JP2019091197A - Information processing apparatus and method - Google Patents

Abstract

To propose an information processing apparatus and method capable of reliably supporting subsequent operations performed based on a processing result of a batch.SOLUTION: An information processing apparatus and method for executing each registered batch and storing a processing result of the executed batch in a corresponding data mart are configured to: set the degree of importance for each record of each data mart; calculate the degree of influence that the change in the data has on subsequent operations when the batch detects a change in data referenced during the processing; determine for each record of the data mart whether or not the degree of influence falls within an allowable range on the basis of the degree of importance set for each record in the data mart for storing the processing result of the batch referring to the changed data and the calculated degree of influence; and reflect the change in the data on the data mart when the degree of influence of any record exceeds the allowable range.SELECTED DRAWING: Figure 30

Description

  The present invention relates to an information processing apparatus and method, and is suitably applied to, for example, a store management system.

  In a company that develops a plurality of stores, such as a convenience store, sales data representing sales results of the stores are transmitted daily from the store terminal installed in each store to the headquarters system installed in the headquarters. Then, the headquarters system side generates periodic data such as daily sales and sales results for each store and product by executing periodical batch processing that refers to (inputs) these sales data. These generated aggregated data are stored and managed in different data marts. The tabulated data stored in these data marts generated in this manner is thereafter utilized for sales management in the accounting department of the headquarters, sales planning in the planning department, and the like.

JP, 2008-97476, A

  By the way, in the head office system, consider a case where sales data or data mart data (hereinafter referred to as original data) referred to by the batch is changed, added or deleted after processing of a certain batch is started. Such a situation occurs, for example, when a store representative corrects an input error after the start of batch processing, or when there is a return after the start of batch processing.

  In this case, the headquarters system side needs to re-execute the batch referring to the original data that has been changed etc., but the completion time of the batch is delayed by the re-execution, so the subsequent process using the processing result of the batch In some cases, it is not possible to complete the batch by the start time of the work (the work in the accounting department, the planning department or the like, which will be referred to as a subsequent work hereinafter).

  In this regard, for example, in Patent Document 1, even if batch target data is not determined, the batch processing end time can be achieved by batching the batch target data when batch target data is generated. An invention is disclosed that can accelerate the

  However, even if the original data is changed or the like, if the change or the like has little influence on the subsequent business, there is no need to re-execute a batch for processing with reference to the original data. Therefore, when the original data is changed or the like, the batch may be re-executed only when the change or the like has a large influence on the subsequent business. However, it is difficult to determine the magnitude of the impact of changes in the original data on subsequent operations, and it is not easy to determine whether it is necessary to re-execute the corresponding batch when there is a change in the original data, etc. There was a problem.

  Therefore, when there is a change or the like in the original data, the head office system side determines the influence of the change or the like on the subsequent business, and determines the necessity of re-execution of the corresponding batch based on the determination result, If it is possible to re-execute only the batch that is really necessary, it is possible to minimize the occurrence of a situation where batch processing is not completed by the start time of the subsequent business due to the re-execution of the batch, and reliable execution of the subsequent business It is thought that it can support. Further, it is considered that such a method makes it unnecessary to check the influence of the change of the original data on the subsequent work, and the burden on the system administrator and the subsequent work person can be alleviated.

  The present invention has been made in consideration of the above points, and an object of the present invention is to propose an information processing apparatus and method capable of reliably supporting subsequent work performed based on the batch processing result.

  In the present invention, in order to solve the problem, in the information processing apparatus, each registered batch is executed, and the processing result of the executed batch is stored in the corresponding data mart, for each record of each data mart. The importance setting unit that sets the importance, the change detection unit that monitors data that each batch refers to during processing, and detects the change when any of the data is changed, and the change When the detection unit detects a change in any of the data, an influence calculation unit that calculates the degree of influence that the change in data has on subsequent operations, and the processing result of the batch that refers to the changed data is The importance degree set for each record of the data mart to be stored, and the influence degree calculated by the influence degree calculation unit Therefore, a batch control unit for reflecting the change of the data to the data mart is provided as necessary, and the allowable range of the influence degree is set in advance for each importance degree, and the batch control unit The degree of influence is based on the degree of importance set in each record of the data mart in which the processing result of the batch referring to is stored, and the degree of influence calculated by the influence degree calculation unit. It is determined for each record of the data mart whether or not it is within the allowable range, and when the degree of influence of any of the records exceeds the allowable range, the change of the data is determined as the data mart To reflect it.

  Further, in the present invention, in the information processing method executed by an information processing apparatus that executes each registered batch and stores the processing result of the executed batch in the corresponding data mart, each record of each data mart The first step of setting the degree of importance for each and the data that each batch refers to at the time of processing is monitored, and when a change in any of the data is detected, the change of the data is a subsequent work The second step of calculating the degree of influence, and the degree of importance calculated for each of the records set in each record of the data mart in which the processing result of the batch referring to the changed data is stored And a third step of reflecting the change of the data in the data mart, as necessary, The allowable range of the influence degree is set in advance for each of the above, and in the third step, the information processing apparatus stores the processing result of the batch referring to the data in each record of the data mart. It is determined for each of the records of the data mart whether or not the degree of influence is within the allowable range based on the set degree of importance and the calculated degree of influence, and any one of the records When the degree of influence exceeds the allowable range, the change of the data is reflected in the data mart.

  According to the information processing apparatus and method, when the data to which the batch refers is updated, the change of the data is reflected in the corresponding data mart only when the change has a large influence on the subsequent business. If the change has a small impact on follow-on operations, the change in the data is not reflected in the data mart. Therefore, only when it is really necessary, such data changes will be reflected in the corresponding data mart, so the process of reflecting such data changes in the corresponding data mart will not be in time for the start of the subsequent business Occurrence of a situation can be suppressed as much as possible.

  According to the present invention, it is possible to realize an information processing apparatus and method that can reliably support subsequent work to be performed based on batch processing results.

It is a block diagram showing the whole composition of the store management system by this embodiment. FIG. 6 is a block diagram for explaining the flow of work in the store management system. (A) is a configuration of sales data, (B) is a daily sales mart, (C) is a sales performance mart, (D) is a popular commodity mart, and (E) is a sales order mart. (A)-(C) are charts used for description of an influence degree calculation rule. (A)-(C) are charts used for description of an influence degree calculation rule. (A)-(C) are charts used for description of an influence degree calculation rule. It is a block diagram showing the logical composition of an application server. It is a block diagram showing the logical composition of an application server. It is a chart showing an example of composition of an influence degree calculation rule decision table. It is a chart showing an example of composition of an importance setting rule determination table. It is a chart showing an example of composition of a person-in-charge contact table. It is a chart showing an example of composition of an input-and-output management table. It is a chart showing an example of composition of a batch execution condition management table. It is a chart showing an example of composition of a batch execution schedule table. It is a chart showing an example of composition of a batch priority management table. It is a chart showing an example of composition of a resource management table. It is a chart showing an example of composition of a batch execution history management table. It is a chart showing an example of composition of a mart reference history management table. It is a chart showing an example of composition of a reference frequency management table. It is a chart showing an example of composition of a degree of importance management table. It is a figure which shows the structural example of a batch registration screen. It is a figure which shows the structural example of an importance setting rule determination screen. It is a figure which shows the structural example of a person-in-charge contact registration screen. It is a flowchart which shows the process sequence of a 1st new batch registration process. It is a flowchart which shows the process sequence of a 2nd new batch registration process. It is a flowchart which shows the process sequence of normal operation processing. It is a flowchart which shows the process sequence of importance level management table update process. It is a chart showing an example of composition of a table for calculation. It is a flowchart which shows the process sequence of a change detection process. It is a flow chart which shows a processing procedure of batch re-execution processing. It is a flowchart which shows the process sequence of an idle time search process. It is a flowchart which shows the process sequence of influence degree calculation processing.

  An embodiment of the present invention will now be described in detail with reference to the drawings.

(1) Configuration of Store Management System According to the Present Embodiment In FIG. 1, reference numeral 1 generally denotes a store management system according to the present embodiment. The store management system 1 is a system used to manage sales and the like of each store 2 in a company operating a plurality of stores 2 such as a convenience store and a family restaurant.

  The store management system 1 includes a store terminal 3 installed in each store 2, and a head office system 8 including a web server 4 installed in the head office, an application server 5, a database server 6, and a plurality of host devices 7. Is configured.

  The store terminal 3 is a computer device such as a POS (Point Of Sales) terminal used to manage the sales situation in the store 2 and includes a central processing unit (CPU) 10, a memory 11, a storage device 12, and a communication device. It comprises 13 and so on.

  The CPU 10 is a processor that controls the operation of the entire shop terminal 3. The memory 11 is also used as a work memory for the CPU 10 in addition to being used to temporarily hold necessary programs. The CPU 10 executes the program stored in the memory 11 to perform various processes of the entire shop terminal 3.

  The storage device 12 is composed of a large-capacity non-volatile storage device such as a hard disk drive or a solid state drive (SSD), for example, and is used to hold necessary information such as programs and daily sales data 14 for a long time. Ru. Furthermore, the communication device 13 is configured by a NIC (Network Interface Card) or the like, and performs protocol control at the time of communication with the head office system 8.

  The web server 4 is connected to the store terminal 3 of each store 2 via the network 9A including the Internet and the like, and the sales data 14 representing daily sales results transmitted daily from the store terminal 3 is stored in the application server 5 It is a general purpose server device having a function of transferring to The web server 4 is configured to include information processing resources such as a CPU 20, a memory 21, a storage device 22, and a communication device 23. The CPU 20, the memory 21, the storage device 22, and the communication device 23 have almost the same functions and configurations as the CPU 10, the memory 11, the storage device 12, and the communication device 13 of the shop terminal 3, and therefore I omit it.

  The application server 5 is a server device having a function of executing various processes in accordance with the installed application program 34 (FIG. 7). The application server 5 is configured of a general-purpose server device including information processing resources such as a CPU 30, a memory 31, a storage device 32, and a communication device 33. The CPU 30, the memory 31, the storage device 32, and the communication device 33 have substantially the same functions and configurations as the CPU 20, the memory 21, the storage device 22, and the communication device 23 of the web server 4, and thus the description thereof is omitted. Do.

