JP5758783B2 - Analysis device, analysis method, and computer program - Google Patents

Description

  The present invention relates to an analysis technique, and more particularly to an analysis apparatus and an analysis method for analyzing a financial product business system.

  Online services that allow users to buy and sell stocks and trade foreign currencies via the Internet are widespread. Due to its simplicity and low fees, the number of online service users is increasing year by year. As the number of users increases, the load on securities business systems that provide online services is also increasing. In online services, immediacy of transactions is particularly important, and system down due to overload must be avoided as much as possible.

  Therefore, the present applicant has proposed a technique for predicting the load on the securities business system in Patent Document 1 and Patent Document 2. Based on the predicted load as described above, it is desirable to perform processing for ensuring that the capacity of the system sufficiently exceeds the predicted load. The present applicant proposes an example of such processing in Patent Document 3 and Patent Document 4.

JP 2007-265244 A JP 2007-265245 A JP 2010-39663 A JP 2010-152818 A

  With the diversification and higher functionality of online services, the configuration of securities business systems that support them, such as server configurations and application configurations, has become more complex. In other words, in the securities business system, a wide variety of servers are organically coupled, and a variety of applications are being implemented on that. This applies not only to securities business systems but also to general financial product business systems. There is a demand for the emergence of analytical techniques adapted to such complex financial product business systems.

  The present invention has been made in view of these problems, and an object thereof is to provide an analysis technique that can be adapted to the complexity of a financial product business system.

  One embodiment of the present invention relates to an analyzer. This analysis device obtains history data in association with the degree of use of a plurality of applications that realize a plurality of functions of a financial product business system and the degree of use of resources used by the plurality of applications as history data Multiple regression analysis is performed on the history data acquired by the data acquisition unit and the history acquisition unit, with each of the usage levels of the plurality of applications as explanatory variables and the resource usage level as an objective variable. A multiple regression analysis unit for deriving coefficients.

  According to this aspect, it is possible to know the influence of the degree of use of the corresponding application on the degree of resource use based on the derived partial regression coefficient.

  It should be noted that any combination of the above-described constituent elements, or those obtained by replacing the constituent elements and expressions of the present invention with each other between apparatuses, methods, systems, computer programs, recording media storing computer programs, and the like are also included in the present invention. It is effective as an embodiment of

  According to the present invention, it is possible to provide an analysis technique that can be adapted to the complexity of a financial product business system.

It is a schematic diagram which shows the structure of the securities business system which is the analysis object of the analyzer which concerns on embodiment. It is a block diagram which shows the function and structure of the analyzer which concerns on embodiment, and the member connected to it. It is a data structure figure which shows an example of the log | history holding part of FIG. It is a representative screen figure of an analysis parameter designation screen. It is a typical screen figure of a result display screen. It is explanatory drawing which shows typically the prediction model about the application of the prediction object produced | generated by the prediction model production | generation part of FIG. It is explanatory drawing for demonstrating the use restriction | limiting mode set by the restriction | limiting setting part of FIG. It is a typical screen figure of a volume ranking screen. It is a typical screen figure of a correlation display screen. It is a flowchart which shows a series of processes in the analyzer of FIG.

  Hereinafter, the same or equivalent components, members, and processes shown in the drawings are denoted by the same reference numerals, and repeated description is appropriately omitted.

  The securities business system which is the analysis target of the analysis apparatus according to the embodiment is a variety of useful services for users of securities companies such as securities trading, foreign currency trading, margin trading, and securities ordering via a network such as the Internet. Provide online services that can acquire various information. The securities business system has a plurality of resources, that is, a server and a communication device, and at least an order inquiry function that allows the user to inquire about his / her order status by the server and the communication device, and an order processing function for processing the order. , Ordering and receiving order processing function, daily transaction detail editing function for editing daily transaction details, contract calculation function for executing contract calculation, and screen display function for displaying top screen etc. on the display of user terminal And an order confirmation processing function for confirming the order and a market price inquiry function that enables the market price of the stock to be queried. The order inquiry function and the order receiving / ordering function are functions that mediate securities transactions via a network. Each of the order inquiry function, the order processing function, the ordering / ordering function, and the screen display function includes a general customer and a branch.

