JP2003044327A - Business startup discriminating device of information system under multiplexing environment and method for predicting business processing time - Google Patents

Abstract

PROBLEM TO BE SOLVED: To make dynamic and exact prediction of end time of a business possible by reflecting difference of degrees of influence by multiplexing degrees by every processing unit to constitute the business. SOLUTION: One business is segmentized by individual processing units having characteristics, processing time is estimated by using it as the minimum unit by a processing time prediction processing means 13 and the total of the processing time is defined as the entire processing time of the business. Since the estimation of the processing end time by processing unit is reflected on changes of multiplexing degrees, the processing time is predicted by taking weight (a weighting coefficient table 162) to the processing time by every multiplexing degree in one processing unit. In addition, the processing time is calculated by reflecting influences of changes of the multiplexing degrees to be generated when a new business is introduced not only on the new business but on the existing business and the end time of the entire businesses in a system is judged on the basis of the last end time among newly introduced businesses including the existing business.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a business activation discriminating apparatus for an information system in a multiplex environment, a business processing time prediction method, a business processing time prediction program in which the method is programmed, and the program are recorded. Recording medium.

[0002]

2. Description of the Related Art For example, in an information system in which the start time of each work is defined by a method called a scheduler on a daily basis, if the previous work is not completed by the work start time, the succeeding work is Not only do they all fall behind, but as a result, important tasks such as backup and information exchange with other systems may not be executed. In the worst case, it may cause serious trouble such as system stop. Especially in the closed system, strict adherence to the end time of each work is the most important design item, and is strictly required as a specification. Conventionally, the estimation and evaluation of the business processing time of an information system in a multiplexed environment is performed by using the whole business as the minimum unit used for estimating the processing end time, and the pre-blocking time is set by measuring the maximum processing time during the maximum possible multiplexing. A fixed value based on the value was used.

[0003]

In such estimation and evaluation, it is difficult to faithfully predict the extension of the business processing time due to multiplexing, and it is said that the error between the estimation result and the actual measurement value is large. Conceivable. Also, since the pre-blocking time was a fixed value, the usage of the system resources before and after that was sparse and dense, and the resources were not effectively used.

The present invention has been made in view of the above circumstances, and it is possible to dynamically and accurately predict the end time of a work by reflecting the difference in the degree of influence due to the multiplicity for each processing unit that constitutes the work. And a business process determination device for an information system in a multiplexed environment, a business process time prediction method, a business process time prediction program in which the method is programmed, and a recording medium in which the program is recorded. To aim. In addition, by implementing this method in an information system, all started tasks can be completed by the specified end time of requirements, and
When applied to the blocking system where the above-mentioned end time is strictly required, not only can serious troubles such as system stop be avoided, but also the optimization of the pre-blocking process of business can be realized.
Apparatus for determining business activation of information system in a multiplex environment in which system resources can be effectively used, business processing time prediction method, business processing time prediction program programmed by the method, and recording medium recording the program It is also intended to provide.

[0005]

In order to solve the above problems, the present invention subdivides an existing business including a newly input business into processing units and estimates a processing time for each processing unit.
A business start determination device for an information system in a multiplexing environment that predicts the total as the total processing time of the business, and for each processing unit of the newly input business, the processing start time from the business processing time management table. , The multiplicity information at the start point is acquired, the overlap with the existing work is taken into consideration, and the weighting coefficient table is referred to reflect it in the calculation of the processing time, and the processing time and the end time are reflected in the business processing time management table. For the first computing means for writing and for existing jobs whose multiplicity changes by receiving new jobs,
Multiplicity information is acquired from the business processing time management table, a new processing time is calculated by referring to the weighting coefficient table in consideration of overlap with a new business, and the processing time and end time are calculated as the business processing time. By repeating in chronological order the second computing means for writing in the time management table and the computations by the first and second computing means, dynamic changes in multiplicity with respect to newly input work and existing work are reflected. And a control means for writing the calculation result in the business processing time management table.

Further, in the business system activation discriminating apparatus of the information system in the multiplex environment of the present invention, the end times of the existing business and the newly input business are acquired from the business processing time management table, and the final end time of them is determined as the end requirement. It is characterized by further comprising business activation determination processing means for determining whether or not to activate the newly input business by comparing with a prescribed time.

Further, in the job activation discriminating apparatus of the information system in the multiplex environment of the present invention, the weighting coefficient table is subjected to a multi-running test in the same job or a plurality of kinds of jobs are mixed, and different multiplicity is obtained. At first, the start time and the end time of each processing unit are measured to calculate the traveling time, and the correlation between the business type, the processing unit type, and the multiplicity of the processing time is analyzed to set and register as the weighting coefficient. And a second weighting coefficient table that is set and registered as a weighting coefficient by measuring and analyzing the correlation between the work load, the processing unit, and the processing time.

In order to solve the above problems, the present invention provides
It is a job activation determination device for an information system in a multiplex environment in which an existing job including a newly input job is subdivided into processing units, a processing time is estimated for each processing unit, and the total is predicted as the total processing time of the job. And a business queuing means for receiving and accumulating a business processing request including the newly input business as a transaction, and a type of the business to be started next among the queued businesses, and a business information table Management target business selection processing means for determining whether or not it is a management target business by collating,
For existing jobs including newly input jobs, the change in multiplicity is dynamically reflected by referring to the weighting coefficient table for each processing unit, and the processing time for each job, processing unit, and multiplicity is calculated. Processing time prediction processing means for predicting the total as the total processing time, business activation determination processing means for determining whether or not to input the new input business based on the prediction result of the processing time, and the business activation And a business activation processing means for activating a new input business determined to be activated by the determination processing means.