  The memory 31 of the application server 5 stores, in addition to the application program 34 (FIG. 7), a plurality of batches 35 (FIG. 7) respectively registered by persons in charge of each division of the head office. The application program 34 stores the sales data 14 transferred from each shop terminal 3 via the web server 4 in the database 40 held by the database server 6 or stored in the data mart 41 held by the database server 6 It is a program having a function of reading data in response to a request from the host device 7 and transferring the data to the host device 7. The batch 35 is a program that inputs one or more data and executes processing, and updates one or more data marts 41 based on the processing result.

  The database server 6 is a server device having a function of managing and maintaining the database 40. The database server 6 is configured of a general-purpose server device provided with information processing resources such as a CPU, a memory, a storage device, and a communication device (all not shown). The CPU, the memory, the storage device and the communication device have almost the same functions and configurations as the CPU 20, the memory 21, the storage device 22 and the communication device 23 of the web server 4, and thus the description thereof is omitted here. The storage device of the database server 6 stores a database 40 in which sales data 14 from each store 2 is registered, and a plurality of data marts 41 in which the processing results of each batch 35 (FIG. 7) are respectively registered. Ru.

  The host device 7 is a computer device lent to each person in charge of each department such as the accounting department, the planning department and the general affairs department in the head office, and each application server 5 is connected via a network 9B such as an in-house LAN (Local Area Network). And connected. The person in charge of each department accesses the application server 5 by using the host device 7 of its own, registers the batch 35 (FIG. 7) for executing predetermined processing in the application server 5, or makes a desired data mart 41 Data is acquired from the database server 6 via the application server 5.

(2) Batch Management Function According to this Embodiment Next, the batch management function installed in the application server 5 according to this embodiment will be described. On this occasion, first, the flow of operations in the store management system 1 will be described with reference to FIG.

  In the present store management system 1, the person in charge 45 of each store 2 sequentially inputs information on sales (sales information) to the store terminal 3. Then, the shop terminal 3 generates, for example, sales data 14 as shown in FIG. 3A daily based on the input sales information, and the generated sales data 14 is stored in the storage device 12 in the shop terminal 3 (FIG. Store in). The store terminal 3 also transmits the sales data 14 stored in the storage device 12 to the head office system 8 via the network 9A on a regular basis every day.

  On the head office system 8 side, the sales data 14 transmitted from each shop terminal 3 is given to the application server 5 via the web server 4 (FIG. 1). Then, the application server 5 stores and manages the sales data 14 in the database 40 (FIG. 1) held by the database server 6 (FIG. 1). In addition, the application server 5 creates a schedule (hereinafter referred to as a batch execution schedule) which defines the execution timing of each batch 35 (FIG. 7) registered in advance by persons in charge 46A to 46C of each division of the head office respectively. And each batch 35 is executed according to the batch execution schedule.

  Here, in the application server 5, the first tabulated batch 35A ("tabulated batch 1") created by the person in charge 46A of the accounting department and the second and third tabulations created by the person in charge 46B of the planning department The batches 35B and 35C ("totaling batch 2" and "totaling batch 3") and the fourth totaling batch 35D ("totaling batch 4") created by the person in charge 46C of the general affairs department are registered. Do.

  The first tabulated batch 35A is a batch for tabulating the daily sales of each store 2. The daily tabulated sales for each store 2 are calculated by this first tabulated batch 35A, and the calculation result is stored in the database server 6 For example, it is stored in the data mart 41 (daily sales mart 41A) configured as shown in FIG. 3 (B). Thus, the person in charge 46A of the accounting department acquires daily sales data of each store stored in the daily sales mart 41A using its own host device 7 (FIG. 1), and based on the acquired sales data Work to manage daily sales of each store 2.

  Further, the second tabulation batch 35B is a batch for tabulating the sales results for each product, and the number of sales per product per day is calculated by this second tabulation batch 35B, and the calculation result is stored in the database server 6 For example, it is stored in the data mart 41 (sales performance mart 41 B) as shown in FIG. Thus, the person in charge 46B of the planning department acquires the number of sales for each product stored in the sales performance mart 41B using the host device 7 of its own, and based on the acquired number of sales for each product, for example, each store 2 The management of whether or not the sales quota of the sales reinforcement designated product has been achieved, and the operation of adjusting the shipment quantity of each product to each store 2 are performed.

  Furthermore, the third tabulation batch 35C is a batch for identifying the popular products of each store 2. By this third tabulation batch 35C, the product with the largest number of sales in each shop 2 is extracted, and the extraction result is stored in the database server 6 Are stored in, for example, the data mart 41 (popular product mart 41C) as shown in FIG. Thus, the person in charge 46B of the planning department uses the host device 7 of its own to acquire the product with the largest number of sales for each store stored in the popular product mart 41C, and plans a sales campaign based on the acquired information. Perform tasks such as

  Furthermore, the fourth tabulation batch 35D is a batch for calculating the sales ranking of the shop 2, the shop 2 is sorted in descending order of sales by this fourth tabulation batch, and the sort result is held by the database server 6, for example 3 (E) are stored in the data mart 41 (sales order mart 41D). Thus, the person in charge 46C of the general affairs department acquires the sales ranking information of each store 2 stored in the store sales order mart 41D using the host device 7 of its own, and based on the acquired information, the top sales are high Do business such as awarding several stores.

  By the way, when the sales data 14 to which the first to fourth totalized batches 35A to 35D refer is changed after the processing of the first to fourth totalized batches 35A to 35D is started, the first to fourth always Assuming that the aggregation batches 35A to 35D of No. 4 are to be re-executed, the completion of the first to fourth aggregation batches 35A to 35D is in time for the start time of the subsequent operation (operation such as the administration department, the planning department or the general affairs department) It may disappear.

  In addition, the second batch uses the processing result of the batch to be re-executed (referred to as the first batch), the third batch uses the processing result of the second batch, and so on. If a setting is made such that each batch is executed in a row (hereinafter, each batch after the second batch is appropriately referred to as a subsequent batch of the first batch), the subsequent batch 35 is identified, or It is also difficult to determine which subsequent batch 35 to rerun.

  Therefore, in the present store management system 1, when the sales data 14 referred to by the batch 35 (FIG. 7) and the data (original data) of the data mart 41 referred to by the batch 35 are changed, the subsequent batch of the batch 35 35, identify the degree of influence of the change of the original data on the subsequent work for these batch 35 and the subsequent batch 35, and evaluate the magnitude of the calculated degree of influence, and based on the evaluation result, the batch 35 or The application server 5 is equipped with a batch management function that determines whether or not to execute the subsequent batch 35 again.

  In practice, the application server 5 refers to data which each batch 35 registered in itself refers (inputs) and a data mart 41 which stores (outputs) processing results in an input / output management table 63 described later with reference to FIG. Registered and managed. The application server 5 also monitors sales data 14 from each store 2 stored in the database 40 (FIG. 1) held by the database server 6 and data of each record of each data mart 41 (FIG. 1). ing.

  Then, when one of the sales data 14 and the data of each data mart 41 is changed, the application server 5 refers to the changed data (the data is used as the original data). All batches 35 and their subsequent batches 35 are identified with reference to the input / output management table 63. Further, the application server 5 determines whether there is free time for re-executing each specified batch 35 and its subsequent batch 35 with reference to the batch execution schedule. Hereinafter, this determination will be referred to as schedule vacancy determination.

  Further, in parallel with the schedule availability judgment, the application server 5 executes the processing with reference to the original data and determines whether the degree of influence of the change of the original data on the subsequent work is within an allowable range. A determination is made on each of the records of the data mart 41 in which the processing result of 35 is stored (hereinafter, this determination is referred to as “impact determination”).

  As means for performing such impact determination, a plurality of types of impact calculation rules are registered in advance in the application server 5 according to the processing content of each batch 35. Then, when performing the influence degree determination, first, the application server 5 calculates the influence degree of the change of the original data on the subsequent work according to the influence calculation rule associated with the batch 35 referring to the original data.

  For example, the batch 35 executed with such original data as an input is the above-described first aggregation batch 35A or second aggregation batch 35B, and aggregation is performed as in the first aggregation batch 35A or second aggregation batch 35B. The influence calculation rule (hereinafter referred to as a first influence calculation rule) prepared in advance for the batch 35 including the process is referred to as “the processing result of the batch executed with reference to the changed original data. Of the data of the data mart storing the data, the absolute value of the difference between the value after the change of the data to be changed with the change of the original data and the value before the change of the data before the change of the data It is assumed that such a degree of influence is calculated by dividing by a value (| value before change−value after change | value before change).

  Also, as shown in FIG. 4A, the sales volume of the product name "chocolate" sold to "2017/2/1 11:00" in the sales data 14 of the store 2 "store 1" is "2" It is assumed that the unit price of the product name "beer" sold to "3" and "2017/2/2 12:00" is changed from "200" to "300".

In this case, the contents of the corresponding data mart 41 (daily sales mart 41A) are "380,000 yen" for the sales of "2/1" and "390,000 for the sales of" 2/2 "as shown in FIG. 4 (B). As shown in Fig. 4 (C), the sales of "2/1" will be changed to "380,800 yen" and the sales of "2/2" will be changed to "390,100 yen" as shown in Fig. 4C. The degree of influence that the change of the sales data 14 (original data) of “2/1” has on subsequent work is
It is calculated as “0.002” as in the above, and the degree of influence that such “2/2” sales data changes have on subsequent operations is
Is calculated as "0.0002".

  In addition, the batch 35 which refers to the original data is the above-mentioned third aggregation batch 35C, and the influence calculation rule prepared in advance for the batch 35 which calculates the maximum value like this third aggregation batch 35C. (Hereafter, this is referred to as a second influence calculation rule) “If the original data is changed, the maximum value exceeding the maximum value before the change of the original data is calculated, the first influence The influence degree is calculated by the same calculation method as the calculation rule, and in other cases, the influence degree is always set to “0”. Further, as shown in FIG. 5A, the sales volume of the product name "chocolate" sold to "2017/2/11 11:00" in the sales data of the store 2 "store 1" is from "2" to "2 It shall be changed to 5 ".

In this case, the content of the corresponding data mart 41 (popular product mart 41C) is changed from the state of FIG. 5 (B) to FIG. 5 (C). Therefore, the degree of influence of the change in sales data 14 (original data) on subsequent work in this case is
Is calculated as "0.00009".