In the securities business system, a plurality of applications for realizing the plurality of functions are installed. The analysis device acquires a history of the degree of use of the installed application, that is, the number of requests processed by the application per unit time (for example, 1 minute) (hereinafter referred to as the number of processing cases).
Although there may be a case where a plurality of equivalent applications for one function are provided in parallel, in this embodiment, these application groups may be handled as one application.

  The analysis apparatus acquires a history of the degree of use such as a CPU (central processing unit) usage rate, power consumption, storage free capacity, and communication packet amount of a server used by a plurality of applications. In particular, the analysis apparatus obtains a history of the CPU usage rate of the database server.

  The analysis apparatus performs a multiple regression analysis using the CPU usage rate of the database server as an objective variable and each of the number of processes of a plurality of applications as an explanatory variable, and derives a partial regression coefficient corresponding to each explanatory variable. This partial regression coefficient is a good statistical indicator of how much load is applied to the database server by application processing. That is, such an index can be obtained without performing a performance test for each application and collecting facts. Currently, with the diversification and high functionality of securities business, a large number of applications are implemented in securities business systems, and it is impossible to perform such performance tests on each of them in terms of cost and time. close. Therefore, when the analysis apparatus according to the present embodiment is used, the degree of application influence on the database server can be derived without performing such a performance test.

(Securities business system 200)
FIG. 1 is a schematic diagram illustrating a configuration of a securities business system 200 that is an analysis target of the analysis apparatus according to the embodiment.
The securities business system 200 includes a presentation layer 210, a business logic layer 220, and a database layer 230. The presentation layer 210 mainly receives data input from the client terminals 202, 204, and 206 and passes the data to the business logic layer 220, or processes the data provided from the business logic layer 220 into a web page and processes the client terminal 202. , 204, 206. The business logic layer 220 has a role in charge of actual processing such as ordering / buying securities and providing information.

  The presentation layer 210 includes a plurality of web servers 212, 214, 216, 218. In the present embodiment, a group of servers that execute web page processing is prepared for each access channel to the securities business system 200. In FIG. 1, it is assumed that the securities business system 200 is accessed from a general customer via the web, a call center, and a branch using the client terminals 202, 204, and 206. Therefore, the presentation layer 210 includes a general web server 212 responsible for processing for access from general customers, a call center web server 214 responsible for processing for access from a call center, and a branch responsible for processing for access from a branch. Web server 216. Further, a spare web server 218 that is normally idle may be prepared. Each web server 212, 214, 216, 218 may be composed of a plurality of servers. Each of the web servers 212, 214, 216, and 218 has a dynamic page generation function and a transaction management function that execute a program in response to a request from the client terminals 202, 204, and 206 and transmit the result. The application that realizes each function is implemented.

  In the business logic layer 220, a group of servers that execute a predetermined application exclusively, not for each access channel, is prepared. The business logic layer 220 includes an information providing application server 222 that executes an application for providing information necessary for securities trading such as current and past stock prices, trading volumes, and company information to the client terminals 202, 204, and 206, An order processing application server 224 that executes an application for receiving trade orders such as investment trusts, and a business processing application server 226 that executes an application that manages user login and logout, customer information, and the like are provided. In addition to this, for example, an external linkage application server that provides economic news in cooperation with an external newspaper company site, a report application server that provides a transaction balance report for each customer, etc. Also good. Each application server 222, 224, 226 may be composed of a plurality of servers. Each of the application servers 222, 224, and 226 has the functions of the general application server in addition to the business processing functions described above, for example, a connection function to the database server of the database layer 230, and a transaction that connects a plurality of processes. It also has management functions, etc., and implements applications that realize those functions.

  In the database layer 230, a database server (database, DB) that is referred to or updated from the business logic layer 220 is prepared. The database layer 230 includes a main DB server 232 serving as a main database of each server of the business logic layer 220, and an information system DB server 234 provided mainly for reference purposes.

  In each application server 222, 224, 226, services 242, 244, 246, 248, 250, 252 for a plurality of client terminals are executed simultaneously. For example, in the information providing application server 222, a service A242 that provides stock price information to general customers and a service B244 that provides stock price information to a call center may be executed simultaneously. In the order processing application server 224, a service C246 for processing stock trading orders from a call center and a service D248 for processing investment trust trading orders from a store may be executed simultaneously.