With the above configuration, one work is subdivided into individual processing units having characteristics (the degree of influence on processing time is different from other portions due to changes in multiplicity), and the processing time is set as the minimum unit. The processing time is estimated by the prediction processing means, and the total is used as the total processing time of the business. Since the estimation of the processing end time of the processing unit is reflected in the change of the multiplicity, the method of adding the weight to the processing time for each multiplicity (weighting coefficient table) and predicting is adopted even in one processing unit. . In addition, the processing time is calculated by reflecting the influence of the change in multiplicity that occurs when a new job is input, not only on the new job but also on the existing job, and the end time of the entire job in the system includes the existing job. Judgment is based on the final end time of new input work. The coefficients set and registered in the weighting coefficient table are subjected to a multi-run test in the same job or a mixture of jobs of multiple types, and the start time and end time of each processing unit are measured at different multiplicities. It shall be obtained by calculating the traveling time and analyzing the correlation between the business type, the processing unit type, and the multiplicity with respect to the processing time. As a result, in an information system under a multi-driving environment, it is possible to reflect the dynamically changing multiplicity in real time for both existing business and new business in the system, and the influence of extension of processing time due to multiplexing. It is possible to more accurately predict the processing time taking into consideration.

The "processing unit" is a conceptual processing unit that constitutes a task, and physically represents a procedure, a module, or an aggregate thereof. For example, when there is a DB update operation, the DB update operation is not only a simple table update process, but also a preliminary check required for update, information acquisition, post-filing, and a difference DB.
May be included, and refers to each process that constitutes these functions. Further, the “weight” is a concept for quantitatively describing how each factor related to the processing time, here, the work, the processing unit, the work amount, and the multiplicity influence the processing time. . Specifically, a change in processing time with respect to a certain increase in each factor is approximately expressed by a mathematical formula described later.

In order to solve the above problems, the present invention provides
A method for predicting the processing time of an information system in a multiplexed environment in which existing operations including newly input operations are subdivided into processing units, the processing time is estimated for each processing unit, and the total is predicted as the overall processing time of the operation. Therefore, the estimation of the processing time for each processing unit is performed by adding a weight to the processing time for each multiplicity even within one processing unit to dynamically reflect the change in multiplicity and when a new job is input. The processing time is calculated by reflecting the change in multiplicity that occurs in the existing work including new work,
The processing time of the entire job is determined based on the final end time of all running jobs in the existing job including the newly input job.

In order to solve the above-mentioned problems, the present invention subdivides an existing work including a newly input work for each processing unit, estimates a processing time for each processing unit, and estimates the total as a total processing time of the work. A business processing time prediction program used for a business activation determination device of an information system in a multiplexed environment, wherein the business processing time prediction program is a business processing time management table for each processing unit of the newly input business. , The processing start time and the multiplicity information at the start time are acquired, and after considering the overlap with the existing work, the weighting factor table is referred to reflect the calculation of the processing time, and the processing time and the end time are described above. For the first step of writing in the processing time management table and for existing jobs whose multiplicity changes due to the input of new jobs,
Multiplicity information is acquired from the business processing time management table, a new processing time is calculated by referring to the weighting coefficient table in consideration of overlap with a new business, and the processing time and end time are calculated as the business processing time. By repeating the second step of writing in the management table and the first step and the second step in time series, the change of the multiplicity with respect to the newly input work and the existing work is dynamically reflected, and the calculation result is shown. A third step of writing the work processing time management table, and causing the computer to execute the first to third steps.

Further, in the business processing time prediction program of the present invention, the end times of the existing business and the newly input business are acquired from the business processing time management table, and the final end time is compared with the prescribed time of the end requirement, It is characterized by further comprising a fourth step of determining whether or not to start a new input work, and causing the computer to execute the first to fourth steps.

In the business processing time prediction program of the present invention, the first step acquires multiplicity information of a processing unit to be calculated from the start and end times of the processing of the business processing time management table. The processing time is divided by the step and the weighting coefficient and the multiplexing time obtained by dividing the processing unit to be calculated into parts with different multiplicities and referring to the first weighting coefficient table for each divided part. It is characterized by comprising a step of calculating and a step of updating the work processing time management table based on the calculated result, and causing a computer to execute each of the steps.

In the business processing time prediction program of the present invention, the second step acquires multiplicity information of a processing unit to be calculated from the start and end times of the processing of the business processing time management table. The processing time is divided by the step and the weighting coefficient and the multiplexing time obtained by dividing the processing unit to be calculated into parts with different multiplicities and referring to the first weighting coefficient table for each divided part. It is characterized by comprising a step of recalculating and a step of updating the business processing time management table based on the calculated result, and causing a computer to execute each of the steps.

Further, in the business processing time prediction program of the present invention, the third step tentatively calculates the processing time during independent traveling according to the business volume by referring to the second weighting coefficient table for each processing unit. Then, the step of updating the business processing time management table, the step of comparing the end time of the processing unit that is the calculation target with the processing time of the entire existing business, and determining the calculation order of the processing units,
According to the calculation order, it is instructed to calculate the processing time for each new business processing unit or for each existing business processing unit, the updated business processing time management table is referred to, and from the start and end times of the processing, It is determined whether or not there is another existing business that overlaps, and the steps of repeating the above steps until the existing business is exhausted, and whether or not there is a processing unit to be calculated after searching the business information table. And repeating the processing of each step until it is determined that there is no subsequent processing unit, and causing the computer to execute each of the steps.

