information to each classified fourth log data, and transmits it to the resource usage rate prediction unit 110.
- the resource usage rate prediction unit 110 calculates the prediction formula and the predicted value of the resource usage rate for the fourth log data with the largest pattern among the classified fourth log data as in the second embodiment. . For example, when classification is performed using the Ward method shown in the classification definition 204, if the pattern A is classified into three, the pattern B is one, and the pattern C is classified into one, the resource usage rate predicting unit 110 determines the three patterns A. The same processing as in the second embodiment is applied to the set. Note that the resource usage rate prediction unit 110 may calculate a resource usage rate prediction formula and a prediction value for each pattern information with reference to the pattern information. For example, the resource usage rate prediction unit 110 can calculate a prediction formula and a prediction value for each of the patterns A, B, and C with respect to the assumed load value defined in the load definition 114.
- pattern information indicating a tendency of processing performed by the monitoring target system is assigned using the correlation definition 302. Then, a prediction formula and a predicted value of the resource usage rate are calculated from the log classified for each pattern information. Thereby, the resource usage can be predicted according to the processing pattern executed by the monitoring target system, that is, according to the characteristics of the processing performed by the monitoring target system.
- FIG. 18 is a block diagram showing the configuration of the capacity management support apparatus 10 according to the fourth embodiment of the present invention.
- the storage unit 102 further includes a safety factor definition 402.
- the safety factor definition 402 defines a safety factor for each resource type or for the entire resource type.
- the safety factor is a coefficient for correcting a predicted value having an influence such as an error.
- FIG. 19 is a diagram illustrating an example of the data of the safety factor definition 402.
- the safety factor definition 402 includes at least a resource type and a safety factor.
- the resource type indicates the type of resource to which the safety factor is applied.
- the safety coefficient indicates a value used for correcting the predicted value.
- the resource usage rate prediction unit 110 reads the safety factor definition 402 and acquires a safety factor corresponding to the calculated resource type (S402).
- the same effects as those of the first embodiment can be obtained.
- the predicted value calculated by the resource usage rate prediction unit 110 is corrected using the safety factor definition 402.
- the resource usage rate prediction unit 110 can predict the resource usage with respect to the assumed load value with a margin, so that the capacity of the monitoring target system can be compared with the case where the safety factor definition 402 is not used. Can be detected early. Therefore, the monitored system can be operated more stably. Note that this embodiment may be applied to the second and third embodiments.
- FIG. 21 is a block diagram showing the configuration of the capacity management support apparatus 10 according to the fifth embodiment of the present invention.
- the capacity management support apparatus 10 further includes a service level determination unit 502, and the storage unit 102 further stores a service level definition 504.
- the storage unit 102 stores a service level definition 504 indicating a performance value required for each load type.
- FIG. 22 is a diagram illustrating an example of the service level definition 504.
- the service level definition 504 includes, for example, a load type and a request value defined for each load type.
- the required value is defined based on a target value of performance required for the system to be constructed.
- the service level determination unit 502 determines the current configuration of the monitoring target system based on the resource usage rate prediction formula predicted by the resource usage rate prediction unit 110 and the requested value of the service level definition 504. It is determined whether or not the above is satisfied.
- the service level determination unit 502 acquires a resource usage rate prediction formula from the resource usage rate prediction unit 110 (S502).
- the service level determination unit 502 calculates the resource amount necessary to achieve the service level based on the resource usage rate prediction formula acquired in S502 and the request value of the service level definition 504 (S504). . Then, the service level determination unit 502 determines whether the monitored system satisfies the service level with the current configuration based on the resource amount calculated in S504. As a result of the determination, if the current system configuration satisfies the required value (YES in S506), the service level determination unit 502 ends the process assuming that there is no problem with the current system configuration.
- the service level determination unit 502 reads the request value “200,000” of “Throughput” from the service level definition 504. Then, the service level determination unit 502 substitutes the read request value into the assumed load value portion of the prediction formula. In this example, the substituted result is “67” and does not exceed “100”. In this case, the service level determination unit 502 can determine that the monitoring target system satisfies the service level.
- the service level determination unit 502 uses a display unit (not shown) to indicate that the current system configuration does not satisfy the request. To the user (S508). For example, the service level determination unit 502 can determine that the current system configuration does not satisfy the service level when the request value is substituted into the prediction formula acquired in S502 and the value exceeds “100”.
- the service level definition 504 may define a required value for resource usage. By substituting the required value of resource usage into the prediction formula, the service level determination unit 502 can calculate the maximum load that can maintain the service level in the current system configuration. In S502, the service level determination unit 502 can also acquire a prediction value calculated by the resource usage rate prediction unit 110 and determine a violation of the service level. For example, the service level determination unit 502 determines that the service level is not satisfied when the predicted value acquired in S502 is larger than the resource usage amount request value of the service level definition 504. On the other hand, the service level determination unit 502 can determine that the service level is satisfied when the predicted value acquired in S502 is equal to or less than the resource usage amount request value of the service level definition 504.
- the resource amount necessary for maintaining the service level at which the monitoring target system is determined is calculated from the predicted value predicted by the resource usage rate prediction unit 110. Further, it is determined whether or not the resource usage predicted for the assumed load value satisfies a predetermined service level. Therefore, with this configuration, the user can easily determine when to reinforce the configuration of the monitoring target system.
- FIG. 24 is a block diagram showing the configuration of the capacity management support apparatus 10 according to the sixth embodiment of the present invention.
- the capacity management support apparatus 10 further includes a configuration determination unit 602.
- the storage unit 102 further stores the configuration definition 604.
- the storage unit 102 stores a configuration definition 604 that is information related to the configuration.
- FIG. 25 is a diagram illustrating an example of the configuration definition 604.
- the configuration definition 604 includes, for example, a resource type, an infrastructure ID, an applied value, an added value, and the like.
- the data of the configuration definition 604 may include a maximum value or a minimum value that each resource type can take.
- the resource type indicates the type of resource that constitutes a system such as a CPU or a memory.
- the infrastructure ID indicates the name of a node including each resource.