  General customer, call center, and branch office staff use the software such as a web browser from the client terminals 202, 204, and 206 to access the securities business system 200 via the network 208 such as the Internet. Client terminals 204 and 206 provided in a call center or a branch may be connected to the securities business system 200 by a dedicated line 209.

  When the securities business system 200 is accessed from the client terminals 202, 204, 206, web pages created by the web servers 212, 214, 216 corresponding to the access channel of the presentation layer 210 are transmitted to the client terminals 202, 204, 206. The The client terminals 202, 204, and 206 display the transmitted web page on the display. When the user performs operations such as login, information browsing, and sales order on the web page, the information is passed to the application servers 222, 224, and 226 of the business logic layer 220 via the web servers 212, 214, and 216. The results of processing by the application servers 222, 224, and 226 are processed into web pages by the web servers 212, 214, and 216, and the data is transmitted to the client terminals 202, 204, and 206. In this way, the user can refer to various information displayed on the web page, or place an order for buying and selling securities through the web page.

  In the securities business system as described above, if the load on the server in the system is excessive due to an increase in the number of accesses, there is a possibility that normal services cannot be provided. Since such a situation inconveniences the customer, the degree of use of resources in the securities business system is predicted, and the system configuration is systematically changed based on the prediction, for example, a process such as bringing a spare server online. It is desirable to do. In particular, since the main DB server 232 of the database layer 230 is referenced by a plurality of applications, if this main DB server 232 malfunctions, many services of the securities business system may be affected. Therefore, the analysis apparatus 100 according to the present embodiment predicts the CPU usage rate of the main DB server 232 and presents it to the administrator. As a result, the administrator can take measures in advance so that the main DB server 232 is not overloaded.

  Further, the analysis apparatus 100 can identify applications that have a relatively large contribution to the CPU usage rate of the main DB server 232 without performing a performance test on each application. As a result, the administrator can make the system more efficient by taking measures such as adjustments and restrictions on congestion when the application has a large contribution.

  FIG. 2 is a block diagram showing functions and configurations of the analyzer 100 according to the present embodiment and members connected thereto. Each block shown here can be realized by hardware and other elements such as a computer CPU and a mechanical device, and software can be realized by a computer program or the like. Draw functional blocks. Therefore, it is understood by those skilled in the art who have touched this specification that these functional blocks can be realized in various forms by a combination of hardware and software.

  The analysis apparatus 100 may be provided inside the securities business system 200 or may be provided outside. In the present embodiment, the analysis device 100 is provided outside the securities business system 200, and the analysis device 100 and the securities business system 200 are connected via a network 208. The analysis apparatus 100 acquires the operation history of each server and the usage frequency of applications from the securities business system 200 via the network 208. The analysis device 100 is connected to an input device 102 such as a mouse and a keyboard and a display 104. The administrator of the securities business system 200 uses the analysis device 100 to analyze the performance of the securities business system 200 and predict the load on the securities business system 200.

The analysis apparatus 100 includes a history acquisition unit 110, a history holding unit 112, a multiple regression analysis unit 114, a presentation unit 116, a restriction unit 118, a prediction unit 120, and a variable selection support unit 170.
The history acquisition unit 110 acquires the processing number of each application of the securities business system 200 and the CPU usage rate of the main DB server 232 in association with each other via the network 208 as history data. The history acquisition unit 110 includes a processing number acquisition unit 122 and a CPU usage rate acquisition unit 124.

  The processing number acquisition unit 122 acquires the processing number of each application from the presentation layer 210 and the business logic layer 220 of the securities business system 200 via the network 208. The processing number acquisition unit 122 acquires the processing number of applications already calculated in the presentation layer 210 and the business logic layer 220 at a predetermined time interval, for example, one minute interval. The processing number acquisition unit 122 registers the processing number of each application and the acquisition time at which the processing number is acquired in the history holding unit 112 in association with each other. In the securities business system 200, since the number of processes of each application is updated as needed, it can be said that the acquisition time when the number of processes is acquired is the time when the number of processes occurred.

  The processing number acquisition unit 122 acquires log data related to the processing start time and processing end time of each application request from the presentation layer 210 and the business logic layer 220, and each application per unit time is acquired from the acquired log data. The number of requests to be processed may be calculated.