In order to solve the above problems, the present invention provides
Used for a job activation determination device of an information system in a multiplex environment in which existing jobs including newly input jobs are subdivided into processing units, the processing time is estimated for each processing unit, and the total is predicted as the overall processing time of the job. A business processing time prediction program, wherein the business processing time prediction program receives a business processing request including the newly input business as a transaction and accumulates the business processing request, and starts next in the queued business. The type of task to be performed is identified and the step of determining whether or not it is a managed task by collating it with the task information table, and the weighting coefficient for each processing unit for existing tasks including newly submitted tasks Dynamically reflect changes in multiplicity by referring to the table, calculate the processing time for each task, processing unit, and multiplicity. A step of predicting the total as the total processing time, a step of determining whether or not to input the new input job based on the prediction result of the processing time, and a new input determined to be activated by the operation activation determination processing means. A step of activating a task, and causing a computer to execute each of the steps.

In order to solve the above problems, the present invention provides
Used for a job activation determination device of an information system in a multiplex environment in which existing jobs including newly input jobs are subdivided into processing units, the processing time is estimated for each processing unit, and the total is predicted as the overall processing time of the job. A computer-readable recording medium on which a business processing time prediction program is recorded, wherein the business processing time prediction program starts processing of each processing unit of the newly input business from a business processing time management table. Time and start time multiplicity information is acquired, after considering overlap with existing work, it is reflected in the calculation of the processing time by referring to the weighting coefficient table, and the processing time and end time are reflected in the business processing time management table. The first step of writing to the job and the job processing time management for the existing job whose multiplicity changes when a new job is input. It acquires multiplicity information from the table,
The second step of calculating a new processing time by referring to the weighting coefficient table in consideration of the overlap with the new business, and writing the processing time and the end time in the business processing time management table; By repeating the step and the second step in time series, the third step dynamically reflecting the change of the multiplicity with respect to the newly input work and the existing work and writing the calculation result in the work processing time management table is provided. , Causing the computer to execute the first to third steps.

In order to solve the above problems, the present invention provides
Used for a job activation determination device of an information system in a multiplex environment in which existing jobs including newly input jobs are subdivided into processing units, the processing time is estimated for each processing unit, and the total is predicted as the overall processing time of the job. A computer-readable recording medium having a business processing time prediction program recorded therein, wherein the business processing time prediction program receives and accumulates a business processing request including the newly input business as a transaction; The type of the next job to be started among the executed jobs and the step of determining whether it is a managed job by comparing it with the job information table, and the existing jobs including newly submitted jobs On the other hand, by referring to the weight coefficient table for each processing unit, the change in multiplicity is dynamically reflected, Processing unit, calculating the processing time for each multiplicity, and then predicting the total as the total processing time; and determining whether or not to input the new input work based on the prediction result of the processing time. And a step of activating a newly input business determined to be activated by the business activation determination processing means, and causing the computer to execute the first to third steps.

[0020]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 is a block diagram showing a functional expansion of the internal configuration of a business activation discriminating apparatus of an information system under a multiplexing environment of the present invention. Each block shown here is specifically a peripheral L including a CPU and a memory.
The function of each block is realized by the SI being configured and the CPU sequentially reading and executing the program recorded in the memory. The business activation discrimination device 1 is composed of a business queuing unit 11, a management target business selection processing unit 12, a processing time prediction processing unit 13, a business activation determination processing unit 14, a business activation processing unit 15, and a database (DB) 16. It is a server and is connected to a client (not shown) via the network 2.

The business queuing unit 11 receives a business processing request including a newly submitted business as a transaction,
Has a function to accumulate. The management target operation selection processing unit 12
A function of identifying the type of a job to be started next among the jobs queued in the job queuing unit 11 and comparing it with a job information table 161 described later to determine whether the job is a managed job have. Here, illegal commands that do not affect the processing time are excluded.

The processing time prediction processing unit 13 dynamically reflects the change in the multiplicity by referring to a weighting coefficient table 162, which will be described later, for each processing unit for existing operations including newly input operations, and , Has a function of calculating the processing time for each processing unit and multiplicity and predicting the total as the total processing time. The internal configuration of the processing time prediction processing unit 13 is shown in FIG. The processing time prediction processing unit 13 has a first
The calculation unit 131, the second calculation unit 132, and the control unit 133 are included. The above-described first calculation unit 131 and second calculation unit 13
2. Peripheral LSI including CPU and memory for both control unit 133
It is assumed that the function of each block is realized by the CPU sequentially reading and executing the program recorded in the memory.

The first arithmetic unit 131, for each processing unit of a new input work, a work processing time management table 1 described later.
The processing start time and the multiplicity information at the start time are acquired from 63, the overlapping with the existing work is taken into consideration, and the weighting coefficient table 162B is referred to and reflected in the calculation of the processing time. Has a function of writing in the business processing time management table 163. The second calculation unit 132
For the existing business whose multiplicity changes due to the input of a new business, the multiplicity information is acquired from the business processing time management table 163, and the weighting coefficient table 162B is referenced after considering the overlap with the new business. Has a function of calculating a new processing time and writing the processing time and the end time in the business processing time management table 163. Control unit 133
Controls the repetition of the operations by the first and second operation units 131 and 132 in time series, thereby dynamically reflecting the change of the multiplicity with respect to the newly input operation and the existing operation, and the calculation result is applied to the operation. It has a function of writing to the processing time management table 163.

The work activation discrimination processing unit 14 has a function of judging whether or not to input a new input work based on the prediction result of the processing time. Specifically, the end times of the existing work and the new input work are acquired from the work processing time management table 163, and the final end time is compared with the specified time of the end requirement to determine whether to start the new input work. To do. The business activation processing unit 15 has a function of activating a newly input business determined to be activated by the business activation determination processing unit 14.