  Furthermore, the batch 35 which refers to the original data is the fourth totalization batch 35D described above, and the influence level calculation rule prepared in advance for the batch 35 including the rearrangement process as in the fourth totalization batch 35D Hereinafter, this will be referred to as the third influence calculation rule) “The difference between the rank n before the change of the record of the data mart whose value is changed after the change of the original data and the rank m after the change of the record This influence degree is calculated by dividing the absolute value of by the number of rows S of the entire data mart after the change (| n−m | / S) to calculate the degree of influence.

  Further, as shown in FIG. 6 (A), the sales volume of the product name "chocolate" sold to "2017/2/1 11:00" in the sales data 14 of "store 1" is "2" to "3". In accordance with this, the content of the corresponding data mart 41 (store sales order mart 41D) is to be changed from the state of FIG. 6 (B) to FIG. 6 (C).

In this case, the rank n of the record of the data mart 41 whose value is changed after the change of the original data is "3", the rank m of the record after the change is "2", and the entire row of the data mart 41 is Since the number S is "100", the degree of influence of the change of the original data on the subsequent work is
Is calculated as "0.01".

  On the other hand, the determination as to whether or not the change in the original data has an influence on the subsequent operation “the degree of influence is within the allowable range” depends on each record of the data mart 41 storing the processing result of the batch referring to the original data. It is performed according to the importance of the data set for each data. In the case of the present embodiment, a plurality of rules according to the processing content of various batches are set as rules for setting the degree of importance of the data of each record of the data mart 41 in this manner (hereinafter referred to as the importance degree setting rule). Several types of importance setting rules are registered in advance in the application server 5.

  Then, the persons 46A to 46C in charge of the follow-on operations trying to register a new batch 35 (first to fourth totalized batches 35A to 35D) in the application server 5 are the data mart 41 in which the processing result of the batch 35 is stored. The importance setting rule used to set the importance of the data of each record (41A to 41D) is determined and set when the batch 35 is registered in the application server 5.

  For example, the importance level setting rule set for the first aggregation batch 35A and the second aggregation batch 35B is a rule that "the demand degree of data of frequently referenced records is set high" (hereinafter referred to as It is assumed that it is called "importance setting rule of 1."

  In this case, the application server 5 stores the daily sales mart 41A (FIG. 3 (B)) and the sales result mart 41B (FIG. 3 (FIG. 3 (FIG. 3)) in which the processing results of the first tabulation batch 35A and the second tabulation batch 35B are stored. As for C), the importance of the data of the top 20% records with many references is “A”, and the importance of the data of the top 20-40% of records is “B” The importance of the data in each of the top 40 to 60% records is "C", the importance of the data in each of the top 60 to 100% records is "D", and the degree of importance is "A" The importance of the data of each record which does not correspond to any of "D" (that is, never referred to) is set as "E", and the importance of each record (each data) is respectively set.

  Also, for example, the importance level setting rule set for the third aggregation batch 35C and the fourth aggregation batch 35D is a rule that “importance level of data of records having different numerical tendencies with other records is set high”. (Hereinafter, this will be referred to as a second importance level setting rule).

  In this case, the application server 5 stores the popular commodity mart 41C (FIG. 3D) and the sales order mart 41D (FIG. 3E (E), in which the processing results of the third totalization batch 35C and the fourth totalization batch 35D are stored. )) For each record of the data mart 41, the absolute value of the difference between the numerical value of the record and the average value of the values of each record of the data mart 41 is determined as the deviation of the record, and the deviation is large Importance of data of top 20% of each record is "A", importance of data of top 20 to 40% of each record is "B" and importance of data of top 40 to 60% of each record The degree of importance of the data of each of the top 60 to 100% records of “C” and “D” respectively, and the degree of importance of each record (each data) are respectively set.

  Then, for example, with respect to the data of the record having the importance “A”, the application server 5 determines that “the influence is within the allowable range” when the influence is “0” or less, and the importance “B”. With regard to the data of the record of “1”, when the degree of influence is “0.5” or less, it is determined that “the degree of influence is within the allowable range”. Further, the application server 5 determines that the “impact is within the allowable range” when the impact is “1” or less for the data of the record having the importance “C”, and the importance is “D”. For the data of the record, if the degree of influence is "10" or less, it is determined that "the degree of influence is within the allowable range". Furthermore, the application server 5 always determines that the “impact is within the allowable range” regardless of the value of the impact (unlimited) for the data of the record that does not correspond to the importance “A” to “D”. .

  Therefore, as described above for the first aggregation batch 35A and the second aggregation batch 35B, when the importance of the data of the record whose influence degree is calculated as "0.002" is "A", the original data It is determined that the impact of the change on subsequent work exceeds the allowable range, and if the importance of the data is “B” to “E”, the impact on the subsequent work of the change of the original data is It will be determined that it is within the allowable range.

  In the case of the present embodiment, as described above, the importance of the data of each record of the data mart 41 relates to the data which changes with time, such as the number of times of reference to the data of the record and the magnitude of deviation from the average value. Since it is set according to a predetermined value, it is necessary to set the importance of each record (each data) based on the latest information as much as possible. Therefore, the application server 5 periodically (for example, every day) updates the importance of each record (data).

  Then, based on the determination result of the above-mentioned schedule vacancy determination and the above determination result of the impact determination, the application server 5 determines, for example, whether or not the influence of the change of the original data on the subsequent operation is within the allowable range. Is determined for each record of the data mart 41 (that is, the data mart 41 in which the processing result of the batch 35 referring to the original data is registered) which is affected by the change of the original data. Then, when there is a record in which the degree of influence exceeds the allowable range, the application server 5 determines that the batch 35 storing the processing result in the data mart 41 should be re-executed.

  Further, even if there is no record whose degree of influence exceeds the allowable range due to the change of the original data, the application server 5 has a vacant time when the data mart 41 can execute the batch 35 storing the processing result again. Is determined to be re-executed, and it is determined that the batch 35 should not be re-executed if there is no such free time.

  When the application server 5 determines that the batch 35 should be re-executed, the application server 5 changes the batch execution schedule to re-execute the batch 35, and re-executes the batch 35 according to the changed batch execution schedule.

  As means for realizing the batch re-execution necessity determination function according to the present embodiment as described above, as shown in FIG. 7, the memory 31 of the application server 5 includes a batch analysis unit 50 and a batch schedule management unit 51. A history management unit 52, an importance setting unit 53, a change detection unit 54, an influence calculation unit 55, a batch control unit 56, and a reflection notification unit 57 are stored. The storage unit 32 of the application server 5 also includes an impact degree calculation rule determination table 60, an importance level setting rule determination table 61, a person-in-charge contact table 62, an input / output management table 63, a batch execution condition management table 64, and a batch execution schedule. Table 65, Batch priority management table 66, Resource management table 67, Batch execution history management table 68, Mart reference history management table 69, Reference frequency management table 70, Importance management table 71, and a plurality of impact calculation rules 72 respectively And an importance setting rule 73 is stored.

  Then, as shown in FIG. 8, the batch analysis unit 50 manages the influence degree calculation rule determination table 60, the importance degree setting rule determination table 61, the person-in-charge contact table 62 and the input / output management table 63, and the batch schedule management unit 51 manages a batch execution condition management table 64, a batch execution schedule table 65, a batch priority management table 66, and a resource management table 67. Further, the history management unit 52 manages the batch execution history management table 68 and the mart reference history management table 69, and the importance setting unit 53 registers in advance by the reference frequency management table 70, the importance management table 71, the system administrator, etc. The several importance setting rules 73 are managed, and the influence calculation unit 55 manages several influence calculation rules 72 registered in advance by a system administrator or the like.

  The batch analysis unit 50 is a program having a function of inputting an execution condition of the batch 35 when a person in charge of the subsequent business registers the batch 35 in the application server 5. In practice, when the person in charge of the subsequent business registers the batch 35 in the application server 5, the batch analysis unit 50 causes the host device 7 of the person in charge to display a batch registration screen 80 described later with reference to FIG.

  The batch analysis unit 50 then executes the execution frequency (daily, weekly, monthly, etc.) of the batch registered by the person in charge using the batch registration screen 80, “execute after midnight”, “after the batch A is completed. "Conditions to be met to start the batch 35 such as" Start "(hereinafter referred to as" start conditions ") and conditions such as the time when the batch 35 must be completed (hereinafter referred to as" completion conditions " Call), the required time required to start and complete the batch 35 (hereinafter referred to as the estimated required time), and the resources to be allocated for executing the batch 35. Acquired as an execution condition of the batch 35, and register the acquired execution condition in the batch execution condition management table 64.

  Further, when the person in charge of the subsequent business registers the batch 35 in the application server 5, the batch analysis unit 50 analyzes the processing content of the batch 35, and determines the above-described influence calculation rule 72 to be applied to the batch 35. Do. For example, when the batch process executed by the batch 35 includes tallying, the batch analysis unit 50 calculates the above-described first influence degree calculation rule, and the batch process executed by the batch 35 calculates the maximum value. When the process is included, the second impact degree calculation rule described above is applied, and when the batch process executed by the batch 35 further includes a sort process, the third impact degree calculation rule described above is applied to the batch 35 It is determined as the power of influence calculation rule 72. Then, the batch analysis unit 50 registers the determined impact degree calculation rule 72 in association with the batch 35 in the impact degree calculation rule determination table 60.

  Furthermore, when the person in charge of the subsequent work registers the batch 35 in the application server 5, the batch analysis unit 50 sets the importance level setting rule determination screen 90 described later with reference to FIG. 22 and the person in charge contact registration screen 100 described later with reference to FIG. And in order to be displayed on the host device 7 of the person in charge. Then, the batch analysis unit 50 registers the above-mentioned importance setting rule to be applied to the batch 35 set by the person in charge using the importance setting rule determination screen 90 in the importance setting rule determination table 61 and manages it. . Further, the batch analysis unit 50 registers the contact address of the person in charge registered by the person in charge using the person-in-charge contact registration screen 100 in the person-in-charge contact table 62 and manages it.