  The CPU usage rate acquisition unit 124 acquires the CPU usage rate of the main DB server 232 from the main DB server 232 via the network 208 at a time interval equivalent to the time interval in the processing number acquisition unit 122. Since the server usually has a function of reporting the current CPU usage rate to the outside, the CPU usage rate acquisition unit 124 may acquire the CPU usage rate of the main DB server 232 using the function. The CPU usage rate acquisition unit 124 registers the CPU usage rate of the main DB server 232 and the acquisition time at which the CPU usage rate was acquired in the history holding unit 112 in association with each other.

  FIG. 3 is a data structure diagram illustrating an example of the history holding unit 112. The history holding unit 112 holds the history data acquired by the history acquisition unit 110. In particular, the history holding unit 112 holds the number of processes for each application, the CPU usage rate of the main DB server 232, and the acquisition time in association with each other. In FIG. 3, the application is indicated by the name of the corresponding function. In FIG. 3, the time interval for acquiring the history is a unit time, that is, 1 minute. In FIG. 3, the CPU usage rate is shown as a percentage.

  Returning to FIG. 2, the multiple regression analysis unit 114 sets each of the processing numbers of a plurality of applications designated by the administrator as explanatory variables for the history data held by the history holding unit 112, and A partial regression coefficient is derived by performing a multiple regression analysis with the CPU usage rate as an objective variable. The multiple regression analysis unit 114 includes a variable designation receiving unit 126, a period designation receiving unit 128, an analysis target data extracting unit 130, a partial regression coefficient deriving unit 132, and a multiple correlation deriving unit 134.

  The variable designation receiving unit 126 receives designation of explanatory variables and objective variables used in the multiple regression analysis from the administrator via the input device 102. The period designation accepting unit 128 accepts designation of a range of history data to be subjected to multiple regression analysis from the administrator via the input device 102. This range is specified by a period.

  FIG. 4 is a representative screen diagram of the analysis parameter designation screen 300. The multiple regression analysis unit 114 causes the display 104 to display an analysis parameter designation screen 300 that accepts designation of parameters for multiple regression analysis. The analysis parameter designation screen 300 has an OK button 302, a target variable selection area 304 configured to allow the administrator to select a variable in a pull-down format, and an explanation configured to allow the administrator to select a plurality of variables in a pull-down format. It has a variable selection area 306 and an analysis target period designation area 308 into which an analysis target period in which the acquisition time of history data to be analyzed is to be entered. When a variable is selected in the target variable selection area 304, a plurality of variables are selected in the explanatory variable selection area 306, an analysis target period is input in the analysis target period specification area 308, and the OK button 302 is pressed, a variable specification receiving unit 126 accepts a variable selected in the objective variable selection area 304 as a designated objective variable and a plurality of variables selected in the explanatory variable selection area 306 as a plurality of designated explanatory variables. The period designation accepting unit 128 accepts the analysis target period input in the analysis target period designation area 308 as the designated period.

Returning to FIG. 2, the analysis target data extraction unit 130 extracts the history data of the specified objective variable and the specified plurality of explanatory variables in the specified period from the history holding unit 112 as analysis target data.
The partial regression coefficient deriving unit 132 determines the partial regression coefficient by the least square method. The partial regression coefficient deriving unit 132 calculates the difference between the predicted value of the objective variable obtained by substituting the values of the plurality of explanatory variables included in the analysis target data into the multiple regression equation and the actual measurement value of the target variable included in the analysis target data. The partial regression coefficient corresponding to each explanatory variable is determined so that the sum of squares is minimized. Specifically, the partial regression coefficient deriving unit 132 calculates an inverse matrix of the deviation square sum / product sum matrix of the explanatory variables. The partial regression coefficient deriving unit 132 calculates the partial regression coefficient by calculating the product of the calculated inverse matrix and the deviation product sum matrix of the explanatory variable and the objective variable. The partial regression coefficient deriving unit 132 may calculate a standard partial regression coefficient as the partial regression coefficient. The partial regression coefficient derived in this way indicates how much the processing of one request by the corresponding application gives a load to the main DB server 232.
The multiple correlation deriving unit 134 derives a multiple correlation coefficient from the ratio of the fluctuation due to the regression among the total fluctuations.