The DB 16 is made up of various data tables. Here, a business information table 162, a weighting coefficient table 162, a business processing time management table 163, and a remaining business management table 164 are allocated. The respective data structures are shown as (a) to (e) in FIG. 3, respectively. Business information table 161 shown in (a)
Has, as an entry, a process ID subdivided for each business ID. There are two types of weighting coefficient tables 162 shown in (b) and (c). One (b) has a work load function set and registered for each process ID subdivided for each work ID, and the other one. In (c), the multiplicity weighting coefficient to be reflected in the calculation of the processing time for each multiplicity is set and registered. Hereinafter, for convenience, the former will be referred to as a work load weighting coefficient table 162A and the latter will be referred to as a multiplicity weighting coefficient table 162B. Multiplicity weight coefficient table 1
62B is illustrated only in the case of 2 multiplexes, and therefore has a matrix structure here.

The work amount weighting table 162A is set and registered as a weighting coefficient by measuring and analyzing the correlation between the work amount, the processing unit, and the processing time. The multiplicity weighting table 162B stores the same work or a plurality of types of work. The multi-run test is conducted in a mixed state, and the running time is calculated by measuring the start time and end time of each processing unit at different multiplicities, and the business type, the processing unit type, and the correlation of the multiplicity with the processing time are calculated. It is assumed that the weight coefficient is set and registered by analyzing the relationship. The task processing time management table 163 shown in FIG.
It has D, multiplicity, start time, end time, processing time, and single travel processing time as entry items, and is referred to and updated by the processing time prediction processing unit 13 and the work activation determination processing unit 14. Further, the remaining work management table 164 shown in (e)
In, a transaction ID and a transaction ID of a transaction including online real time and batch are set and registered as a pair.

FIG. 4 is a flow chart showing a procedure for realizing the business processing time prediction method for an information system in a multiple environment according to the present invention. Also, FIGS.
Corresponds to each step of the flowchart shown in FIG.
FIG. 7 is an operation conceptual diagram quoted for explaining the details of the operation procedure for each step. 2 to 3 in FIGS. 5 to 13.
The blocks to which the same numbers are attached are the same as those in FIGS. 2 and 3. Hereinafter, referring to FIG. 4 to FIG.
The operation of the embodiment of the present invention shown in FIG. 3 will be described in detail below.

First, the business activation determination device 1 stores the business that has arrived at the system in the business queuing section 11 as a new input business. Here, as shown in FIG. 5, 12:
Job A (DB update job) is queued at 00, and the job amount is 200 records. Further, it is assumed that the job A has processing units A1 to A3, and there is a job B having processing units B1 to B3 as a higher-ranking existing job. As shown in FIG. 6, the management target business selection processing unit 12 uses the business ID as a key for the business information table 1
By searching 61, it is determined whether or not the business ID of the queued business exists in the business information table 161, whether the queued business is a management target by the process of the present invention (step S31). That is, here, the monitoring of the processing time by the irregular command that does not affect the processing time is omitted. If it is determined to be a management target here, the process proceeds to the "processing unit information acquisition" process of step S32, and if it is not a management target, step S4.
Proceed to the "business activation" process in 2.

In step S32, as shown in FIG. 7, the number information of the processing units constituting the business is acquired from the business information table 161 using the business ID of the newly input business as a key. Then, the processing unit ID that is the calculation target of the new business is set in the business processing time management table 163, and is registered as a new record together with the processing start time. next,
Step S3 for the processing unit (A1) to be calculated
The "higher-level existing business determination" process 3 is performed. That is, FIG.
As shown in, the processing time in the case of traveling alone is temporarily calculated by acquiring the work amount function from the work amount weighting coefficient table 162A and substituting the work amount V. Then, by referring to the work processing time management table 163, it is checked whether or not there is a higher-ranking process from the start and end times of the process which is tentatively calculated previously (step S33). Here, if there is a higher-level existing business, the process proceeds to the "new business process unit end time determination" process of step S34, and if not, to the "new business subsequent process unit determination" process of step S40. Here, 1
2: 00-12: 04 is 2 multiplexes with the processing unit B2, 1
At 2: 04-12: 10, 2 multiplexes with the processing unit B3 are recognized. Note that, among the higher-ranking existing jobs, those that are completed before the start time of the new job are deleted from the processing time management table 163.

In the "new work processing unit end time determination" processing of step S34, the end time of the processing unit to be processed is compared with the end time (T) of the entire upper existing work, and the calculation order of the processing units is calculated. to decide. Here, FIG.
As shown in (a), when the target processing unit ends first, the calculation is performed in the order of the target processing unit and the upper existing business.
Further, as shown in FIG. 9B, when the upper-layer existing work is finished first, the upper-layer existing work and the target processing unit are calculated in this order. Here, the reason why the calculation order is changed depending on the end time is that, as shown in FIG. 9C, the multiplexing time with the new business processing unit A1 changes depending on how long the processing time of the existing business processing unit C2 extends. So processing unit A1
This is because if the processing time is calculated first, it is necessary to calculate A2 again after calculating C2.

The new business process unit calculation process (steps S36 and S37) will be described. Here, the multiplicity information and the multiplicity weighting coefficient (weighting coefficient table 162
The processing time is predicted and calculated from B), and the contents of the processing time management table 163 are updated. That is, as shown in FIG. 10, the first calculation unit 131 of the processing time prediction processing unit 13
Refers to the business processing time management table 163, and from the start time and end time of the processing, multiplicity information (processing and how long,
Multilevel) and divide it into parts with different multiplicities. Then, the processing time is calculated from the multiplexing time and the multiplicity weighting coefficient for each of the divided parts.