  In addition, each time the batch 35 is registered in the application server 5, the batch analysis unit 50 specifies the table (the database 40 or the data mart 41) to which the batch 35 refers and the data mart 41 storing the processing result. And registers and manages the specific result in the input / output management table 63. Then, for example, when the sales data 14 or the data mart 41 (original data) is changed, the batch analysis unit 50 refers to the input / output management table 63 in response to a request from the batch control unit 56 as described later. The records of the batch 35 and the data mart 41 that are affected by the change are identified.

  The batch schedule management unit 51 is a program having a function of creating and managing the above-mentioned batch execution schedule. In practice, when a new batch 35 is registered in the application server 5 by the person in charge of the subsequent work, the batch schedule management unit 51 executes the above-described execution condition of the batch 35 and the priority of each batch 35 described later with reference to FIG. The execution timing of the batch 35 is determined based on the order and the determined execution timing is registered in the batch execution schedule table 65 and managed. The batch schedule management unit 51 registers and manages, in the batch priority management table 66, the priority of each batch 35 registered in the application server 5 set by the system administrator or the like.

  Further, the batch schedule management unit 51 registers and manages the resources of the application server 5 in the resource management table 67, and as described later, can the batch control unit 56 designate the batch 35 and re-execute it? If an inquiry as to whether or not it has been issued is given, the resource management table 67 and the batch execution schedule table 65 are referenced to determine whether or not the batch 35 can be re-executed in terms of resources and time. The determination result is notified to the batch control unit 56. In this case, if the batch schedule management unit 51 determines that the batch 35 can be re-executed, the batch execution schedule is updated to incorporate the execution schedule for re-executing the batch 35 (that is, the execution is performed). The schedule is registered in the batch execution schedule table 65).

  The history management unit 52 executes the execution history of each batch 35 registered in the application server 5 or each record of the data mart 41 performed by the person in charge of the subsequent business using the host device 7 of the own via the application server 5. It is a program having a function of managing reference history. The history management unit 52 registers and manages the execution history of the batch 35 in the batch execution history management table 68 each time the batch 35 registered in the application server 5 is executed, and at the same time, the history management unit 52 Each time the record data is referred to by the person in charge of the subsequent work, the reference history is registered and managed in the mart reference history management table 69.

  The degree of importance setting unit 53 is a program having a function of setting the degree of importance of each record of the data mart 41 for each data mart 41 and updating the degree of importance periodically. In practice, the importance setting unit 53 refers to the mart reference history management table 69 managed by the history management unit 52, and registers and manages the reference frequency for each data in the reference frequency management table 70. Then, when a new batch 35 is registered, the importance setting unit 53 acquires the importance setting rule 73 (FIG. 7) set for the batch 35 from the importance setting rule determination table 61, and acquires the acquired importance The importance of each record (each data) of the data mart 41 in which the processing result of the batch 35 is registered is set with reference to the setting rule 73 and the reference frequency management table 70 described above. Further, the importance setting unit 53 periodically updates the importance of each record of the data mart 41 set as described above.

  The change detection unit 54 monitors each of the sales data 14 from each store terminal 3 and the data of each record of each data mart 41 stored in the database 40 (FIG. 1), and either the sales data 14 or the data mart This is a program having a function of detecting the 41 data when it is updated. When the change detection unit 54 detects a change in the sales data 14 or data in the data mart 41, the change detection unit 54 refers to the input / output management table 63 to refer to the batch 35 referring to the sales data 14 or the data of the record in the data mart 41. The batch control unit 56 is notified of the batch names of the batches 35 and the values before and after the change of the sales data 14 and the data of the data mart 41 records.

  The influence degree calculation unit 55 is a program having a function of calculating the degree of influence of the change on the subsequent business when the sales data 14 or the data of the data mart 41 is changed. The influence calculation unit 55 acquires the influence calculation rule 72 set for the batch 35 from the influence calculation rule determination table 60 for each batch 35 referring to the changed data (original data), and acquires the influence According to the degree calculation rule 72, the influence of the change of the original data on the subsequent work is calculated.

  The batch control unit 56 is a program having a function of starting and executing each batch 35 in accordance with a batch execution schedule managed by the batch schedule management unit 51 under normal conditions. When the change detection unit 54 detects a change in the sales data 14 or the data (original data) of the data mart 41, the batch control unit 56 batch-specifies the batch 35 to which the changed original data is input. The analysis unit 50 is requested.

  Then, the batch control unit 56 requests the batch schedule management unit 51 to determine whether there is an idle time for re-executing the batch 35 specified by the batch analysis unit 50 (schedule availability determination). Further, the batch control unit 56 requests the influence degree calculation unit 55 to calculate the degree of influence of the change of the original data on subsequent work, and based on the calculation result, the degree of influence of the change of the original data on subsequent work Is judged to be within the allowable range for each record of the data mart 41 affected by the original data (impact determination).

  The batch control unit 56 stores the processing result in the data mart 41 when there is a record in which the degree of influence exceeds the allowable range, based on the judgment result of the above-mentioned schedule vacancy judgment and the judgment result of the influence degree judgment. It is determined that the batch 35 to be executed should be re-executed, and the batch schedule management unit 51 and the reflection notification unit 57 are notified of re-execution of the batch 35.

  The reflection notifying unit 57 is a program having a function of notifying the person in charge of the subsequent business who has registered the batch of the re-execution of the batch 35. In fact, when the re-execution of the batch 35 is notified from the batch control unit 56, the reflection notification unit 57 acquires the contact information of the person who has registered the batch 35 from the person-in-charge contact table 62, and the batch Send a notification to that contact to re-execute 35.

  On the other hand, the influence degree calculation rule determination table 60 is a table used to manage the influence degree calculation rules 72 set by the batch analysis unit 50 for each batch 35 registered in the application server 5, as shown in FIG. As shown in FIG. 6, the target batch field 60A and the influence degree calculation rule field 60B are provided.

  The batch name of each batch 35 registered in the application server 5 is stored in the target batch field 60A, and the impact degree calculation rule 72 set for the corresponding batch 35 is stored in the impact degree calculation rule field 60B. The identifier of is stored.

  Therefore, in the case of the example of FIG. 9, for example, “impact calculation rule 1” is set as “impact calculation rule 72” for “total batch 1” and “total batch 2”, and “coarse batch 3” "Influence calculation rule 2" is set as the influence degree calculation rule 72, and "influence calculation rule 3" is set as the influence calculation rule 72 for "total batch 3". ing.

  Further, the importance level setting rule determination table 61 is a setting rule of the importance level for each record of the data mart 41 which registers the processing result of the batch 35 set when the person in charge of the subsequent business registers the batch 35 in the application server 5 It is a table used to manage (the importance setting rule 73). As shown in FIG. 10, the importance level setting rule determination table 61 is configured to include a target batch column 61A, an importance level setting rule column 61B, and a start condition column 61C.

  The batch name of each batch 35 registered in the application server 5 is stored in the target batch column 61A, and the importance level setting rule 73 set for the corresponding batch 35 is stored in the importance level setting rule column 61B. The identifier of is stored. Further, the condition for executing the process of updating the importance of each record of the data mart 41 in which the batch 35 stores the processing result, which is set when the corresponding batch 35 is registered in the start condition column 61C Condition) is stored.

  Therefore, in the case of the example of FIG. 10, for example, "importance setting rule 1" is set as the importance setting rule 73 for "aggregate batch 1", and the "aggregate batch 1" stores the processing result. It is shown that the importance set for the data of each record of the data mart 41 to be updated is set to be updated to “every day at midnight”.

  The person-in-charge contact table 62 is used to manage contacts (hereinafter referred to as e-mail addresses) to the person-in-charge set when the person-in-charge in the subsequent business registers the batch 35 in the application server 5. And, as shown in FIG. 11, is configured to include a target batch field 62A and an e-mail address field 62B.

  The batch name of each batch 35 registered in the application server 5 is stored in the target batch field 62A, and the e-mail address field 62B stores the e-mail address of the person who registered the corresponding batch 35 in the application server 5. Stored.

  Therefore, in the case of the example of FIG. 11, for example, the e-mail address of the person in charge who has registered "aggregate batch 1" in the application server 5 is "person" who registered "batch1_tantou@gyoumu.com" and "aggregate batch 2" in the application server 5 'S email address is "batch2_tantou@gyoumu.com", and the email address of the person in charge who has registered "aggregate batch 3" in application server 5 is "batch3_tantou@gyoumu.com", "aggregate batch 4" in application server 5 It is shown that the registered person's email address is "batch4_tantou@gyoumu.com".

  The input / output management table 63 includes a table (a table of the database 40 or the data mart 41) to be referred to when each batch 35 registered in the application server 5 executes a process, and a data mart 41 for storing the processing result. This table is used for management, and as shown in FIG. 12, is configured to include a batch name column 63A, a reference table column 63B, and an update table column 63C.

  Then, the batch name of each batch 35 registered in the application server 5 is stored in the batch name column 63A. The reference table column 63B stores the identifier of the table to be referred to when the corresponding batch 35 executes the processing, and the update table column 63C stores the identifier of the data mart 41 whose batch 35 registers the processing result. Stored.

  Therefore, in the case of the example of FIG. 12, for example, the batch 35 "total batch 1" executes processing with data stored in a table (table of the database 40) "store 1 sales data (copy)" as an input. It is shown that the processing result is stored in a data mart 41 called “daily sales mart”.

  The batch execution condition management table 64 is a table used to manage the execution conditions of the batch 35 registered by the person in charge of the subsequent business when registering the batch 35 in the application server 5 as described above. As shown in FIG. 13, the batch execution condition management table 64 comprises a batch name column 64A, an execution frequency column 64B, a start condition column 64C, a completion condition column 64D, an estimated required time column 64E, and an allocated resource column 64F. Be done.

  The batch name field 64A stores the name (batch name) of the batch 35 set by the person in charge of registering the corresponding batch 35 when the corresponding batch 35 is registered, and the execution frequency field 64B stores the name. The execution frequency of the batch 35 set by the person in charge is stored. The start condition column 64C stores the start condition of the batch 35 set by the person in charge, and the completion condition column 64D stores the completion condition of the batch 35 set by the person in charge.

  Furthermore, the estimated required time of the batch 35 registered by the person in charge is stored in the estimated required time column 64E, and allocated in the allocated resource column 64F when the batch 35 registered by the person in charge is executed. The resources to be stored are stored.