The presentation unit 116 presents the analysis result in the multiple regression analysis unit 114 to the administrator. In particular, the presentation unit 116 causes the display 104 to display a result display screen 310 that indicates the size of the partial regression coefficient and the multiple correlation coefficient corresponding to each explanatory variable, that is, each application.
FIG. 5 is a representative screen diagram of the result display screen 310. The result display screen 310 shows the size of the partial regression coefficient for each application as a bar graph. Further, the result display screen 310 shows the multiple correlation coefficient as a numerical value. The administrator can know the degree of contribution of each application to the CPU usage rate of the main DB server 232 by referring to the result display screen 310.

  Returning to FIG. 2, the prediction unit 120 includes a prediction model generation unit 136, a processing number prediction unit 138, and a CPU usage rate prediction unit 140. The prediction model generation unit 136 generates a prediction model based on the history of the number of processed applications held by the history holding unit 112. In particular, the prediction model generation unit 136 refers to the history holding unit 112, and identifies the day in which the sum total of the number of requests processed by the prediction target application is relatively large in the past several months. The prediction model generation unit 136 extracts the number of processing cases from 9:00 to 15:00 on the identified date from the history holding unit 112. The prediction model generation unit 136 calculates the sum of the number of requests processed by the prediction target application in the time zone from 9:00 to 15:00 on the identified date from the extracted number of processing cases. The prediction model generation unit 136 calculates the ratio of the number of processes at each acquisition time to the calculated total. The prediction model generation unit 136 generates a prediction model by associating the acquisition time with the calculated ratio.

  FIG. 6 is an explanatory diagram schematically illustrating a prediction model for a prediction target application generated by the prediction model generation unit 136. In this prediction model, the time period from 9:00 to 15:00 in one day is divided in units of one minute, and the ratio of the number of processing cases is associated with the divided sections. In this prediction model, it is predicted that the number of processing cases will be maximized around 9:13.

  Returning to FIG. 2, the processing number prediction unit 138 uses the prediction model generated by the prediction model generation unit 136 to predict the future processing number of the prediction target application. In particular, the processing number prediction unit 138 obtains a predicted value of the total number of processing cases from 9:00 to 15:00 on the prediction target day for the prediction target application from the user via the input device 102. The processing number prediction unit 138 calculates the product of the ratio associated with the time in the prediction model and the predicted value of the total processing number as the predicted value of the processing number at that time. For example, assuming that the predicted value of the total number of processes on the prediction target day is 10000 and the ratio associated with 9:20 on the prediction target day is 0.001, the prediction of the processing number on the prediction target day 9:20 The value is 10000 × 0.001 = 10.

  In the above description, the case where the processing number acquisition time interval in the processing number acquisition unit 122 and the unit time for deriving the processing number are both equal to 1 minute, but is not limited thereto. May be different. In that case, the prediction model configuration and the processing number prediction formula take into account the ratio between the time interval and the unit time in the processing number acquisition unit 122.

The CPU usage rate prediction unit 140 obtains, from the outside or the processing number prediction unit 138, the predicted value of the processing number of the plurality of applications adopted as the explanatory variable in the multiple regression analysis unit 114. The CPU usage rate prediction unit 140 predicts the CPU usage rate of the main DB server 232 by substituting the acquired predicted value into a multiple regression equation based on the partial regression coefficient derived by the multiple regression analysis unit 114. The multiple regression equation is, for example, as follows.
CPU usage of main DB server 232 (%)
= 3 (%) + Correction value (CPU usage rate of base of main DB server 232) +
(Number of branch order processing applications x 0.12) +
(Number of branch order inquiry application processing times × 0.16) +
(Number of processed general customer order processing applications x 0.02) +
(General customer order inquiry application processing count × 0.05) +
(General customer screen display application processing count × 0.23) +
(Number of deal calculation application processes × 0.01)
Coefficients such as “0.12” and “0.16” in the multiple regression equation are partial regression coefficients derived by the multiple regression analysis unit 114.
The CPU usage rate prediction unit 140 causes the display 104 to display the predicted CPU usage rate value.

  The restriction unit 118 includes an application specifying unit 142, a monitoring unit 144, a determination unit 146, and a restriction setting unit 148. The application specifying unit 142 acquires the partial regression coefficient derived by the multiple regression analysis unit 114. The application specifying unit 142 compares the acquired partial regression coefficient with a predetermined coefficient threshold value. When the acquired partial regression coefficient is larger than the coefficient threshold, the application specifying unit 142 specifies an application corresponding to the partial regression coefficient.