- the applied value indicates the performance value of each resource currently installed in the monitored system.
- the added value indicates a resource unit to be added when reinforcing each resource.
- the configuration determination unit 602 determines whether or not to change the system configuration based on the prediction formula and predicted value of the resource usage rate predicted by the resource usage rate prediction unit 110 and the application value of the configuration definition 604.
- the configuration determination unit 602 acquires a resource usage rate prediction formula and a predicted value from the resource usage rate prediction unit 110 (S602).
- the configuration determination unit 602 compares the predicted value of the resource usage rate acquired in S602 with the application value of the configuration definition 604. As a result of the comparison, if the predicted value of the resource usage exceeds the application value (YES in S604), the configuration determination unit 602 determines that the performance of the resource currently installed in the system is insufficient. For example, when the predicted value of the usage rate of the resource “CPU” of the node “WEB001” exceeds “100”, the configuration determination unit 602 has the resource “CPU” of the node “WEB001” currently installed in the system. It can be determined that the performance is insufficient.
- the configuration determination unit 602 adds the addition value by 2 units (2.0 GHz). It can be judged that it is good. Then, the configuration determining unit 602 outputs the calculated result to a display unit (not shown).
- the display unit outputs a message such as “processable” or “ ⁇ ”.
- the display unit outputs information indicating the amount of resources to be added in addition to a message such as “cannot be processed” or “ ⁇ ”.
- the display unit presents the received information to the user (S608).
- the display unit displays the received information on a display, for example. Further, the display unit may present the received information to the user by outputting it to a form using a printer or the like.
- the current system configuration is predicted based on the current system performance value determined from the configuration definition 604 and the prediction formula and the predicted value calculated by the resource usage rate prediction unit 110. It is determined whether or not it can withstand the load. Thereby, it can be shown to a user whether the performance of the present system is enough. Further, when the system performance is insufficient with respect to the load on the system, it is possible to present how much resources should be added to reinforce the performance.
- a type definition that associates a log related to a resource and a log related to a load corresponding to the log related to the resource, and a load definition that defines an assumed load value that is a value of the load assumed for the monitored system
- Storage means for storing
- Input means for acquiring input information for specifying an association between the log related to the resource and the log related to the load from the type definition
- a log type to be acquired is determined based on the input information and the type definition, and first log data obtained by extracting data related to the determined log type from a log held by the monitored system Log acquisition means to acquire;
- Second log data which is data obtained by extracting a correspondence relationship between the specific resource and the specific load, is obtained from the first log data, and based on the second log data, resource use is acquired.
- a distribution density function indicating a true distribution of quantity and load value, a range satisfying a specific condition is selected from the distribution density function, and third log data that is data belonging to the range among the second log data Log distribution estimation means for obtaining
- a prediction formula related to a resource usage rate is calculated based on data of a certain threshold value or more in the third log data, and a predicted value of the resource usage rate is calculated based on the prediction formula and the load definition.
- a capacity management support apparatus comprising: a resource usage rate prediction unit.
- the storage means Further storing a classification definition defining a method and conditions for classifying the data contained in the third log data;
- the capacity management support device is Log classification means for classifying data included in the third log data on the basis of the classification definition to obtain a plurality of fourth log data;
- the resource usage rate predicting means includes: A capacity management support apparatus that calculates a prediction formula for the resource usage based on the fourth log data.
- the storage means The primary log type that is a log type determined based on the input information and the load definition is associated with the secondary log type that is a log type correlated with the primary log type, and the primary log type resource Further storing a correlation definition that defines a pattern of the fourth log data based on a usage amount and a resource usage amount of the sub-log type;
- the log acquisition means includes Based on the correlation definition, the secondary log type information is further added to the first log data;
- the log distribution estimating means includes Based on the resource data and the load data related to the main log type among the second log data, the distribution density function is estimated,
- the log classification means includes A capacity management support apparatus that further determines to which pattern the plurality of fourth log data belongs based on the correlation definition.
- the storage means Further storing a safety factor definition including a safety factor corresponding to the resource type;
- the resource usage rate predicting means includes: A capacity management support apparatus that corrects the prediction formula and the predicted value of the resource usage rate based on the safety factor.
- the storage means Further storing a service level definition including a request value that is a load value corresponding to a service level required for the monitored system; Capacities further comprising service level determination means for determining whether or not the monitored system satisfies a service level based on the service level definition and the prediction formula or the predicted value calculated by the resource usage rate prediction means.
- City management support device Further storing a service level definition including a request value that is a load value corresponding to a service level required for the monitored system;
- Capacities further comprising service level determination means for determining whether or not the monitored system satisfies a service level based on the service level definition and the prediction formula or the predicted value calculated by the resource usage rate prediction means.
- the storage means Further storing a configuration definition for storing an application value indicating the current performance of the monitored system and an added value indicating a unit for adding the resource; Capacity management further comprising configuration determination means for determining whether the performance of the monitored system needs to be reinforced based on the configuration definition and the prediction formula and the prediction value calculated by the resource usage rate prediction means. Support device.
- a type definition that associates a log related to a resource and a log related to a load corresponding to the log related to the resource from the storage means, and an assumed load value that is a value of the load assumed for the monitored system
- Read the load definition that defines From the type definition obtain input information for specifying the association between the log related to the resource and the log related to the load
- a log type to be acquired is determined based on the input information and the type definition, and first log data obtained by extracting data related to the determined log type from a log held by the monitored system Acquired
- second log data which is data obtained by extracting a correspondence relationship between the specific resource and the specific load, is obtained from the first log data, and based on the second log data, resource use is acquired.
- a distribution density function indicating a true distribution of quantity and load value, a range satisfying a specific condition is selected from the distribution density function, and third log data that is data belonging to the range among the second log data Get
- a prediction formula related to a resource usage rate is calculated based on data of a certain threshold value or more in the third log data, and a predicted value of the resource usage rate is calculated based on the prediction formula and the load definition.