  Therefore, in the case of the example of FIG. 13, for the batch 35 “total batch 1”, the estimated required time is “3 hours”, and processing is started “daily”, “after midnight”, “next morning It is shown that it is necessary to complete the process by 9 o'clock, and that at the time of process execution, it is registered that one CPU and 2 GB worth of memory (“CPU = 1, memory = 2 GB”) should be allocated. There is.

  Also, the batch execution schedule table 65 is a table created by the batch schedule management unit 51 (FIGS. 7 and 8) for managing the execution schedule of each batch 35 registered in the application server 5, respectively. As shown, a batch name field 65A and an execution timing field 65B are provided.

  The batch name column 65A stores the batch name of each batch 35 registered in the application server 5, and the execution timing column 65B stores the execution timing (corresponding to the corresponding batch 35 scheduled by the batch schedule management unit 51). The start date and time of the batch 35 is stored.

  Thus, in the case of the example of FIG. 14, for example, the batch 35 "total batch 1" is scheduled to be executed at "everyday", "midnight", and the batch 35 "total batch 2" is "every Friday" It is shown that it is scheduled to run at "3 am".

  The batch priority management table 66 is a table used to manage the priority of each batch 35 set by the system administrator or the like for each batch 35 registered in the application server 5, as shown in FIG. , A batch name field 66A and a priority order field 66B.

  The batch name column 66A stores the batch name of each batch 35 registered in the application server 5, and the priority column 66B stores the batch set for the corresponding batch 35 by the system administrator or the like. The 35 priorities are stored. Note that the subsequent batch 35 of a certain batch 35 is set to have a lower priority than the batch 35.

  Accordingly, in the case of the example of FIG. 15, “total batch 2” has the highest priority among the batches 35 registered in the application server 5, and thereafter “total batch 3”, “total batch 4” and “total batch It is shown that the priorities of “aggregate batch 1” to “aggregate batch 4” are set such that the priority is lowered in the order of “1”.

  The resource management table 67 is a table used to manage the resources of the application server 5 and, as shown in FIG. 16, is configured to include a resource name column 67A and a resource amount column 67B.

  The resource name column 67A stores the resource name of each resource necessary for executing the batch 35 among the resources held by the application server 5, and the resource amount column 67B stores the resource amount of the corresponding resource. Ru.

  Therefore, in the case of the example of FIG. 16, the application server 5 holds “12” (correctly, the number of CPU cores is 12) “CPU” as the resource and “32” GB as the “memory”. Is shown.

  The batch execution history management table 68 is a table used to manage the execution history of each batch 35 registered in the application server 5, and as shown in FIG. 17, the batch name column 68A, the start date and time column 68B, and It is configured to have an end date and time field 68C.

  The batch name column 68A stores the batch name of the executed batch 35 among the batches 35 registered in the application server 5. The start date and time column 68B stores the date and time when the execution of the batch 35 was started, and the end date and time column 68C stores the date and time when the batch 35 ended.

  Accordingly, in the case of the example of FIG. 17, the execution of the batch 35 "aggregate batch 1" is started at "2017/2/3 1:00", and the process is completed at "2017/2/2 4:05", The execution of the batch 35 "aggregate batch 2" is started at "2017/2/3 3:00", and "2017/2/3 5:21" indicates that the processing is completed.

  The mart reference history management table 69 is a table used to manage the reference history of persons in charge of subsequent operations to the data mart 41 held by the database server 6, and as shown in FIG. 18, the target mart field 69A , Target column column 69B, target value column 69C, and reference date and time column 69D.

  The target mart field 69A stores the name of the referenced data mart 41, and the reference date and time field 69D stores the date and time when the reference was made. Also, in the target column field 69B, the name of the column (field) referenced at that time in the data mart 41 is stored, and in the target value field 69C, the name of the value referenced at that time in that column is stored. Ru.

  Therefore, in the case of the example of FIG. 18, reference to the product name “chocolate” in the column “product name” of the “sales performance mart” is performed on “2017/2/2 9:00” It is shown.

  The reference frequency management table 70 is a table used to manage the number of times of reference for each keyword used when referring to the data stored in the data mart 41 desired by the person in charge of the subsequent work or the like. As shown in the figure, the object mart field 70A, the object column field 70B, the object value field 70C, the composite key field 70D, and the reference frequency field 70E are provided.

  Then, in the target mart field 70A, the identifier of the data mart 41 that is to be searched at that time is stored, and in the target value field 70C, the keyword used in the search is stored as the target value.

  Also, the identifier of the column corresponding to the target value in the data mart 41 is stored in the target column column 70B, and the number of times the target value of the data mart 41 is referred to in the reference frequency column 70E. Stored.

  Furthermore, in the composite key field 70D, different numerical values are stored for each type of keyword used in the search. In this case, when a search is performed by combining a plurality of keywords, the same numerical value is stored in the composite key field of the plurality of keywords.

  Therefore, in the case of the example of FIG. 19, the data mart 41 of “sales record” is used as a combination of two keywords “chocolate” and “2/1 to 2/28” (two keywords and condition). It has been shown that there have been 10 searches.

  The importance management table 71 is a table used to manage the importance of each record of the data mart 41 set by the importance setting unit 53 (FIGS. 7 and 8), as shown in FIG. An importance level ID column 71A, an object mart column 71B, an object column column 71C, an object value column 71D, an importance level column 71E, and a setting date and time column 71F are provided.

  In the target mart column 71B, the target column column 71C, and the target value column 71D, the same information as the information stored in the target mart column 70A, the target column column 70B, and the target value column 70C of the reference frequency management table 70 respectively The same information as that of the compound key field 70D of the reference frequency management table 70 is stored in the importance level ID field 71A. The importance degree column 71E stores the importance degree set for the record (data) of the target value stored in the target value column 71D, and the setting date and time column 71F stores the date and time when the importance degree is set. Is stored.

  Therefore, in the case of the example of FIG. 20, for the data mart 41 of "sales record", for example, a record in which "product name" is "chocolate" within a period of "sales day" It is shown that the importance level of “data” is set to “A” on “2017/2/12”.

  In addition, the person in charge of the succeeding business who registered the batch 35 for registering the processing result in the data mart 41 in the application server 5 or the system administrator of the store management system 1 etc. It may be freely set and changed.

(3) Configuration of Various Screens FIG. 21 shows that when the person in charge of the subsequent business has performed a predetermined operation for registering a new batch 35 in the application server 5 using the host device 7 of its own, the host device 7 The schematic structure of the batch registration screen 80 displayed is shown. The batch registration screen 80 is configured to include a batch main body registration method designation field 81, a batch name setting field 82, a batch execution condition setting field 83, an OK button 84, and a cancel button 85.

  The batch main body registration method designation field 81 is a field for designating how to register the main body of the batch 35 which is the object at that time with the application server 5, and the registration method designation column 81A and the pull-down button 81B Prepare.

  Then, in the batch main body registration method designation field 81, by clicking the pull-down button 81B, a pull-down menu (not shown) is displayed on which several methods available for registering the target batch 35 in the application server 5 are listed. It can be done. Thus, the person in charge can designate the registration method as the registration method of the target batch 35 to the application server 5 by selecting the desired method from the registration methods listed in the pull-down menu. it can. Then, the registration method designated at this time is displayed in the registration method designation column 81A.

  The batch name setting field 82 is a field for specifying a batch name of the batch 35 to be registered at that time, and a batch name input text box 82A for inputting the batch name is provided.

  Further, the batch execution condition setting field 83 is a field for setting the execution condition of the batch 35 to be registered at that time, and the execution frequency of the batch 35, the start condition and the completion condition, and the estimated required time of the batch 35 , And the resource to be allocated to the batch 35 can be specified.

  Thus, the person in charge of the follow-up operation causes the registration method designation column 81A to display the registration method of the batch 35 to be registered at that time in the application server 5 as described above, and enters the batch name of the batch 35 as the batch name. The batch 35 can be registered in the application server 5 by entering the text box 82A and setting the execution condition of the batch 35 in the batch execution condition setting field 83 and clicking the OK button 84.

  At this time, in the application server 5, under control of the batch analysis unit 50 (FIGS. 7 and 8), data of the batch 35 is taken in and stored in the storage device 32 (FIGS. 1 and 7). The execution condition of the batch 35 set by the person in charge of the succeeding operation on the batch registration screen 80 is stored in the batch execution condition management table 64 (FIG. 13).

  Also, the person in charge of the subsequent job can close the batch registration screen 80 after canceling all the settings performed on the batch registration screen 80 by clicking the cancel button 85.

  On the other hand, FIG. 22 shows a schematic configuration of the importance level setting rule determination screen 90 displayed on the host device 7 after registering the batch 35 in the application server 5 using the batch registration screen 80 of the person in charge of the subsequent business. The importance level setting rule determination screen 90 is a screen for determining the importance level setting rule 73 for the batch 35 registered at that time. The importance level setting rule determination screen 90 is configured to include an importance level setting rule determination field 91, a start condition setting field 92, an OK button 93, and a cancel button 94.

  The importance setting rule determination field 91 is a field for determining and setting the importance setting rule for the batch 35 registered at that time, and the names of all the importance setting rules 73 registered in advance by the system administrator or the like. (Hereinafter, this will be referred to as an importance setting rule name) 91A, a brief explanation 91B of the importance setting rule, and a toggle button 91C arranged on the left side of each importance setting rule name are displayed.

  Thus, the person in charge turns on the toggle button 91C displayed in front of the name (importance setting rule name 91A) of the importance setting rule 73 considered to be appropriate for the batch 35 in the importance setting rule determination screen 90. By setting an OK button 93 after setting, the importance level setting rule 73 can be specified as an importance level setting rule to be applied to the batch 35.

  The start condition setting field 92 is a field for setting a cycle for updating the importance of each record of the data mart 41 in which the batch 35 selected at that time registers the processing result, and specifies the frequency and date and time of the update. It is made to be able.

  Thus, the person in charge designates the importance setting rule 73 to be applied to the batch 35 registered in the application server 5 at that time in the importance setting rule determination field 91, and further the corresponding data mart 41 according to the importance setting rule 72. By specifying the update cycle of the importance of each record in the start condition setting field 92 and clicking the OK button 93, the importance setting rule 73 for the batch 35 can be determined and set.