  The monitoring unit 144 acquires the processing number of applications specified by the application specifying unit 142 (hereinafter referred to as a specific application) from the securities business system 200 via the network 208. The monitoring unit 144 monitors the number of processes by comparing the acquired number of processes with a predetermined number threshold.

  The determination unit 146 determines whether or not to restrict the use of the specific application based on the monitoring result in the monitoring unit 144. The determination unit 146 determines that the use of the specific application should be restricted when a monitoring result indicating that the number of processing cases is greater than the threshold value is obtained from the monitoring unit 144.

  When the determination unit 146 determines that the restriction should be restricted, the restriction setting unit 148 changes the operation of the securities business system 200 via the network 208 so that the use of the specific application is restricted. For example, the restriction setting unit 148 may restrict a new login to a service provided by the securities business system 200, or may restrict the number of activations when a plurality of specific applications are provided in parallel. The queue depth of a specific application may be redesigned.

FIG. 7 is an explanatory diagram for explaining the use restriction mode set by the restriction setting unit 148. FIG. 7 shows a case where the application specifying unit 142 specifies the application 260 that realizes the general customer order processing function, and the determining unit 146 determines that the use of the application should be restricted. A plurality of applications 260 that realize the general customer order processing function are activated in parallel. The restriction setting unit 148 sets the order processing application server 224 so as to limit the number of activations to a predetermined number or less via the network 208 in the use restriction mode. In addition, the queue depth 262 to the application 260 that realizes the general customer order processing function is set so that the queued request is processed with a waiting time that is equal to or shorter than a predetermined maximum waiting time. In other words, when the request is concentrated, the idea is to immediately return an error with a relatively higher probability than to increase the response time.
By setting such a use restriction mode, the load on the main DB server 232 can be suppressed.

  Returning to FIG. The case where the variable designation receiving unit 126 receives designation of explanatory variables and objective variables used in the multiple regression analysis from the administrator via the input device 102 has been described above. Basically, the administrator can arbitrarily select a variable. In this case, a method of finding a “good” explanatory variable by trial and error while observing the partial regression coefficient or multiple correlation presented by the presentation unit 116 is used. May be.

  However, since a large number of applications are generally implemented in the securities business system 200, there is a possibility that an appropriate explanatory variable cannot be found even if trial and error are repeated based on the experience and intuition of the manager. is there. Therefore, the analysis apparatus 100 includes a variable selection support unit 170 that supports the selection of variables by the administrator.

  The variable selection support unit 170 includes a ranking generation unit 172 that generates a volume ranking in which applications are ranked according to the load on the securities business system 200, and an application that is positioned higher in the volume ranking generated by the ranking generation unit 172. A correlation presentation unit 174 that calculates and presents a correlation between the number of processing cases and the CPU usage rate of the main DB server 232.

  The ranking generation unit 172 calculates, for each application, the product of the total number of processing cases and the average response time within a predetermined calculation target period as a volume. The ranking generation unit 172 generates a volume ranking by ranking applications in ascending or descending order of the calculated volume. The response time is the time taken from when the application receives a request until it returns a response. The ranking generation unit 172 displays a volume ranking screen 320 indicating the generated volume ranking on the display 104.

  Note that the ranking generation unit 172 may refer to the history holding unit 112 when calculating the product. In addition, the ranking generation unit 172 acquires log data related to the process start time and process end time of each application request from the presentation layer 210 and the business logic layer 220, and the total number of processes and the average response time from the acquired log data. May be calculated.

  FIG. 8 is a representative screen diagram of the volume ranking screen 320. The volume ranking screen 320 displays application names, calculated volume values, and ranks in association with each other. The administrator refers to the displayed volume ranking screen 320 and selects an application whose volume is relatively higher. The administrator notifies the selected application to the correlation presentation unit 174 via the input device 102.

  Returning to FIG. 2, the correlation presentation unit 174 calculates the correlation between the notified number of processed applications and the CPU usage rate of the main DB server 232 with reference to the history holding unit 112. The correlation presentation unit 174 causes the display 104 to display a correlation display screen 322 showing the calculation result.