- a type definition that associates a log related to a resource and a log related to a load corresponding to the log related to the resource, and a load definition that defines an assumed load value that is a value of the load assumed for the monitored system Means for storing, Means for acquiring input information for identifying an association between a log related to the resource and a log related to the load from the type definition;
- a log type to be acquired is determined based on the input information and the type definition, and first log data obtained by extracting data related to the determined log type from a log held by the monitored system Means to obtain,
- second log data which is data obtained by extracting a correspondence relationship between the specific resource and the specific load, is obtained from the first log data, and based on the second log data, resource use is acquired.
- a distribution density function indicating a true distribution of quantity and load value, a range satisfying a specific condition is selected from the distribution density function, and third log data that is data belonging to the range among the second log data Means to obtain the A prediction formula related to a resource usage rate is calculated based on data of a certain threshold value or more in the third log data, and a predicted value of the resource usage rate is calculated based on the prediction formula and the load definition.
- the storage means Further storing a classification definition defining a method and conditions for classifying the data included in the third log data;
- the computer is Based on the classification definition, the data included in the third log data is classified as a plurality of fourth log data, A capacity management method for calculating the prediction formula of the resource usage based on the fourth log data.
- the storage means The primary log type that is a log type determined based on the input information and the load definition is associated with the secondary log type that is a log type correlated with the primary log type, and the primary log type resource Further storing a correlation definition that defines a pattern of the fourth log data based on a usage amount and a resource usage amount of the sub-log type;
- the computer is Based on the correlation definition, the secondary log type information is further added to the first log data; Based on the resource data and the load data related to the main log type among the second log data, the distribution density function is estimated, A capacity management method for further determining to which pattern the plurality of fourth log data belong based on the correlation definition.
- the storage means Further storing a safety factor definition including a safety factor corresponding to the resource type;
- the computer is A capacity management method for correcting the prediction formula and the predicted value of the resource usage rate based on the safety factor.
- (Appendix 12) In the capacity management method according to any one of appendices 7 and 9 to 11,
- the storage means A service level definition including a request value that is a load value corresponding to a service level required for the monitoring target system;
- the computer is A capacity management method for determining whether or not the monitored system satisfies a service level based on the service level definition and the prediction formula or the predicted value calculated by the resource usage rate prediction means.
- the storage means A configuration definition for storing an application value indicating the current performance of the monitored system and an addition value indicating a unit for adding the resource;
- the computer is A capacity management method for determining whether or not it is necessary to reinforce the performance of the monitored system based on the configuration definition and the prediction formula and the predicted value calculated by the resource usage rate prediction unit.