  At this time, in the application server 5, under the control of the batch analysis unit 50, the importance setting rule 73 for the batch 35 set by the person in charge using the importance setting rule determination screen 90 corresponds to the batch 35. And stored in the importance level setting rule determination table 61 (FIG. 10).

  Also, the person in charge may close the importance level setting rule determination screen 90 after canceling all the settings performed on the importance level setting rule determination screen 90 by clicking the cancel button 94. be able to.

  On the other hand, FIG. 23 shows the contact information of the person in charge displayed on the host apparatus 7 after the person in charge of the succeeding operation decides the importance degree setting rule 73 of the batch 35 registered at that time using the importance degree setting rule determination screen 90 The schematic structure of the registration screen 100 is shown.

  In the person-in-charge contact registration screen 100, a person-in-charge contact input text box 101 for the person in charge to input his / her e-mail address, an OK button 102, and a cancel button 103 are displayed.

  Thus, the person in charge can register his / her e-mail address by clicking the OK button 102 after inputting his / her e-mail address in the person-in-charge contact input text box 101. In this case, in the application server 5, under the control of the batch analysis unit 50, the e-mail address entered in the contact person contact address input text box 101 is associated with the batch 35 registered at that time, and the contact person in charge is contact person. It is registered in the table 62.

  Further, by clicking the cancel button 103, the person in charge can close the person-in-charge contact registration screen 100 without registering his / her e-mail address in the application server 5.

(4) Various processes related to the batch management function according to the present embodiment Next, as to various processes to be executed in the application server 5 in relation to the batch management function according to the present embodiment These processes will be described in this order, divided into “various processes”, “various processes performed during normal operation”, and “various processes performed when data change is detected”.

  In the following, although the processing subject of various types of processing will be described as a program ("~ unit"), in practice, the CPU 30 (Fig. 1) of the application server 5 executes the processing based on the program.

(4-1) Various types of processing executed at registration of new batch (4-1-1) First new batch registration processing FIG. 24 shows one of the host devices (FIG. 24) according to the predetermined operation of the person in charge of the subsequent business. Hereinafter, when a request to register a new batch 35 (hereinafter, this will be referred to as a new batch registration request) is given to the application server 5 from the target host device 7, the batch analysis unit 50 (see FIG. 7, FIG. 8) shows a processing procedure of the first new batch registration processing.

  The batch analysis unit 50 starts the first new batch registration process shown in FIG. 24 when such a new batch registration request is given, and first displays the batch registration screen 80 described above with reference to FIG. (S1). Further, the batch analysis unit 50 waits for the person in charge to complete necessary settings such as execution conditions of the batch 35 to be registered (hereinafter referred to as a registration target batch) using the batch registration screen 80 (S2) ).

  When the necessary setting for the registration target batch 35 is eventually completed, the batch analysis unit 50 executes the execution conditions of the registration target batch 35 input to the batch registration screen 80 in the batch execution condition management table 64 (FIG. 13). Store (S3).

  Subsequently, the batch analysis unit 50 inquires of the batch schedule management unit 51 (FIGS. 7 and 8) whether or not there is an idle time for executing the registration target batch 35 so as to satisfy the execution condition (S4). Then, when the batch schedule management unit 51 notifies the batch analysis unit 50 of the result of the inquiry, the batch analysis unit 50 determines whether there is an idle time to execute the registration target batch 35 based on the notified inquiry result ( S5).

  If the batch analysis unit 50 obtains a negative result in this determination, it causes the target host device to display an error display indicating that the registration target batch 35 can not be registered under the current execution condition, and the batch execution condition management table 64 in step S3. The execution condition of the registration target batch 35 registered in is deleted from the batch execution condition management table 64 (S6). After that, the batch analysis unit 50 ends the first new batch registration process.

  On the other hand, when the batch analysis unit 50 obtains a positive result in the determination of step S5, it analyzes the registration target batch 35 to extract necessary information, and the extracted information is input / output management table 63 (FIG. 12 (FIG. 12). And the influence degree calculation rule determination table 60 (FIG. 9) (S7).

  Specifically, the batch analysis unit 50 analyzes the registration target batch 35 to detect the data referred to by the registration target batch 35 and the data mart 41 outputting the processing result, and the detection result is the registration target It is registered in the input / output management table 63 in association with the batch 35. Further, the batch analysis unit 50 should apply the registration target batch 35 to the registration target batch 35 based on the processing content of the registration target batch 35 (including the aggregation processing, the calculation processing of the maximum value, or the sorting processing). The influence calculation rule 72 is determined, and the determined influence calculation rule 72 is associated with the registration target batch 35 and registered in the influence calculation rule determination table 60 (FIG. 9).

  Subsequently, the batch analysis unit 50 causes the target host device 7 to display the importance level setting rule determination screen 90 described above with reference to FIG. 22 (S8). The batch analysis unit 50 then waits for the importance level setting rule 73 to be applied to the registration target batch 35 to be set by the person in charge using the importance level setting rule determination screen 90 (S9).

  Then, when the importance level setting rule to be applied to the registration target batch 35 is set by the person in charge, the batch analysis unit 50 associates the importance level setting rule 73 set at that time with the registration target batch 35. The importance level setting rule determination table 61 (FIG. 10) is registered (S10).

  Next, the batch analysis unit 50 causes the target host device 7 to display the person-in-charge contact address registration screen 100 described above with reference to FIG. 23 (S11). Further, after that, the batch analysis unit 50 uses the person-in-charge contact setting screen 100 to set the contact address (e-mail address) of the person-in-charge of the subsequent business who registered the registration target batch 35 in the application server 5. Wait for (S12).

  Then, when the contact address of the person in charge who has registered the registration target batch 35 is set, the batch analysis unit 50 associates the contact address set at that time with the registration target batch 35, and the contact person contact address table 62 (FIG. 11) is registered (S13). After that, the batch analysis unit 50 ends the first new batch registration process.

(4-1-2) Second New Batch Registration Process On the other hand, FIG. 25 shows idle time for executing the registration target batch 35 from the batch analysis unit 50 in step S4 of the first new batch registration process described above with reference to FIG. The process procedure of the 2nd new batch registration process performed by the batch schedule management part 51 (FIG. 7, FIG. 8) inquired whether or not exists is shown.

  The batch schedule management unit 51 starts the second new batch registration process shown in FIG. 25 when such a query is given from the batch analysis unit 50, and first obtains the batch name of the registration target batch 35 included in the query. To do (S20).

  Subsequently, the batch schedule management unit 51 acquires the execution condition of the registration target batch 35 by searching the batch execution condition management table 64 (FIG. 13) using the batch name acquired in step S20 as a keyword (S21). Further, after that, the batch schedule management unit 51 refers to the batch execution schedule table 65 (FIG. 14) to search for free time that satisfies the execution condition of the registration target batch 35 (S22).

  Next, the batch schedule management unit 51 determines whether a vacant time satisfying the execution condition of the registration target batch 35 has been detected by the search of step S22 (S23). When the batch schedule management unit 51 obtains a negative result in this determination, it notifies the batch analysis unit 50 that the registration target batch 35 can not be registered in the batch execution schedule as a query result (S24), and thereafter, End the new batch registration process of 2.

  On the other hand, when the batch schedule management unit 51 obtains a positive result in the determination of step S23, it notifies the batch analysis unit 50 that the registration target batch 35 can be registered in the batch execution schedule. Further, the batch schedule management unit 51 updates the batch execution schedule table 65 to determine the start time and the like of the registration target batch 35 and incorporate the execution schedule of the registration target batch 35 into the batch execution schedule (that is, the registration target batch 35 The execution schedule is registered in the batch execution schedule table 65) (S25). Then, the batch schedule management unit 51 ends this second new batch registration process.

(4-2) Various processing executed during normal operation (4-2-1) Normal operation processing FIG. 26 shows the normal operation executed by the batch control unit 56 (FIG. 7, FIG. 8) during normal operation of the application server 5. The processing procedure of operation processing is shown. When the application server 5 is started, the batch control unit 56 starts this normal operation processing, and first, referring to the batch execution schedule table 65 (FIG. 14) (S30), the batch 35 to be started at the current time is It is determined whether it exists (S31).

  If the batch control unit 56 obtains a negative result in this determination, the process returns to step S30, and thereafter repeats the loop of step S30-step S31-step S30 until a positive result is obtained in step S31.

  Then, when the batch control unit 56 obtains a positive result in step S31 by detecting the batch 35 to be executed at the current time on the batch execution schedule table 65, the data to be referred to by the batch 35 (sales data 14 or The difference data from the previous execution time of the data of the record of the data mart 41 is acquired from the database server 6 (FIG. 1). Further, the batch control unit 56 executes the batch 35 with the acquired difference data as an input, and stores the execution result in the corresponding data mart 41 (S32).

  Further, the batch control unit 56 notifies the batch schedule management unit 51 that the execution of the batch 35 is completed, and registers the next execution schedule of the batch 35 in the batch execution schedule table 65 (S33). Then, the batch control unit 56 returns to step S30, and thereafter repeats the processing after step S30.

(4-2-2) Importance Management Table Update Processing FIG. 27 shows the importance executed by the importance setting unit 53 (FIG. 7, FIG. 8) in parallel with the normal operation processing described above with reference to FIG. The processing procedure of degree management table update processing is shown.

  The importance degree setting unit 53 starts the importance degree setting process shown in FIG. 27 when the application server 5 is started, and first, the importance degree setting rule determination table 61 (FIG. 10) is referred to. The activation conditions of the respective importance setting rules set respectively for the registered batches 35 are confirmed (S40).

  Subsequently, the importance level setting unit 53 determines, on the basis of the check result in step S40, whether or not there is an importance level setting rule that satisfies the activation condition at the current time (S41). When the importance degree setting unit 53 obtains a negative result in the determination of step S41, the process returns to step S40, and thereafter repeats the loop of step S40-step S41-step S40.

  When the importance setting unit 53 obtains an affirmative result in step S41 due to the fact that the activation condition of any of the importance setting rules is eventually satisfied, the processing result of the corresponding batch 35 is stored according to the importance setting rule. The importance of each record of the data mart (hereinafter referred to as target data mart) 41 is calculated, and the importance management table 71 (FIG. 19) is updated according to the calculation result (S42).