In this case, 12: 00-12: 04 is multiplexed with the processing unit B2, and 12: 04-12: 10 is multiplexed with the processing unit B3. Therefore, the processing unit A1 is divided into parts of 4: 6 according to the processing time of the processing unit B2, and calculation is performed for each part. That is, the processing unit A
1 and B2 multiplexing time 4 minutes, processing units A1 and B2 multiplexing coefficient 1.25 (see multiplexing weight coefficient table 162B)
The processing time of one part is 5 minutes (4 × 1.5), the multiplexing time of the processing units A1 and B3 is 6 minutes, and the processing units A1 and B3 are
Multiplicity coefficient of 1.5 (multiplicity weight coefficient table 162B
Processing time of the other part is 9 minutes (6 × 1.
5), the processing time of the processing unit A1 that takes multiplicity into consideration is 14 minutes (5 + 9). Processing time prediction processing unit 13
Further updates the contents of the work processing time management table 163 according to the result calculated as described above.

Next, the "processing unit recalculation of existing work" processing in steps S35 and S38 will be described. Here, the processing time of the upper-layer existing business that changes due to the addition of a new business is recalculated. The calculation procedure is the same as the above-mentioned "new business processing unit calculation" processing (steps S36 and S37), and description thereof is omitted to avoid duplication. Here, as shown in FIG. 11, since the multiplexing of the processing units B2 and A1 is 4 minutes and the multiplicity coefficient of the processing units B2 and A1 is 1.5, the processing unit B
2 is 4 minutes, the multiplexing of the processing units B3 and A1 is 8 minutes, and the multiplicity coefficient of the processing units B3 and A1 is 1.25, so the processing unit B3 is 10 minutes. Also in this case, of course, the result of recalculation of the processing time is written in the business processing time management table 163 for the part where the new business processing unit is multiplexed with the processing unit of the upper existing business.

In the "higher-level existing business determination process" in step S39, the existence of a higher-level existing business is checked from the business processing time management table 163. That is, as shown in FIG. 12, after the processing for the processing unit B1 is completed, it is determined whether or not there is a higher-ranking existing job that has not been processed, from the processing start time and the processing end time to the job processing time management table 163. Check by referring to. If there is, the process returns to step S33, and if not, the process proceeds to "new job subsequent process unit determination process" of step S40.
In the “new job subsequent processing unit determination” processing of step S40, as shown in FIG.
To determine whether or not there is a subsequent processing unit in the same job following the processing unit A1. If it exists, the process returns to the "higher-level existing work determination" process of step S33, and if it does not exist, the process proceeds to the "business activation determination" process of step S41. The processing of steps S39 and S40 described above is executed by the control unit 133 of the processing time prediction processing unit 13.

In the "business activation determination" processing of step S40, should the newly input business be activated by comparing the latest processing end time from the processing end time of the business processing time management table 163 with the prescribed time of the requirement? Determine whether or not. This determination is performed by the business activation determination processing unit 14, obtains all business end times from the business processing time management table 163, determines the latest time to end, and compares it with the prescribed time of the requirement. , Whether or not the new input work can be started is determined. If it is determined to start, the business activation processing unit 15 is activated to activate the newly input business (step S42), and if it is determined not to activate, the business processing time management table 163 is changed to a table related to the new business. It deletes and rolls back to the remaining work management table 164 (step S43).

Here, a supplementary explanation will be given on the setting and registration of the weighting factors in the weighting factor table 162. FIG. 14 shows an operation concept for setting / registration of the work load coefficient, and FIG. 15 shows an operation concept for setting / registration of the multiplicity weight coefficient.

The weighting coefficient table 162A is set and registered as a weighting coefficient by measuring and analyzing the correlation with the workload, the processing unit, and the processing time. FIG. 14 shows a weighting factor setting example exemplifying the processing unit A2. Here, first, the processing time at the base workload is measured (step S141). Note that the processing for performing the processing unit A2 is a table update, and the processing time when the workload is 100 (the number of tables) is 1 minute. The workload is, for example, a numerical expression of the processing performed by the processing unit, such as the number of records to be updated and the number of lines to be drawn. next,
This work amount is changed and each processing time is measured (step S142). Here, when the measurement result is 200 workload, 1 minute 12 seconds, and when the workload is 300, 1 minute 48.
It is assumed that the time is 2 minutes and 24 seconds when the workload is 400. Then, it is approximated by an arithmetic expression that satisfies the relationship between the workload and the measured processing time (step S143). Here, the processing time is 36 + 24 * V / 100 [seconds]. Similarly, the relationship between the workload and the processing time is calculated for all processing units. The coefficient derived from the above result is set and registered in the weight coefficient table 162A.

The weighting coefficient table 162B calculates a running time by performing a multiple running test in the same work or a mixture of works of a plurality of types and measuring the start time and the end time for each processing unit at different multiplicities. Business type,
The value is set and registered by analyzing the correlation between the processing unit type and the multiplicity with respect to the processing time. In Figure 15,
An example of weighting factor setting in the case of two multiplexing of processing units A2 and B1 is illustrated. In FIG. 15, first, the processing time when the A2 is traveling alone is measured (S151). Here, it is assumed that it was one minute. Next, the processing time is measured by multiple running with other processing units (S152). Here, it is assumed that it is 1 minute and 12 seconds when the vehicle is run in a multiplex with B1.
Then, the stretching rate of the processing time is obtained (step S153), and the stretching rate of A2 when the processing unit B1 and the processing unit B2 are doubled is 1 minute 12 seconds / 1 minute = 1.25. This value is written in the matrix intersection of A2 and B1 of the weight coefficient table 162B (step S154).

As described above, the business type, the processing unit type, the multiplicity J, and the business volume V with respect to the business processing time.
Multivariate analysis is performed on the correlation of and the weighting factor is calculated. Specifically, the same job is subdivided into processing units with characteristics (the degree of influence on processing time differs from other parts due to changes in multiplicity), and in the environment where the same job or jobs of multiple types coexist, Perform a multi-run test and measure the start time and end time for each processing unit at different multiplicities to calculate the run time.
Then, the weight is obtained by analyzing the correlation between the type of work, the type of processing unit, and the processing time of the multiplicity J. Further, separately, the correlation between the work volume V, the processing unit, and the processing time is measured and analyzed, and is used as the weight. Finally, the weights derived from the above results are set in the weight coefficient table 16
Set to 2 and register.