FIG. 9 is a representative screen diagram of the correlation display screen 322. The correlation display screen 322 shows a graph in which the horizontal axis indicates the CPU usage rate of the main DB server 232 and the vertical axis indicates the number of branch order inquiry processes. The correlation display screen 322 shows the value of the correlation coefficient. The administrator may determine an application having a relatively high correlation shown on the correlation display screen 322 as an application that uses the explanatory variable. As a result, among the applications positioned higher in the volume ranking, an application having a high correlation between the CPU usage rate of the main DB server 232 and the number of processing cases is selected as an explanatory variable. Thereby, the administrator can select an effective explanatory variable more easily and accurately.
In the variable selection support unit 170, automatic selection based on a predetermined threshold value may be used instead of selection by the administrator.

  In the above-described embodiment, examples of the holding unit are a hard disk and a memory. Based on the description in this specification, each unit is realized by a CPU (not shown), an installed application program module, a system program module, a memory that temporarily stores the contents of data read from the hard disk, and the like. It is understood by those skilled in the art who have touched this specification that they can do this.

The operation of the analyzer 100 having the above configuration will be described.
FIG. 10 is a flowchart showing a series of processes in the analysis apparatus 100. The history acquisition unit 110 acquires history data via the network 208 (S402) and registers it in the history holding unit 112. The multiple regression analysis unit 114 receives designation of an objective variable, an explanatory variable, and a data range to be used in the multiple regression analysis from the administrator (S404). The multiple regression analysis unit 114 extracts analysis target data from the history holding unit 112 (S406). The multiple regression analysis unit 114 derives a partial regression coefficient by performing multiple regression analysis on the analysis target data (S408). The presenting unit 116 presents the derived partial regression coefficient to the administrator (S410).

  According to the analysis apparatus 100 according to the present embodiment, it is possible to see which application can have a large influence on the main DB server 232 by comparing the derived partial regression coefficients between applications. The effect in this case is the effect of each process. For example, it can be determined that an application with a large partial regression coefficient may consume excessive processing time of the main DB server 232. Therefore, it is possible to more efficiently suppress the CPU usage rate of the main DB server 232 by individually adjusting such applications. Further, by limiting the number of such applications activated during request concentration or resetting the queue depth, an overload of the main DB server 232 can be avoided more reliably.

  As described above, at present, the securities business system is becoming more complicated both in terms of server configuration and application configuration. From the experience as a person skilled in the art, the present inventor is suitable for multivariate analysis using the degree of application use as a variable in order to predict the degree of resource use in such a complicated securities business system. I came up with it. Therefore, the analysis apparatus 100 according to the present embodiment predicts the CPU usage rate of the main DB server 232 using a multiple regression equation based on the derived partial regression coefficient. Thereby, even in a complicated securities business system, the degree of use of the main DB server 232 can be predicted more accurately.

  In the present embodiment, the processing number prediction unit 138 predicts the processing number based on the prediction model generated by the prediction model generation unit 136. Thereby, even when the predicted value of the number of processing cases cannot be obtained from the outside, the predicted value can be internally generated from the past results.

  The configuration and operation of the analyzer 100 according to the embodiment have been described above. This embodiment is an exemplification, and it is understood by those skilled in the art that various modifications can be made to each component and combination of processes, and such modifications are within the scope of the present invention.

  In the embodiment, the case where the analysis apparatus 100 analyzes the securities business system 200 has been described. However, the present invention is not limited to this. For example, the analysis apparatus may analyze another financial product business system such as a bank business system. It should be noted that the securities business system 200 has a large fluctuation in the degree of use compared to other financial product business systems due to its nature. Therefore, the merit of tuning the application of the securities business system 200 using the analysis apparatus 100 and accurately predicting the degree of use is relatively great.

  DESCRIPTION OF SYMBOLS 100 Analysis apparatus, 110 History acquisition part, 114 Multiple regression analysis part, 120 Prediction part, 136 Prediction model production | generation part, 142 Application specification part, 144 Monitoring part, 146 Judgment part, 200 Securities business system.