  For example, if the importance level setting rule satisfying the start condition is the first importance level setting rule described above, the importance level setting unit 53 refers to the mart reference history management table 69 (FIG. 18) at that time. The frequency management table 70 (FIG. 19) is updated. Further, the importance setting unit 53 sets the importance of the data of each of the top 20% records having a large number of references to “A” for each target data mart 41 based on the updated reference frequency management table 70. The importance of data in the top 20 to 40% records is "B", the importance of data on each top 40 to 60% records is "C", and the number of references is top 60 to 100% The importance degree management table 71 (FIG. 20) is updated so that the importance degree of the data in (4) is "D" (S43).

  If the importance level setting rule satisfying the start condition is the second importance level setting rule described above, the importance level setting unit 53 determines that the target data mart 41 (for example, one shown in FIG. 3E). After creating the calculation table 111 as shown in FIG. 28 obtained by copying B), an average value of target values (numerical values stored in the sales column 111C) of each record is calculated. Further, the importance setting unit 53 stores the absolute value of the difference between the target value of the record and the average value for each record in the deviation column 111D of the record using the calculated average value.

  Then, based on the deviation for each record obtained as described above, the importance degree setting unit 53 sets the importance degree of the top 20% records with large deviation to “A”, and the importance degree of the top 20 to 40% records. The importance management table 71 is updated so that “B”, the top 40 to 60% of the record importance is “C”, and the top 60 to 80% of the record importance is “D”.

  Next, the importance level setting unit 53 sorts the records of the importance level management table 71 updated as described above in the descending order of the setting date and time stored in the setting date and time column 71F, and one lower x% importance level Reset each to the lower importance. Note that "x" is a preset value. Therefore, for example, the target record is re-selected as "B" if the previous importance is "A", "C" if "B", and "D" if "C". It will be set.

  Then, the importance setting unit 53 ends the importance setting processing.

(4-2-3) Change Detection Processing FIG. 29 shows a processing procedure of change detection processing executed by the change detection unit 54 (FIGS. 7 and 8) during normal operation. When the application server 5 is activated, the change detection unit 54 starts the importance level setting process shown in FIG. 29, and first, the sales data 14 registered in the database 40 (FIG. 1) of the database server 6 (FIG. 1). With respect to the data of each record of each data mart 41, the presence or absence of the change of the data is sequentially confirmed based on the access log or the like to the data (S50).

  Subsequently, the change detection unit 54 determines whether or not any data has been changed based on the confirmation result of step S50 (S51). When the change detection unit 54 obtains a negative result in this determination, the process returns to step S50, and thereafter repeats the loop of step S50-step S51-step S50.

  In addition, when the change detection unit 54 obtains an affirmative result in step S51 due to any data being changed in advance, a notification indicating that the data has been changed is designated (hereinafter referred to as “changed data”). The notification of change detection is sent to the batch control unit 56 (FIG. 7, FIG. 8) (S52). The change detection notification also includes information representing the change content of the data. Then, the change detection unit 54 returns to step S50, and thereafter repeats the processing after step S50.

(4-3) Various processing executed when data change is detected (4-3-1) Batch reexecution determination processing On the other hand, in FIG. 30, the above-mentioned change detection notification is given to the batch control unit 56. The processing procedure of the batch re-execution determination processing executed by the batch control unit 56 will be described. When the change detection unit receives the change detection notification from the change detection unit, the batch control unit 56 starts the batch re-execution determination process shown in FIG. 30, and first changes are made with reference to the input / output management table 63 (FIG. 12). All the batches 35 which refer to the data and the records of all the data marts 41 in which the batches 35 respectively store the processing results are specified (S60).

  Subsequently, the batch control unit 56 selects one batch 35 from each of the batches 35 specified in step S60 (S61), and frees the selected batch (hereinafter referred to as a selected batch) 35 to be re-executed. While inquiring of the batch schedule management unit 51 whether or not time exists (S62), the influence calculation request for requesting the calculation of the influence of the change of the data on the subsequent business is the influence calculation unit 55 (FIG. 7, FIG. FIG. 8) is sent (S63). The influence calculation request includes the change content of the changed data and the information of the selected batch 35 identified in step S60.

  Next, the batch control unit 56 waits for receipt of the response from the impact degree calculation unit 55 to the impact degree calculation request (that is, the impact degree of each record of the data mart 41 identified in step S60) (S64). When such an answer is received, the importance of the record of the data mart 41 in which the selected batch 35 registers the processing result among the data mart 41 specified in step S60 is acquired from the importance degree management table 71 (FIG. 20). S65). At this time, the batch control unit 56 sets the importance to “E” for the data of which the importance is not set in the importance management table 71 among the data of the record of the data mart 41.

  Furthermore, the batch control unit 56 determines the degree of importance calculated for the corresponding data mart 41 obtained in step S65 and the degree of importance for each record of the corresponding data mart 41 acquired in step S65. It is determined whether it is necessary to re-execute the selection batch 35 (S66). This determination is made by determining whether there is a record whose influence exceeds the allowable range.

  When the batch control unit 56 obtains a positive result in this determination, it notifies the batch schedule management unit 51 that the selected batch 35 should be re-executed (S68). The batch control unit 56 also notifies the reflection notifying unit 57 that the selected batch 35 is to be re-executed (S69), and then the process proceeds to step S72. Thus, the reflection notification unit 57 that receives this notification acquires the contact of the person in charge who has registered the selected batch 35 from the person-in-charge contact table 62 (FIG. 11), and re-executes the selected batch 35 for that contact. Notify that.

  On the other hand, if the batch control unit 56 obtains a negative result in the determination of step S66, the application is executed even if the selected batch 35 is re-executed based on the answer from the batch schedule management unit 51 to the inquiry of step S62. It is determined whether all the other batches 35 registered in the server 5 are in time for the completion deadline defined by the execution condition (S67). The contents of the response from the batch schedule management unit 51 to the inquiry in step S62 will be described later (step S90 in FIG. 31).

  When the batch control unit 56 obtains a positive result in this determination, the process proceeds to step S68, and thereafter processes step S68 and step S69 in the same manner as described above, and then proceeds to step S72.

  On the other hand, when the batch control unit 56 obtains a negative result in the determination of step S67, it does not re-execute the selected batch 35, and the data mart 41 corresponds to the change of the data at the next execution of the selected batch 35. (S70), and notifies the batch schedule management unit 51 to that effect (S71).

  Thereafter, the batch control unit 56 determines whether or not the processing after step S61 has been completed for all the batches 35 identified in step S60 (S72). Then, when the batch control unit 56 obtains a negative result in this determination, the process returns to step S61, and thereafter repeats the processing after step S61 while sequentially switching the batch 35 selected in step S61 to another unprocessed batch 35. .

  When the batch control unit 56 obtains an affirmative result in step S72 by completing the processing in step S61 and subsequent steps on all the batches 35 identified in step S60, the batch re-execution determination processing ends.

  When the batch 35 is re-executed, if the processing of the batch 35 is already completed and the processing result is registered in the corresponding data mart 41, the data mart 41 is executed along with the re-execution of the batch 35. The processing result of the batch 35 stored in is also updated. And in this case, since the change of data of this data mart 41 is detected by the change detection unit 54, the same batch re-execution is carried out also for the subsequent batch 35 of the batch 35 which executes the process with reference to the processing result. Execution determination processing is executed.

  Thus, if it is determined that a rerun is performed for a certain batch 35, the batch rerun is performed for each subsequent batch 35 for which processing of the batch 35 has already been completed, until it is determined that rerun is not performed for the immediately subsequent batch 35. The determination process is to be executed.

(4-3-2) Free time search process On the other hand, FIG. 31 shows the batch 35 (selected batch selected in step S61 of the batch reexecution determination process in step S62 of the batch reexecution determination process described above with reference to FIG. The processing procedure of the idle time search process executed by the batch schedule management unit 51 (FIG. 7, FIG. 8) inquired whether there is an idle time for re-executing 35) is shown.

  When the query is given, the batch schedule management unit 51 starts this free time search process, and first acquires the batch name of the selected batch 35 included in the query (S80). Further, the batch schedule management unit 51 acquires the execution condition of the selected batch 35 from the batch execution condition management table 64 (FIG. 13) (S81).

  Subsequently, the batch schedule management unit 51 refers to the batch execution schedule table 65 (FIG. 14) to search for free time that satisfies the execution condition acquired in step S81 (S82), and can detect such free time. It is determined whether or not it is (S83).

  When the batch schedule management unit 51 obtains a positive result in this determination, it adds the start timing of the selected batch 35 to the batch execution schedule table 65 so as to re-execute the selected batch 35 in the idle time (S89). In addition, even if the batch schedule management unit 51 transmits to the batch control unit 56 a response to the effect that all the batches 35 can be completed by the completion deadline determined by the execution condition even if the selected batch 35 is re-executed (S90 ), End this free time search process.

  On the other hand, when the batch schedule management unit 51 obtains a negative result in the determination of step S83, it refers to the batch priority management table 66 (FIG. 15) and registers the execution schedule in the batch execution schedule table 65 at that time. Among the batches 35 being processed, the batch 35 having the lowest priority is identified (S84).

  Next, whether the batch schedule management unit 51 can unlead a vacant time for re-executing the selected batch 35 by delaying the execution time of the batch 35 specified in step S84 within the range of the execution condition of the batch 35 It is determined whether or not it is (S85). At this time, if there is a subsequent batch 35 in the selected batch 35, the batch schedule management unit 51 determines an idle time for re-executing the selected batch 35, also taking into consideration the execution conditions of the subsequent batch 35. Determine if you can.

  When the batch schedule management unit 51 obtains a positive result in this determination, the process proceeds to step S89, and after processing step S89 as described above, all batches 35 are executed even if the selected batch 35 is re-executed. A response to the effect that the completion can be completed by the completion deadline determined by the conditions is transmitted to the batch control unit 56 (S90). Then, the batch schedule management unit 51 ends this free time search process.

  On the other hand, when the batch schedule management unit 51 obtains a negative result in the determination of step S85, whether or not the process of step S85 has been executed for all batches 35 for which the execution schedule is registered in the batch execution schedule table 65. It is determined (S86).

  Then, if the batch schedule management unit 51 obtains a negative result in this determination, it selects a batch 35 whose priority is higher by one than the batch 35 identified in step S 84 with reference to the batch priority management table 66 ( S87), returns to step S85, and thereafter repeats the processing of step S85 to step S87 until an affirmative result is obtained in step S85 or step S86.