- Appendix 14 In the program described in Appendix 8, The computer, Means for further storing a classification definition defining a method and conditions for classifying data included in the third log data; Means for classifying data included in the third log data based on the classification definition into a plurality of fourth log data; A program for further functioning as means for calculating the prediction formula of the resource usage based on the fourth log data.
- the primary log type that is a log type determined based on the input information and the load definition is associated with the secondary log type that is a log type correlated with the primary log type, and the primary log type resource Means for further storing a correlation definition that defines a pattern of the fourth log data based on a usage amount and a resource usage amount of the sub-log type; Means for further adding the sub-log type information to the first log data based on the correlation definition; Means for estimating the distribution density function based on the resource data and the load data of the main log type among the second log data; A program for further functioning as means for further determining to which pattern a plurality of the fourth log data belong based on the correlation definition.
- Appendix 18 In the program according to any one of appendices 8, 14 to 17, The computer, Means for further storing a configuration definition for storing an applied value indicating the current performance of the monitored system and an added value indicating a unit for adding the resource; A program for further functioning as a means for determining whether or not it is necessary to reinforce the performance of the monitoring target system based on the configuration definition, the prediction formula of the resource usage rate, and the predicted value.
- the plurality of operations are not limited to being executed at different timings.
- another operation may occur during the execution of a certain operation, or the execution timing of a certain operation and another operation may partially or entirely overlap.
- each of the embodiments described above a certain operation is described as a trigger for another operation, but the description does not limit all relationships between the certain operation and other operations. For this reason, when each embodiment is implemented, the relationship between the plurality of operations can be changed within a range that does not hinder the contents.
- the specific description of each operation of each component does not limit each operation of each component. For this reason, each specific operation
- movement of each component may be changed in the range which does not cause trouble with respect to a functional, performance, and other characteristic in implementing each embodiment.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Quality & Reliability (AREA)
- Computer Hardware Design (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Probability & Statistics with Applications (AREA)
- Debugging And Monitoring (AREA)
Abstract
Description
リソースに係るログと、前記リソースに係るログに対応する負荷に係るログとを関連付けるタイプ定義、および、監視対象システムに対して想定される前記負荷の値である想定負荷値を定義する負荷定義を記憶する記憶手段と、
前記タイプ定義の中から、前記リソースに係るログと前記負荷に係るログとの関連付けを特定する入力情報を取得する入力手段と、
前記入力情報と前記タイプ定義とに基づいて取得するログタイプを決定し、前記監視対象システムが保持するログから、決定された前記ログタイプに関連するデータを抽出したものである第1ログデータを取得するログ取得手段と、
前記タイプ定義に基づき、前記第1ログデータから、特定の前記リソースと特定の前記負荷との対応関係を抽出したデータである第2ログデータを取得し、前記第2ログデータに基づき、リソース使用量と負荷値の真の分布を示す分布密度関数を推定し、前記分布密度関数から特定の条件を満たす範囲を選択し、前記第2ログデータのうち前記範囲に属するデータである第3ログデータを取得するログ分布推定手段と、
前記第3ログデータのうち、一定の閾値以上のデータに基づいて、リソース使用率に関する予測式を算出し、前記予測式と前記負荷定義とに基づいて、前記リソース使用率の予測値を算出する資源使用率予測手段と、を有するキャパシティ管理支援装置が提供される。
コンピュータが、
記憶手段から、リソースに係るログと、前記リソースに係るログに対応する負荷に係るログとを関連付けるタイプ定義、および、監視対象システムに対して想定される前記負荷の値である想定負荷値を定義する負荷定義を読み出し、
前記タイプ定義の中から、前記リソースに係るログと前記負荷に係るログとの関連付けを特定する入力情報を取得し、
前記入力情報と前記タイプ定義とに基づいて取得するログタイプを決定し、前記監視対象システムが保持するログから、決定された前記ログタイプに関連するデータを抽出したものである第1ログデータを取得し、