The time of each processing unit is linked to the change of the multiplicity by using the weight calculated by the multivariate analysis, and a more accurate processing time is predicted by the following arithmetic expression. The basic idea of processing time is based on the following arithmetic expressions (1), (2), and (3).

[Equation 1] Here, T is the total processing time of the business, Ttask (I) is the total processing time of the processing unit I, I is the number of processing units included in the business,
Ttask (I) (J) is the processing time when the processing unit I is J multiplex, J is the multiplicity, V is the workload, N is the maximum multiplicity,
Let T be the time width that is the multiplicity J.

As described above, the present invention relates to a newly introduced job including an existing job when predicting the processing time of the job including the online real-time processing and the batch processing in a multi-running environment, and has a characteristic processing. It is subdivided into units and is calculated by using the type of work, the type of processing unit, multiplicity, and the correlation between the amount of work and processing time as a weight. Is linked in real time, and is estimated by the total processing time in which stretching due to multiplexing is added for each multiplicity. In addition, in a multi-running environment, the dynamically changing multiplicity of existing jobs and newly entered jobs is reflected in real time, and whether or not to start a new job is determined from the prediction results based on the final end time. be able to. This makes it possible to more accurately predict the processing time taking into consideration the influence of the processing time extension due to the multiplexing. Note that, according to the above-described embodiment of the present invention, only the case of two-multiplexing has been described for ease of description, but the case of n (where n is an integer of 2 or more) multiplexing is also described. The same application is possible by preparing a combination of the above described multiplicity weighting tables.

The business queuing unit 11, the management target business selection processing unit 12, the processing time prediction processing unit 13, the business activation determination processing unit 14, and the business activation processing unit 15 in FIG.
Each of the programs for realizing the functions of the computer is recorded on a computer-readable recording medium, and the program recorded on the recording medium is read by a computer system that constitutes an information system, and the computer executes the program. By sequentially reading and executing, the business activation determining apparatus of the information system in the multiplexing environment of the present invention is constructed. The computer system mentioned here includes an OS and hardware such as peripheral devices.

Further, the computer system also includes a homepage providing environment (or display environment) if the WWW system is used. Further, the computer-readable recording medium refers to a portable medium such as a flexible disk, a magneto-optical disk, a ROM, a CD-ROM, or a storage device such as a hard disk built in a computer system. Further, the computer-readable recording medium is, for example, a volatile memory (RAM) inside a computer system which serves as a server or a client when a program is transmitted via a network such as the Internet or a communication line such as a telephone line.
Those that hold the program for a certain period of time are also included.

Further, the above program may be transmitted from a computer system in which the program is stored in a storage device or the like to another computer system via a transmission medium or by a transmission wave in the transmission medium. Here, the "transmission medium" for transmitting the program refers to a medium having a function of transmitting information, such as a network (communication network) such as the Internet or a communication line (communication line) such as a telephone line. Further, the program may be a program for realizing some of the functions described above. further,
It may be a so-called difference file (difference program) that can realize the above-mentioned functions in combination with a program already recorded in the computer system.

Although the embodiment of the present invention has been described in detail above with reference to the drawings, the specific configuration is not limited to this embodiment, and includes a design etc. within the scope not departing from the gist of the present invention. Be done.

[0046]

As described above, according to the present invention, regarding one work, it is subdivided into individual processing units having characteristics, the processing time is estimated by using it as the minimum unit, and the total is taken as the total processing time of the work. Since the estimation of the processing end time of each processing unit is reflected in the change of the multiplicity, it is possible to make a prediction by adding a weight to the processing time for each multiplicity within one processing unit. In addition, the processing time is calculated by reflecting the influence of the change in multiplicity that occurs when a new job is input, not only on the new job but also on the existing job, and the end time of the entire job in the system includes the existing job. Since it is judged based on the final end time of the newly input work, it was applied to the blocking system where the above-mentioned end time is strictly required by ending all the started work by the specified end time of the requirement. In this case, not only serious troubles such as system stop can be avoided, but also the pre-blocking process of the work can be optimized and the system resources can be effectively used.

The coefficients set and registered in the weighting coefficient table are subjected to a multiple running test in the same work or a mixture of works of a plurality of types, and the start time and the end time for each processing unit at different multiplicities are determined. It is assumed that it is obtained by measuring and calculating the traveling time and analyzing the correlation between the business type, the processing unit type, and the processing time of the multiplicity. By using this weighting coefficient, the information system under the multi-driving environment can be obtained. In processing time prediction in, it is possible to reflect the dynamically changing multiplicity in real time for both existing jobs and new jobs in the system.
It is possible to more accurately predict the processing time in consideration of the influence of the processing time stretching due to the multiplexing.

[Brief description of drawings]

FIG. 1 is a block diagram showing a functional expansion of an internal configuration of a business activation determination device of an information system under a multiplexing environment of the present invention.

FIG. 2 is a block diagram showing functional development of an internal configuration of a processing time prediction processing unit shown in FIG.

FIG. 3 is a diagram showing a data structure of various tables that constitute the database shown in FIG.

FIG. 4 is a flowchart representing a procedure for realizing a business processing time prediction method for an information system in a multiple environment according to the present invention.

FIG. 5 is an operation conceptual diagram quoted for explaining details of an operation procedure for each step corresponding to each step of the flowchart shown in FIG. 4;

FIG. 6 is an operation conceptual diagram quoted for explaining details of an operation procedure for each step corresponding to each step of the flowchart shown in FIG. 4.