  Then, if the batch schedule management unit 51 obtains an affirmative result in step S86 by completing the processing of step S85 for all the batches 35 for which the execution schedule is registered in the batch execution schedule table 65 in the end, The selected batch 35 is executed by delaying the execution time of the batch 35 having the lowest priority among the batches 35 for which the execution schedule is registered in the batch execution schedule table 65 without considering the execution conditions of the batch 35. Unoccupied the free time for (S88).

  After that, the batch schedule management unit 51 proceeds to step S89, and after processing step S89 as described above, delays the completion time of one or more batches 35 more than the completion time limit defined by the execution condition. Thus, the response to the effect that the selected batch 35 can be executed is sent to the batch control unit 56 (S90). Then, the batch schedule management unit 51 ends this free time search process.

  When the batch schedule management unit 51 receives the notification of step S71 of the batch re-execution process described above with reference to FIG. 30, the batch schedule management unit 51 re-executes the selected batch 35 registered in the batch execution schedule table 65 in step S89. The execution timing is deleted from the batch execution schedule table 65, and the batch execution schedule table 65 is returned to the state before the update in step S89.

(4-3-3) Influence calculation process On the other hand, FIG. 32 shows the influence calculation to which the above-mentioned influence calculation request is given from the batch control unit 56 in step S63 of the batch reexecution judgment process described above with reference to FIG. The processing procedure of the influence degree calculation process performed by the part 55 (FIG. 7, FIG. 8) is shown.

  The influence degree calculation unit 55 starts the influence degree calculation processing when the influence degree calculation request is given from the batch control unit 56, and firstly, the influence degree calculation unit 60 refers to the influence degree calculation rule determination table 60 (FIG. 9). An influence degree calculation rule to be applied to the selected batch 35 specified in the calculation request is specified (S100).

  Subsequently, the influence degree calculation unit 55 calculates the influence degree that the change of the data included in the influence degree calculation request has on the subsequent work according to the influence degree calculation rule specified in step S100 (S101). In addition, since the calculation method of this influence degree is as having mentioned above about FIGS. 4-6, the description for the second time here is abbreviate | omitted.

  Then, the influence degree calculation unit 55 notifies the batch control unit 56 of the calculated influence degree (S102), and thereafter ends the influence degree calculation processing.

(5) Effects of the Present Embodiment As described above, in the store management system 1 of the present embodiment, the original data (data of the sales data 14 and the data mart 41) to which the batch 35 refers after the processing of the batch 35 starts. When the change is made, the application server 5 calculates the degree of influence of the change on the subsequent business, and determines whether or not the calculated degree of influence falls within the allowable range for each record of the corresponding data mart 41. , Re-execute the batch 35 only when there is a record whose influence exceeds the allowable range.

  Therefore, according to the present store management system 1, when the original data is changed, the corresponding batch 35 is re-executed only when the change has a large influence on the subsequent business, and the change has an influence on the subsequent business. If the size is small, the batch 35 is not re-executed, so occurrence of a situation where batch processing is not completed by the start time of the subsequent work due to the re-execution of the batch can be suppressed as much as possible, and the execution of the subsequent work is highly reliable. Can help.

(6) Other Embodiments In the above embodiment, the present invention is applied to the store management system 1 configured as shown in FIG. 1, but the present invention is limited to this. However, the present invention can be widely applied to information processing systems of various other configurations.

  In the above embodiment, although the case where the head office system 8 is configured by the web server 4, the application server 5, and the database server 6 has been described, the present invention is not limited to this. You may make it comprise with a server apparatus. In the above embodiment, the network 9A is configured by the Internet or the like, and the network 9B is configured by the in-house LAN or the like. However, the present invention is not limited thereto, and the networks 9A and 9B can be configured. A wide area network (WAN), a dedicated line, a wireless communication line, or the like may be used, and various other configurations can be widely applied as the configuration of the networks 9A and 9B.

  Furthermore, in the above-described embodiment, as a method of reflecting the change of data in the corresponding data mart, the case where the corresponding batch 35 is re-executed has been described, but the present invention is not limited to this. The data change may be reflected on the corresponding data mart by the method disclosed in JP 2010-122880 A, and various other methods may be used to reflect the data change on the corresponding data mart. Can be widely applied.

  The present invention can be widely applied to information processing apparatuses of various configurations that execute each registered batch and store the processing result of the executed batch in the corresponding data mart.

1 ... store management system, 2 ... store, 3 ... store terminal, 5 ... application server, 6 ... database server, 7 ... host device, 8 ... head office system, 10, 30 ... CPU, 14 ...... Sales data, 35, 35A to 35D ...... batch, 40 ...... database, 41, 41A to 41 D ...... data mart, 50 ...... batch analysis unit, 51 ...... batch schedule management unit, 52 ...... history management unit 53: importance setting unit 54: change detection unit 55: influence calculation unit 56: batch control unit 57: reflection notification unit 60: influence calculation rule determination table 61 ... Importance setting rule determination table, 62 ... Contact person in charge contact table, 63 ... I / O management table, 64 ... Batch execution condition management table, 65 ... Batch execution schedule table 66, batch priority management table 67, resource management table 68, batch execution history management table 69, mart reference history management table 70, reference frequency management table 71, importance management Table 72: Impact calculation rule 73: Importance setting rule.

Claims (12)

In an information processing apparatus that executes each registered batch and stores the processing result of the executed batch in the corresponding data mart,
An importance setting unit that sets an importance for each record of each data mart;
A change detection unit that monitors data that each of the batches refers to at the time of processing, and detects the change when any of the data is changed;
An influence degree calculation unit that calculates an influence degree that the change of the data has on subsequent work when the change detection unit detects any change of the data;
Based on the degree of importance set in each of the records of the data mart in which the processing result of the batch referring to the changed data is stored, and the degree of influence calculated by the influence degree calculation unit , And a batch control unit for reflecting the change of the data in the data mart as needed;
The allowable range of the influence degree is set in advance for each of the importance degrees,
The batch control unit
The influence based on the degree of importance set in each record of the data mart in which the processing result of the batch referring to the data is stored, and the degree of influence calculated by the influence degree calculation unit It is determined for each record of the data mart whether or not the degree is within the allowable range, and when the degree of influence of any of the records exceeds the allowable range, the change of the data is determined An information processing apparatus characterized by reflecting on a data mart. The batch control unit
When the degree of influence of any of the records in the data mart exceeds the allowable range, the data referred to by the batch is re-executed by storing the processing result in the data mart. The information processing apparatus according to claim 1, wherein the change of is reflected in the data mart. The batch control unit
If none of the records in the data mart has the impact degree exceeding the allowable range, all other registered batches are set for the batch even if the batch is re-executed. The information processing apparatus according to claim 2, wherein the batch is re-executed when the time limit is reached. A batch schedule management unit that manages an execution schedule of each of the registered batches;
The batch control unit
When the change detection unit detects any change in the data, the batch schedule management unit is inquired as to whether there is an idle time for re-executing the batch referring to the data;
The batch schedule management unit
If there is no free time to re-execute the batch inquired from the batch control unit, a range of execution conditions including the completion time limit set for the batch for the execution time of the other registered batch By delaying the delay in the memory and unrolling the idle time for re-executing the batch inquired from the batch control unit, and updating the execution schedule to re-execute the batch in the unoccupied idle time The information processing apparatus according to claim 3, characterized in that The importance setting unit is
Setting the importance of each record of each data mart on the basis of a predetermined value which changes with time for the data stored in the record;
The information processing apparatus according to claim 1, wherein the importance of each set record is periodically updated. A plurality of types of influence calculation rules corresponding to the processing content of each of the batches are registered in advance,
The influence degree calculation unit
The information processing apparatus according to claim 1, wherein the influence degree of the batch is calculated according to the influence degree calculation rule set for the batch at the time of registration of the batch. In an information processing method executed by an information processing apparatus, which executes each registered batch and stores processing results of the executed batch in corresponding data marts,
A first step of setting an importance level for each record of each data mart;
A second step of monitoring data referred to at the time of processing by each of the batches, and calculating a degree of influence of the change of the data on the subsequent work when any change of the data is detected;
Based on the degree of importance set in each of the records of the data mart in which the processing result of the batch referring to the changed data is stored, and based on the calculated degree of influence, if necessary, the data And a third step of reflecting the change in the data mart.
The allowable range of the influence degree is set in advance for each of the importance degrees,
In the third step, the information processing apparatus
The degree of influence falls within the allowable range based on the degree of importance set for each record of the data mart in which the processing result of the batch referring to the data is stored, and the calculated degree of influence It is determined for each record of the data mart whether or not there is any, and when the degree of influence of any of the records exceeds the allowable range, the change of the data is reflected in the data mart. Characteristic information processing method. In the third step, the information processing apparatus
When the degree of influence of any of the records in the data mart exceeds the allowable range, the data referred to by the batch is re-executed by storing the processing result in the data mart. The information processing method according to claim 7, wherein the change of is reflected in the data mart. In the third step, the information processing apparatus
If none of the records in the data mart has the impact degree exceeding the allowable range, all other registered batches are set for the batch even if the batch is re-executed. The information processing method according to claim 8, wherein the batch is re-executed when the time limit is met. The information processing apparatus is
Manage the execution schedule of each registered batch,
In the third step, the information processing apparatus
When a change in any of the data is detected, it is determined based on the execution schedule whether there is an idle time for re-executing the batch that refers to the data.
If there is no free time to re-execute the batch, change the execution time of the other registered batch by delaying within the range of execution conditions including the completion deadline set for the batch. The execution schedule is updated so as to find out the idle time for re-executing the batch referring to the stored data, and to re-execute the batch in the excavated free time. Information processing method described. The information processing apparatus is
Setting the importance of each record of each data mart on the basis of a predetermined value which changes with time for the data stored in the record;
8. The information processing method according to claim 7, wherein the importance level of each set record is periodically updated. A plurality of types of influence calculation rules corresponding to the processing content of each of the batches are registered in advance,
In the second step, the information processing apparatus
The information processing method according to claim 7, wherein the influence degree of the batch is calculated according to the influence calculation rule set for the batch at the time of registration of the batch.