前記タイプ定義に基づき、前記第1ログデータから、特定の前記リソースと特定の前記負荷との対応関係を抽出したデータである第2ログデータを取得し、前記第2ログデータに基づき、リソース使用量と負荷値の真の分布を示す分布密度関数を推定し、前記分布密度関数から特定の条件を満たす範囲を選択し、前記第2ログデータのうち前記範囲に属するデータである第3ログデータを取得し、
前記第3ログデータのうち、一定の閾値以上のデータに基づいて、リソース使用率に関する予測式を算出し、前記予測式と前記負荷定義とに基づいて、前記リソース使用率の予測値を算出するキャパシティ管理方法が提供される。
コンピュータを、
リソースに係るログと、前記リソースに係るログに対応する負荷に係るログとを関連付けるタイプ定義、および、監視対象システムに対して想定される前記負荷の値である想定負荷値を定義する負荷定義を記憶する手段、
前記タイプ定義の中から、前記リソースに係るログと前記負荷に係るログとの関連付けを特定する入力情報を取得する手段、
前記入力情報と前記タイプ定義とに基づいて取得するログタイプを決定し、前記監視対象システムが保持するログから、決定された前記ログタイプに関連するデータを抽出したものである第1ログデータを取得する手段、
前記タイプ定義に基づき、前記第1ログデータから、特定の前記リソースと特定の前記負荷との対応関係を抽出したデータである第2ログデータを取得し、前記第2ログデータに基づき、リソース使用量と負荷値の真の分布を示す分布密度関数を推定し、前記分布密度関数から特定の条件を満たす範囲を選択し、前記第2ログデータのうち前記範囲に属するデータである第3ログデータを取得する手段、
前記第3ログデータのうち、一定の閾値以上のデータに基づいて、リソース使用率に関する予測式を算出し、前記予測式と前記負荷定義とに基づいて、前記リソース使用率の予測値を算出する手段として機能させるためのプログラムが提供される。
図1は、本発明の第1の実施形態に係るキャパシティ管理支援装置10の構成を示すブロック図である。キャパシティ管理支援装置10は、記憶部102、入力部104、ログ取得部106、ログ分布推定部108、資源使用率予測部110を有する。
本実施形態は、以下の点を除き、第1の実施形態と同様である。
本実施形態は、以下の点を除き、第2の実施形態と同様である。
本実施形態は、以下の点を除き、第1の実施形態と同様である。
本実施形態は、以下の点を除き、第1の実施形態と同様である。
本実施形態は、以下の点を除き、第1の実施形態と同様である。
(付記1)
リソースに係るログと、前記リソースに係るログに対応する負荷に係るログとを関連付けるタイプ定義、および、監視対象システムに対して想定される前記負荷の値である想定負荷値を定義する負荷定義を記憶する記憶手段と、
前記タイプ定義の中から、前記リソースに係るログと前記負荷に係るログとの関連付けを特定する入力情報を取得する入力手段と、
前記入力情報と前記タイプ定義とに基づいて取得するログタイプを決定し、前記監視対象システムが保持するログから、決定された前記ログタイプに関連するデータを抽出したものである第1ログデータを取得するログ取得手段と、
前記タイプ定義に基づき、前記第1ログデータから、特定の前記リソースと特定の前記負荷との対応関係を抽出したデータである第2ログデータを取得し、前記第2ログデータに基づき、リソース使用量と負荷値の真の分布を示す分布密度関数を推定し、前記分布密度関数から特定の条件を満たす範囲を選択し、前記第2ログデータのうち前記範囲に属するデータである第3ログデータを取得するログ分布推定手段と、
前記第3ログデータのうち、一定の閾値以上のデータに基づいて、リソース使用率に関する予測式を算出し、前記予測式と前記負荷定義とに基づいて、前記リソース使用率の予測値を算出する資源使用率予測手段と、を有するキャパシティ管理支援装置。
(付記2)
付記1に記載のキャパシティ管理支援装置において、
前記記憶手段は、
前記第3ログデータに含まれるデータを分類するための方法および条件を定義する分類定義をさらに記憶し、
当該キャパシティ管理支援装置は、
前記分類定義に基づき、前記第3ログデータに含まれるデータを分類し、複数の第4ログデータとするログ分類手段をさらに有し、
前記資源使用率予測手段は、
前記第4ログデータに基づいて、前記リソース使用量の予測式を算出するキャパシティ管理支援装置。
(付記3)
付記2に記載のキャパシティ管理支援装置において、
前記記憶手段は、
前記入力情報と前記負荷定義とに基づき決定されるログタイプである主ログタイプと、前記主ログタイプと相関を持つログタイプである従ログタイプとを対応づけ、また、前記主ログタイプのリソース使用量および前記従ログタイプのリソース使用量に基づき、前記第4ログデータのパターンを定める相関定義をさらに記憶し、
前記ログ取得手段は、
前記相関定義に基づき、前記第1ログデータに前記従ログタイプの情報をさらに付加し、
前記ログ分布推定手段は、
前記第2ログデータのうち、前記主ログタイプに係る前記リソースのデータと前記負荷のデータとに基づき、前記分布密度関数を推定し、
前記ログ分類手段は、
前記相関定義に基づき、複数の前記第4ログデータがどの前記パターンに属するかをさらに判定するキャパシティ管理支援装置。
(付記4)
付記1~3のいずれか一項に記載のキャパシティ管理支援装置において、
前記記憶手段は、
前記リソースの種類に対応する安全係数を含む安全率定義をさらに記憶し、
前記資源使用率予測手段は、
前記安全係数に基づき、前記リソース使用率の前記予測式および前記予測値を補正するキャパシティ管理支援装置。
(付記5)
付記1~4のいずれか一項に記載のキャパシティ管理支援装置において、
前記記憶手段は、
前記監視対象システムに要求されるサービスレベルに応じた負荷の値である要求値を含むサービスレベル定義をさらに記憶し、
前記サービスレベル定義と、前記資源使用率予測手段が算出した前記予測式または前記予測値とに基づき、前記監視対象システムがサービスレベルを満たしているか否かを判定するサービスレベル判定手段をさらに有するキャパシティ管理支援装置。
(付記6)
付記1~5のいずれか一項に記載のキャパシティ管理支援装置において、
前記記憶手段は、
前記監視対象システムの現在の性能を示す適用値および前記リソースを増設する単位を示す加算値を記憶する構成定義をさらに記憶し、
前記構成定義と、前記資源使用率予測手段が算出した前記予測式および前記予測値とに基づき、前記監視対象システムの性能の補強が必要か否かを判定する構成判定手段をさらに有するキャパシティ管理支援装置。
(付記7)
コンピュータが
記憶手段から、リソースに係るログと、前記リソースに係るログに対応する負荷に係るログとを関連付けるタイプ定義、および、監視対象システムに対して想定される前記負荷の値である想定負荷値を定義する負荷定義を読み出し、
前記タイプ定義の中から、前記リソースに係るログと前記負荷に係るログとの関連付けを特定する入力情報を取得し、
前記入力情報と前記タイプ定義とに基づいて取得するログタイプを決定し、前記監視対象システムが保持するログから、決定された前記ログタイプに関連するデータを抽出したものである第1ログデータを取得し、
前記タイプ定義に基づき、前記第1ログデータから、特定の前記リソースと特定の前記負荷との対応関係を抽出したデータである第2ログデータを取得し、前記第2ログデータに基づき、リソース使用量と負荷値の真の分布を示す分布密度関数を推定し、前記分布密度関数から特定の条件を満たす範囲を選択し、前記第2ログデータのうち前記範囲に属するデータである第3ログデータを取得し、
前記第3ログデータのうち、一定の閾値以上のデータに基づいて、リソース使用率に関する予測式を算出し、前記予測式と前記負荷定義とに基づいて、前記リソース使用率の予測値を算出するキャパシティ管理方法。
(付記8)
コンピュータを、
リソースに係るログと、前記リソースに係るログに対応する負荷に係るログとを関連付けるタイプ定義、および、監視対象システムに対して想定される前記負荷の値である想定負荷値を定義する負荷定義を記憶する手段、
前記タイプ定義の中から、前記リソースに係るログと前記負荷に係るログとの関連付けを特定する入力情報を取得する手段、
前記入力情報と前記タイプ定義とに基づいて取得するログタイプを決定し、前記監視対象システムが保持するログから、決定された前記ログタイプに関連するデータを抽出したものである第1ログデータを取得する手段、
前記タイプ定義に基づき、前記第1ログデータから、特定の前記リソースと特定の前記負荷との対応関係を抽出したデータである第2ログデータを取得し、前記第2ログデータに基づき、リソース使用量と負荷値の真の分布を示す分布密度関数を推定し、前記分布密度関数から特定の条件を満たす範囲を選択し、前記第2ログデータのうち前記範囲に属するデータである第3ログデータを取得する手段、
前記第3ログデータのうち、一定の閾値以上のデータに基づいて、リソース使用率に関する予測式を算出し、前記予測式と前記負荷定義とに基づいて、前記リソース使用率の予測値を算出する手段として機能させるためのプログラム。
(付記9)
付記7に記載のキャパシティ管理方法において、
前記記憶手段は、
前記第3ログデータに含まれるデータを分類するための方法および条件を定義する分類定義をさらに記憶しており、