FIG. 7 is an operation conceptual diagram quoted for explaining details of an operation procedure of each step corresponding to each step of the flowchart shown in FIG. 4;

FIG. 8 is an operation conceptual diagram cited for explaining the details of the operation procedure for each step corresponding to each step of the flowchart shown in FIG.

FIG. 9 is an operation conceptual diagram cited for explaining details of an operation procedure for each step corresponding to each step of the flowchart shown in FIG. 4.

FIG. 10 is an operation conceptual diagram cited for explaining details of the operation procedure for each step corresponding to each step of the flowchart shown in FIG. 4.

FIG. 11 is an operation conceptual diagram quoted for explaining details of an operation procedure for each step corresponding to each step of the flowchart shown in FIG. 4.

FIG. 12 is an operation conceptual diagram cited for explaining details of the operation procedure of each step corresponding to each step of the flowchart shown in FIG. 4;

FIG. 13 is an operation conceptual diagram cited for explaining details of an operation procedure for each step corresponding to each step of the flowchart shown in FIG. 4.

FIG. 14 is a diagram cited for explaining setting registration of coefficients in the weighting coefficient table shown in FIG.

FIG. 15 is a diagram quoted for explaining setting registration of coefficients in the weighting coefficient table shown in FIG. 3;

[Explanation of symbols]

1 ... Business activation determination device (server), 2 ... Network,
11 ... Business queuing unit, 12 ... Managed work selection processing unit, 13 ... Processing time prediction processing unit, 14 ... Business activation determination processing unit, 15 ... Business activation processing unit, 16 ... Database (DB), 131 ... First Arithmetic unit, 132 ... Second arithmetic unit,
133 ... Control unit, 161, ... Business information table, 162 ...
Weighting coefficient table, 163 ... Business processing time management table, 164 ... Remaining business management table

   ─────────────────────────────────────────────────── ─── Continued front page    (72) Inventor Xu             1-9-1, Konan, Minato-ku, Tokyo             IT Comware Co., Ltd. (72) Inventor Takashi Yasumoto             1-9-1, Konan, Minato-ku, Tokyo             IT Comware Co., Ltd. (72) Inventor Toshitaka Sugimoto             1-9-1, Konan, Minato-ku, Tokyo             IT Comware Co., Ltd. (72) Inventor Masanori Maeda             1-9-1, Konan, Minato-ku, Tokyo             IT Comware Co., Ltd. F-term (reference) 5B042 GA23 HH07 HH20 MC33

Claims (13)