Claims (9)

A history acquisition unit that acquires the degree of use of a plurality of applications that realize a plurality of functions of the financial product business system and the degree of use of resources used by the plurality of applications as history data;
The partial regression coefficient is obtained by performing multiple regression analysis on the history data acquired by the history acquisition unit with each of the usage levels of a plurality of applications as explanatory variables and the usage level of resources as an objective variable. A multiple regression analysis unit to derive,
An application specifying unit for specifying an application having a partial regression coefficient derived by the multiple regression analysis unit greater than a threshold;
A monitoring unit for monitoring the degree of use of the application specified by the application specifying unit;
An analysis device comprising: a determination unit that determines whether or not use of the application specified by the application specifying unit should be restricted based on a monitoring result in the monitoring unit. A history acquisition unit that acquires the degree of use of a plurality of applications that realize a plurality of functions of the financial product business system and the degree of use of resources used by the plurality of applications as history data;
The partial regression coefficient is obtained by performing multiple regression analysis on the history data acquired by the history acquisition unit with each of the usage levels of a plurality of applications as explanatory variables and the usage level of resources as an objective variable. A multiple regression analysis unit to derive,
A ranking generation unit that generates a ranking in which applications are ranked according to the load on the financial product business system;
A correlation presentation unit that calculates a correlation between the degree of use of the application extracted based on the ranking generated by the ranking generation unit and the degree of resource use and presents the correlation to the user. apparatus.   A resource prediction unit that predicts the degree of resource use by substituting predicted values of the degree of use of a plurality of applications into a multiple regression equation based on the partial regression coefficient derived by the multiple regression analysis unit; The analyzer according to claim 1 or 2. A prediction model generation unit that generates a prediction model based on data on the degree of use of the application acquired by the history acquisition unit;
An application prediction unit that predicts the degree of use of the application using the prediction model generated by the prediction model generation unit;
The analysis apparatus according to claim 3, wherein the resource prediction unit predicts a resource usage level by substituting the usage level predicted by the application prediction unit into the multiple regression equation.   5. The analysis apparatus according to claim 1, wherein the financial product business system is a securities business system. A method executed by an analysis apparatus including a history acquisition unit, a multiple regression analysis unit, an application identification unit, a monitoring unit, and a determination unit,
The history acquisition unit associates and acquires the degree of use of a plurality of applications that realize a plurality of functions of the financial product business system and the degree of use of resources used by the plurality of applications as history data. When,
The multiple regression analysis unit performs multiple regression analysis on the acquired history data by using multiple regression analysis with each of the usage levels of a plurality of applications as explanatory variables and the usage level of resources as an objective variable. Deriving coefficients,
The application identifying unit identifying an application for which the derived partial regression coefficient is greater than a threshold;
The monitoring unit monitoring the degree of use of the identified application;
And a step of determining whether the use of the identified application should be restricted based on a monitoring result. A function of associating a degree of use of a plurality of applications that realize a plurality of functions of the financial product business system with a degree of use of resources used by the plurality of applications as history data;
A function for deriving a partial regression coefficient by performing multiple regression analysis using the degree of use of multiple applications as an explanatory variable and the degree of resource use as an objective variable for the acquired history data,
The ability to identify applications whose derived partial regression coefficients are greater than a threshold;
The ability to monitor the usage of identified applications;
A computer program for causing a computer to realize a function of determining whether or not to restrict use of a specified application based on a monitoring result. A method executed by an analysis apparatus comprising a history acquisition unit, a multiple regression analysis unit, a ranking generation unit, and a correlation presentation unit,
The history acquisition unit associates and acquires the degree of use of a plurality of applications that realize a plurality of functions of the financial product business system and the degree of use of resources used by the plurality of applications as history data. When,
The multiple regression analysis unit performs multiple regression analysis on the acquired history data by using multiple regression analysis with each of the usage levels of a plurality of applications as explanatory variables and the usage level of resources as an objective variable. Deriving coefficients,
The ranking generation unit generates a ranking in which applications are ranked according to the amount of load on the financial product business system;
The correlation presenting unit comprises a step of calculating a correlation between the degree of use of the application extracted based on the generated ranking and the degree of use of the resource and presenting it to the user. . A function of associating a degree of use of a plurality of applications that realize a plurality of functions of the financial product business system with a degree of use of resources used by the plurality of applications as history data;
A function for deriving a partial regression coefficient by performing multiple regression analysis using the degree of use of multiple applications as an explanatory variable and the degree of resource use as an objective variable for the acquired history data,
A function that generates a ranking of applications ranked by the amount of load on the financial product business system,
A computer program for causing a computer to realize a function of calculating a correlation between a degree of use of an application extracted based on a generated ranking and a degree of use of a resource and presenting it to a user.

Images