前記コンピュータが、
前記分類定義に基づき、前記第3ログデータに含まれるデータを、複数の第4ログデータとして分類し、
前記第4ログデータに基づいて、前記リソース使用量の前記予測式を算出するキャパシティ管理方法。
(付記10)
付記9に記載のキャパシティ管理方法において、
前記記憶手段は、
前記入力情報と前記負荷定義とに基づき決定されるログタイプである主ログタイプと、前記主ログタイプと相関を持つログタイプである従ログタイプとを対応づけ、また、前記主ログタイプのリソース使用量および前記従ログタイプのリソース使用量に基づき、前記第4ログデータのパターンを定める相関定義をさらに記憶しており、
前記コンピュータが、
前記相関定義に基づき、前記第1ログデータに前記従ログタイプの情報をさらに付加し、
前記第2ログデータのうち、前記主ログタイプに係る前記リソースのデータと前記負荷のデータとに基づき、前記分布密度関数を推定し、
前記相関定義に基づき、複数の前記第4ログデータがどの前記パターンに属するかをさらに判定するキャパシティ管理方法。
(付記11)
付記7、9、10のいずれか一項に記載のキャパシティ管理方法において、
前記記憶手段は、
前記リソースの種類に対応する安全係数を含む安全率定義をさらに記憶しており、
前記コンピュータが、
前記安全係数に基づき、前記リソース使用率の前記予測式および前記予測値を補正するキャパシティ管理方法。
(付記12)
付記7、9~11のいずれか一項に記載のキャパシティ管理方法において、
前記記憶手段は、
前記監視対象システムに要求されるサービスレベルに応じた負荷の値である要求値を含むサービスレベル定義をさらに記憶しており、
前記コンピュータが、
前記サービスレベル定義と、前記資源使用率予測手段が算出した前記予測式または前記予測値とに基づき、前記監視対象システムがサービスレベルを満たしているか否かを判定するキャパシティ管理方法。
(付記13)
付記7、9~12のいずれか一項に記載のキャパシティ管理方法において、
前記記憶手段は、
前記監視対象システムの現在の性能を示す適用値および前記リソースを増設する単位を示す加算値を記憶する構成定義をさらに記憶しており、
前記コンピュータが、
前記構成定義と、前記資源使用率予測手段が算出した前記予測式および前記予測値とに基づき、前記監視対象システムの性能の補強が必要か否かを判定するキャパシティ管理方法。
(付記14)
付記8に記載のプログラムにおいて、
前記コンピュータを、
前記第3ログデータに含まれるデータを分類するための方法および条件を定義する分類定義をさらに記憶する手段、
前記分類定義に基づき、前記第3ログデータに含まれるデータを分類し、複数の第4ログデータとする手段、
前記第4ログデータに基づいて、前記リソース使用量の前記予測式を算出する手段としてさらに機能させるためのプログラム。
(付記15)
付記14に記載のプログラムにおいて、
前記コンピュータを、
前記入力情報と前記負荷定義とに基づき決定されるログタイプである主ログタイプと、前記主ログタイプと相関を持つログタイプである従ログタイプとを対応づけ、また、前記主ログタイプのリソース使用量および前記従ログタイプのリソース使用量に基づき、前記第4ログデータのパターンを定める相関定義をさらに記憶する手段、
前記相関定義に基づき、前記第1ログデータに前記従ログタイプの情報をさらに付加する手段、
前記第2ログデータのうち、前記主ログタイプに係る前記リソースのデータと前記負荷のデータとに基づき、前記分布密度関数を推定する手段、
前記相関定義に基づき、複数の前記第4ログデータがどの前記パターンに属するかをさらに判定する手段としてさらに機能させるためのプログラム。
(付記16)
付記8、14、15のいずれか一項に記載のプログラムにおいて、
前記コンピュータを、
前記リソースの種類に対応する安全係数を含む安全率定義をさらに記憶する手段、
前記安全係数に基づき、前記リソース使用率の前記予測式および前記予測値を補正する手段としてさらに機能させるためのプログラム。
(付記17)
付記8、14~16のいずれか一項に記載のプログラムにおいて、
前記コンピュータを、
前記監視対象システムに要求されるサービスレベルに応じた負荷の値である要求値を含むサービスレベル定義をさらに記憶する手段、
前記サービスレベル定義と、前記リソース使用率の前記予測式または前記予測値とに基づき、前記監視対象システムがサービスレベルを満たしているか否かを判定する手段としてさらに機能させるためのプログラム。
(付記18)
付記8、14~17のいずれか一項に記載のプログラムにおいて、
前記コンピュータを、
前記監視対象システムの現在の性能を示す適用値および前記リソースを増設する単位を示す加算値を記憶する構成定義をさらに記憶する手段、
前記構成定義と、前記リソース使用率の前記予測式および前記予測値とに基づき、前記監視対象システムの性能の補強が必要か否かを判定する手段としてさらに機能させるためのプログラム。
Claims (8)
- リソースに係るログと、前記リソースに係るログに対応する負荷に係るログとを関連付けるタイプ定義、および、監視対象システムに対して想定される前記負荷の値である想定負荷値を定義する負荷定義を記憶する記憶手段と、
前記タイプ定義の中から、前記リソースに係るログと前記負荷に係るログとの関連付けを特定する入力情報を取得する入力手段と、
前記入力情報と前記タイプ定義とに基づいて取得するログタイプを決定し、前記監視対象システムが保持するログから、決定された前記ログタイプに関連するデータを抽出したものである第1ログデータを取得するログ取得手段と、
前記タイプ定義に基づき、前記第1ログデータから、特定の前記リソースと特定の前記負荷との対応関係を抽出したデータである第2ログデータを取得し、前記第2ログデータに基づき、リソース使用量と負荷値の真の分布を示す分布密度関数を推定し、前記分布密度関数から特定の条件を満たす範囲を選択し、前記第2ログデータのうち前記範囲に属するデータである第3ログデータを取得するログ分布推定手段と、
前記第3ログデータのうち、一定の閾値以上のデータに基づいて、リソース使用率に関する予測式を算出し、前記予測式と前記負荷定義とに基づいて、前記リソース使用率の予測値を算出する資源使用率予測手段と、を有するキャパシティ管理支援装置。 - 請求項1に記載のキャパシティ管理支援装置において、
前記記憶手段は、
前記第3ログデータに含まれるデータを分類するための方法および条件を定義する分類定義をさらに記憶し、
当該キャパシティ管理支援装置は、
前記分類定義に基づき、前記第3ログデータに含まれるデータを分類し、複数の第4ログデータとするログ分類手段をさらに有し、
前記資源使用率予測手段は、
前記第4ログデータに基づいて、前記リソース使用量の予測式を算出するキャパシティ管理支援装置。 - 請求項2に記載のキャパシティ管理支援装置において、
前記記憶手段は、
前記入力情報と前記負荷定義とに基づき決定されるログタイプである主ログタイプと、前記主ログタイプと相関を持つログタイプである従ログタイプとを対応づけ、また、前記主ログタイプのリソース使用量および前記従ログタイプのリソース使用量に基づき、前記第4ログデータのパターンを定める相関定義をさらに記憶し、
前記ログ取得手段は、
前記相関定義に基づき、前記第1ログデータに前記従ログタイプの情報をさらに付加し、
前記ログ分布推定手段は、
前記第2ログデータのうち、前記主ログタイプに係る前記リソースのデータと前記負荷のデータとに基づき、前記分布密度関数を推定し、
前記ログ分類手段は、
前記相関定義に基づき、複数の前記第4ログデータがどの前記パターンに属するかをさらに判定するキャパシティ管理支援装置。 - 請求項1~3のいずれか一項に記載のキャパシティ管理支援装置において、
前記記憶手段は、
前記リソースの種類に対応する安全係数を含む安全率定義をさらに記憶し、
前記資源使用率予測手段は、
前記安全係数に基づき、リソース使用率の前記予測式および前記予測値を補正するキャパシティ管理支援装置。 - 請求項1~4のいずれか一項に記載のキャパシティ管理支援装置において、
前記記憶手段は、
前記監視対象システムに要求されるサービスレベルに応じた負荷の値である要求値を含むサービスレベル定義をさらに記憶し、
前記サービスレベル定義と、前記資源使用率予測手段が算出した予測式または予測値とに基づき、前記監視対象システムがサービスレベルを満たしているか否かを判定するサービスレベル判定手段をさらに有するキャパシティ管理支援装置。 - 請求項1~5のいずれか一項に記載のキャパシティ管理支援装置において、
前記記憶手段は、
前記監視対象システムの現在の性能を示す適用値および前記リソースを増設する単位を示す加算値を記憶する構成定義をさらに記憶し、
前記構成定義と、前記資源使用率予測手段が算出した予測式および予測値とに基づき、前記監視対象システムの性能の補強が必要か否かを判定する構成判定手段をさらに有するキャパシティ管理支援装置。 - コンピュータが、
記憶手段から、リソースに係るログと、前記リソースに係るログに対応する負荷に係るログとを関連付けるタイプ定義、および、監視対象システムに対して想定される前記負荷の値である想定負荷値を定義する負荷定義を読み出し、