[Claims] 1. A business of an information system in a multiplexing environment in which an existing business including a newly input business is subdivided for each processing unit, a processing time is estimated for each processing unit, and the total is predicted as a total processing time of the business. It is a start-up determination device, and for each processing unit of the newly input work, acquires the processing start time and the multiplicity information at the start time from the work processing time management table, and considers the overlap with the existing work. Then, referring to the weighting coefficient table, the processing time is reflected in the calculation, and the processing time and the end time are written in the business processing time management table, and the multiplicity is changed by receiving a new business. For existing work to be performed, multiplicity information is acquired from the work processing time management table, and considering the overlap with new work, the weighting coefficient table is referred to and new data is added. By newly calculating the processing time and writing the processing time and the end time in the business processing time management table, and the calculation by the first and second calculating means in time series, the newly input business And a control means for dynamically reflecting the change of the multiplicity with respect to the existing business and writing the calculation result in the business processing time management table. . 2. The end times of the existing work and the new input work are acquired from the work processing time management table, and the final end time among them is compared with the prescribed time of the end requirement to determine whether to start the new input work. The business activation determination device for an information system in a multiplexing environment according to claim 1, further comprising business activation determination processing means for determining. 3. The weight coefficient table is subjected to a multi-run test in the same work or a mixture of works of a plurality of types, and the run time is measured by measuring a start time and an end time for each processing unit at different multiplicities. Is calculated,
The first weighting coefficient table that is set and registered as a weighting coefficient by analyzing the correlation between the business type, the processing unit type, and the processing time of the multiplicity, and the correlation between the business volume, the processing unit, and the processing time is measured. 2. A business activation discriminating apparatus for an information system in a multiplexing environment according to claim 1, further comprising: a second weighting coefficient table that is set and registered as a weighting coefficient by analysis. 4. A business of an information system under a multiplexing environment in which an existing business including a newly input business is subdivided for each processing unit, a processing time is estimated for each processing unit, and the total is predicted as a total processing time of the business. A startup determination device, which identifies a business queuing unit that receives a business processing request including the newly submitted business as a transaction and accumulates the business processing request, and a type of the business that is started next in the queued business. , Refer to the weighting factor table for each processing unit for the management target business selection processing means that determines whether or not the business is a management target business by collating with the business information table, and for existing business including newly input business By dynamically reflecting the change in multiplicity, calculate the processing time for each task, processing unit, and multiplicity, and then predict the total as total processing time Processing time estimation processing means, a business activation determination processing means for determining whether or not to input the new input business based on the estimation result of the processing time, and a new activation processing determined by the business activation determination processing means. A business activation determination device for an information system in a multiplex environment, comprising: a business activation processing means for activating an input transaction. 5. A business of an information system under a multiplexing environment in which an existing business including a newly input business is subdivided for each processing unit, a processing time is estimated for each processing unit, and the total is predicted as a total processing time of the business. A method of predicting a processing time, wherein the processing time for each processing unit is estimated by dynamically reflecting a change in the processing load by adding a weight to the processing time for each processing unit within a single processing unit. The processing time is calculated by reflecting the change in the multiplicity that occurs when a work is input to the existing work including the new work, and the processing time of the entire work is the total processing time of the existing work including the new work. A method for predicting a business processing time of an information system in a multi-environment, characterized in that the final end time of a running business is judged. 6. A business of an information system under a multiplexing environment in which an existing business including a newly input business is subdivided for each processing unit, a processing time is estimated for each processing unit, and the total is predicted as a total processing time of the business. A business processing time prediction program used for a start-up determination device, wherein, for each processing unit of the new input business, processing start time and multiplicity information at the start point are acquired from a business processing time management table, Considering the overlap with the work, refer to the weighting coefficient table, reflect it in the calculation of the processing time, and write the processing time and the end time in the business processing time management table. For the existing business whose multiplicity changes by receiving it, after acquiring multiplicity information from the business processing time management table and considering the overlap with the new business, A second step of calculating a new processing time by referring to the weighting coefficient table and writing the processing time and the end time in the business processing time management table, and the first step and the second step in time series A business processing time prediction program for causing a computer to execute a third step of dynamically reflecting changes in multiplicity with respect to newly input business and existing business by repeating, and writing the calculation result in the business processing time management table. . 7. The end times of the existing work and the new input work are acquired from the work processing time management table, and the final end time among them is compared with the specified time of the end requirement to determine whether to start the new input work. The business processing time prediction program according to claim 6, further comprising a fourth step of making a judgment, and causing a computer to execute the first to fourth steps. 8. The first step is a step of acquiring multiplicity information of a processing unit to be calculated from the start and end times of processing of the business processing time management table, and is a calculation target. Dividing the processing unit into parts having different multiplicities, and calculating the processing time by the weighting coefficient and the multiplex time obtained by referring to the first weighting coefficient table for each of the divided parts; 8. The business processing time prediction program according to claim 7, further comprising the step of updating the business processing time management table based on the above, and causing a computer to execute each of the steps. 9. The second step is a step of acquiring multiplicity information of a processing unit to be a calculation target from the start and end times of processing of the business processing time management table, and is a calculation target. Dividing the processing unit into parts with different multiplicities, and recalculating the processing time by the weighting factor and the multiplex time obtained by referring to the first weighting factor table for each of the divided parts; 8. The business processing time prediction program according to claim 7, further comprising the step of updating the business processing time management table on the basis of the above. 10. The third step tentatively calculates a processing time during independent traveling according to a work amount by referring to a second weighting coefficient table for each processing unit, and updates the work processing time management table. And a step of comparing the end time of the processing unit to be calculated with the processing time of the entire existing work to determine the calculation order of the processing units, according to the calculation order, for each new work processing unit or for the existing work Instruct to calculate the processing time for each processing unit, refer to the updated business processing time management table, and determine whether there is another overlapping existing business from the start and end times of the processing. Then, the steps of repeating the steps described above are performed until the existing work is exhausted, and the work information table is searched to determine whether there is a subsequent processing unit to be calculated. Wherein the step of repeating the processing of the respective steps until it is determined that the connection processing unit is not, the, job processing time prediction program according to claim 7 to execute the respective steps in a computer. 11. An existing job including a newly input job is subdivided for each processing unit, and a processing time is estimated for each processing unit,
A business processing time prediction program used in a business activation determination device of an information system in a multiplexed environment for predicting the total as a total processing time of a business, which receives a business processing request including the newly input business as a transaction, A step of accumulating, a step of identifying the type of a task to be started next among the queued tasks, and a step of judging whether or not the task is a managed task by collating with the task information table; For existing operations including input operations, the change in multiplicity is dynamically reflected by referring to the weighting coefficient table for each processing unit, and the processing time for each operation, processing unit, and multiplicity is calculated. Predicting the total as the total processing time, and determining whether or not to input the new input operation based on the processing time prediction result. A business processing time prediction program for causing a computer to execute the step of starting a new input business that is determined to be started by the business start determination processing means. 12. An existing business including a newly input business is subdivided for each processing unit, and a processing time is estimated for each processing unit,
A computer-readable recording medium recording a business processing time prediction program used in a business activation determination device of an information system in a multiplexing environment for predicting the total as a total business processing time, the business processing time prediction program For each processing unit of the newly input work, the processing start time and the multiplicity information at the start point are acquired from the work processing time management table, and the overlapping with the existing work is taken into consideration. The first step of reflecting the processing time calculation in the calculation of the processing time and writing the processing time and the end time in the business processing time management table, and the existing business whose multiplicity changes by receiving a new business , The weighting factor table after obtaining the multiplicity information from the business processing time management table and considering the overlap with the new business. By calculating a new processing time with reference and writing the processing time and the end time in the business processing time management table, by repeating the first step and the second step in time series, A record that records a business processing time prediction program for causing a computer to execute the third step of dynamically reflecting changes in multiplicity with respect to newly input business and existing business and writing the calculation result in the business processing time management table. Medium. 13. An existing business including a newly input business is subdivided for each processing unit, and a processing time is estimated for each processing unit,
A computer-readable recording medium recording a business processing time prediction program used in a business activation determination device of an information system in a multiplexing environment for predicting the total as a total business processing time, the business processing time prediction program Receiving a business processing request including the newly input business as a transaction and accumulating it, and identifying the type of the business to be started next among the queued businesses, and collating with the business information table. To determine whether or not it is a management target job, and for existing jobs, including newly submitted jobs, dynamically reflect changes in multiplicity by referring to the weighting factor table for each processing unit. , The task, the processing unit, and the processing time for each multiplicity, and then the total is predicted as the total processing time. Computer, a step of determining whether or not to input the new job input based on the prediction result of the processing time, and a step of activating the new job input determined to be started by the job start determination processing means. A recording medium on which a business processing time prediction program to be executed by a computer is recorded.