前記タイプ定義の中から、前記リソースに係るログと前記負荷に係るログとの関連付けを特定する入力情報を取得し、
前記入力情報と前記タイプ定義とに基づいて取得するログタイプを決定し、前記監視対象システムが保持するログから、決定された前記ログタイプに関連するデータを抽出したものである第1ログデータを取得し、
前記タイプ定義に基づき、前記第1ログデータから、特定の前記リソースと特定の前記負荷との対応関係を抽出したデータである第2ログデータを取得し、前記第2ログデータに基づき、リソース使用量と負荷値の真の分布を示す分布密度関数を推定し、前記分布密度関数から特定の条件を満たす範囲を選択し、前記第2ログデータのうち前記範囲に属するデータである第3ログデータを取得し、
前記第3ログデータのうち、一定の閾値以上のデータに基づいて、リソース使用率に関する予測式を算出し、前記予測式と前記負荷定義とに基づいて、前記リソース使用率の予測値を算出するキャパシティ管理方法。 - コンピュータを、
リソースに係るログと、前記リソースに係るログに対応する負荷に係るログとを関連付けるタイプ定義、および、監視対象システムに対して想定される前記負荷の値である想定負荷値を定義する負荷定義を記憶する手段、
前記タイプ定義の中から、前記リソースに係るログと前記負荷に係るログとの関連付けを特定する入力情報を取得する手段、
前記入力情報と前記タイプ定義とに基づいて取得するログタイプを決定し、前記監視対象システムが保持するログから、決定された前記ログタイプに関連するデータを抽出したものである第1ログデータを取得する手段、
前記タイプ定義に基づき、前記第1ログデータから、特定の前記リソースと特定の前記負荷との対応関係を抽出したデータである第2ログデータを取得し、前記第2ログデータに基づき、リソース使用量と負荷値の真の分布を示す分布密度関数を推定し、前記分布密度関数から特定の条件を満たす範囲を選択し、前記第2ログデータのうち前記範囲に属するデータである第3ログデータを取得する手段、
前記第3ログデータのうち、一定の閾値以上のデータに基づいて、リソース使用率に関する予測式を算出し、前記予測式と前記負荷定義とに基づいて、前記リソース使用率の予測値を算出する手段、として機能させるためのプログラム。
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US14/002,679 US20140289735A1 (en) | 2012-03-02 | 2013-01-17 | Capacity management support apparatus, capacity management method and program |
JP2013528157A JP5354138B1 (ja) | 2012-03-02 | 2013-01-17 | キャパシティ管理支援装置、キャパシティ管理方法およびプログラム |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2012047305 | 2012-03-02 | ||
JP2012-047305 | 2012-03-02 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2013128789A1 true WO2013128789A1 (ja) | 2013-09-06 |
Family
ID=49082008
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2013/000188 WO2013128789A1 (ja) | 2012-03-02 | 2013-01-17 | キャパシティ管理支援装置、キャパシティ管理方法およびプログラム |
Country Status (3)
Country | Link |
---|---|
US (1) | US20140289735A1 (ja) |
JP (1) | JP5354138B1 (ja) |
WO (1) | WO2013128789A1 (ja) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2016053803A (ja) * | 2014-09-03 | 2016-04-14 | 株式会社東芝 | 電子機器、方法及びプログラム |
JP2018124878A (ja) * | 2017-02-02 | 2018-08-09 | 富士通株式会社 | 性能要件推定プログラム、性能要件推定装置、および性能要件推定方法 |
CN111046045A (zh) * | 2019-12-13 | 2020-04-21 | 中国平安财产保险股份有限公司 | 处理数据倾斜的方法、装置、设备及存储介质 |
US10642864B2 (en) | 2014-03-18 | 2020-05-05 | Nec Corporation | Information processing device and clustering method |
CN113312485A (zh) * | 2021-06-25 | 2021-08-27 | 展讯通信(上海)有限公司 | 日志自动化分类方法及装置、计算机可读存储介质 |
CN117331795A (zh) * | 2023-12-01 | 2024-01-02 | 南京研利科技有限公司 | 服务指标计算方法与系统 |
Families Citing this family (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP6307862B2 (ja) * | 2013-02-06 | 2018-04-11 | 株式会社リコー | 情報処理システム、情報処理方法およびプログラム |
EP3021251A1 (en) | 2014-11-14 | 2016-05-18 | Hewlett-Packard Development Company, L.P. | Dynamically adjusting a model for a security operations center |
US9507636B2 (en) * | 2015-04-20 | 2016-11-29 | International Business Machines Corporation | Resource management and allocation using history information stored in application's commit signature log |
KR102461022B1 (ko) * | 2015-10-26 | 2022-11-01 | 에스케이플래닛 주식회사 | 로그 데이터 분석 방법 및 장치 |
CN110019110B (zh) * | 2017-07-28 | 2022-11-18 | 腾讯科技(深圳)有限公司 | 一种业务系统的容量管理方法、装置、设备及业务系统 |
CN111061697B (zh) * | 2019-12-25 | 2023-06-13 | 中国联合网络通信集团有限公司 | 日志数据处理方法、装置、电子设备及存储介质 |
KR102405933B1 (ko) * | 2020-02-27 | 2022-06-08 | 숭실대학교 산학협력단 | 학습에 의한 머신 러닝 실행 관리 플랫폼 시스템 및 방법 |
JP2022107464A (ja) * | 2021-01-08 | 2022-07-21 | 富士通株式会社 | 情報収集プログラム、情報収集方法および情報処理装置 |
US20230222005A1 (en) * | 2022-01-11 | 2023-07-13 | Adobe Inc. | Shared Resource Interference Detection involving a Virtual Machine Container |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2009008279A1 (ja) * | 2007-07-10 | 2009-01-15 | Nec Corporation | コンピュータシステム、管理装置、及びコンピュータシステム管理方法 |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5481694A (en) * | 1991-09-26 | 1996-01-02 | Hewlett-Packard Company | High performance multiple-unit electronic data storage system with checkpoint logs for rapid failure recovery |
US7512940B2 (en) * | 2001-03-29 | 2009-03-31 | Microsoft Corporation | Methods and apparatus for downloading and/or distributing information and/or software resources based on expected utility |
KR101207510B1 (ko) * | 2008-12-18 | 2012-12-03 | 한국전자통신연구원 | 클러스터 데이터 관리시스템 및 클러스터 데이터 관리 시스템에서 공유 재수행 로그를 이용한 데이터 재구축 방법 |
JP5825120B2 (ja) * | 2012-01-27 | 2015-12-02 | 富士通株式会社 | 制御システム、演算処理装置及びログ配信方法 |
-
2013
- 2013-01-17 WO PCT/JP2013/000188 patent/WO2013128789A1/ja active Application Filing
- 2013-01-17 US US14/002,679 patent/US20140289735A1/en not_active Abandoned
- 2013-01-17 JP JP2013528157A patent/JP5354138B1/ja not_active Expired - Fee Related
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2009008279A1 (ja) * | 2007-07-10 | 2009-01-15 | Nec Corporation | コンピュータシステム、管理装置、及びコンピュータシステム管理方法 |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10642864B2 (en) | 2014-03-18 | 2020-05-05 | Nec Corporation | Information processing device and clustering method |
JP2016053803A (ja) * | 2014-09-03 | 2016-04-14 | 株式会社東芝 | 電子機器、方法及びプログラム |
JP2018124878A (ja) * | 2017-02-02 | 2018-08-09 | 富士通株式会社 | 性能要件推定プログラム、性能要件推定装置、および性能要件推定方法 |
CN111046045A (zh) * | 2019-12-13 | 2020-04-21 | 中国平安财产保险股份有限公司 | 处理数据倾斜的方法、装置、设备及存储介质 |
CN111046045B (zh) * | 2019-12-13 | 2023-09-29 | 中国平安财产保险股份有限公司 | 处理数据倾斜的方法、装置、设备及存储介质 |
CN113312485A (zh) * | 2021-06-25 | 2021-08-27 | 展讯通信(上海)有限公司 | 日志自动化分类方法及装置、计算机可读存储介质 |
CN113312485B (zh) * | 2021-06-25 | 2022-11-29 | 展讯通信(上海)有限公司 | 日志自动化分类方法及装置、计算机可读存储介质 |
CN117331795A (zh) * | 2023-12-01 | 2024-01-02 | 南京研利科技有限公司 | 服务指标计算方法与系统 |
CN117331795B (zh) * | 2023-12-01 | 2024-01-26 | 南京研利科技有限公司 | 服务指标计算方法与系统 |
Also Published As
Publication number | Publication date |
---|---|
JPWO2013128789A1 (ja) | 2015-07-30 |
US20140289735A1 (en) | 2014-09-25 |
JP5354138B1 (ja) | 2013-11-27 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP5354138B1 (ja) | キャパシティ管理支援装置、キャパシティ管理方法およびプログラム | |
US10318366B2 (en) | System and method for relationship based root cause recommendation | |
JP5831558B2 (ja) | 運用管理装置、運用管理方法、及びプログラム | |
JP6248560B2 (ja) | 管理プログラム、管理方法、および管理装置 | |
JP5874936B2 (ja) | 運用管理装置、運用管理方法、及びプログラム | |
US20190245766A1 (en) | Performance evaluation method, apparatus for performance evaluation, and non-transitory computer-readable storage medium for storing program | |
US8214365B1 (en) | Measuring confidence of file clustering and clustering based file classification | |
JP6183450B2 (ja) | システム分析装置、及び、システム分析方法 | |
US8874642B2 (en) | System and method for managing the performance of an enterprise application | |
US9558091B2 (en) | Information processing device, fault avoidance method, and program storage medium | |
WO2020062702A1 (zh) | 短信发送的方法、装置、计算机设备和存储介质 | |
US9524223B2 (en) | Performance metrics of a computer system | |
JP5768983B2 (ja) | 契約違反予測システム、契約違反予測方法および契約違反予測プログラム | |
JP6183449B2 (ja) | システム分析装置、及び、システム分析方法 | |
US20210048811A1 (en) | Model generation device for life prediction, model generation method for life prediction, and recording medium storing model generation program for life prediction | |
WO2017126585A1 (ja) | 情報処理装置、情報処理方法、及び、記録媒体 | |
KR20190055613A (ko) | 클라우드 컴퓨팅 시스템에서의 자원 제어 장치 및 방법 | |
US8307011B2 (en) | System and method for determining overall utilization | |
KR102269647B1 (ko) | 서버 성능 모니터링 장치 | |
KR101908377B1 (ko) | 계량데이터 관리 시스템 및 컴퓨터 판독가능 기록 매체 | |
JP6599049B2 (ja) | データ収集装置 | |
WO2022247600A1 (zh) | 数据处理方法及设备、计算机可读存储介质 | |
US11616704B2 (en) | Information processing method, storage medium, and information processing device | |
CN109766243B (zh) | 一种基于幂函数的多核主机性能监控方法 | |
US9054995B2 (en) | Method of detecting measurements in service level agreement based systems |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
ENP | Entry into the national phase |
Ref document number: 2013528157 Country of ref document: JP Kind code of ref document: A |
|
WWE | Wipo information: entry into national phase |
Ref document number: 14002679 Country of ref document: US |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 13755146 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 13755146 Country of ref document: EP Kind code of ref document: A1 |