JP2002268879A - A program design support device, a program design support method, and a program for causing a computer to execute the program design support method. - Google Patents

Abstract

(57) [Summary] [PROBLEMS] To predict the execution time of each action included in a program and improve the development efficiency and quality of the program. SOLUTION: A state transition editor unit 1 describes state transition information of a program, and a state transition time prediction comparing unit 4
Of the actions described in the state transition information, for the action for which the time (execution limit time) allowed to execute the process is specified, the action recorded in the program component information database 3 is used to execute the process. The required time (execution prediction time) is predicted, the predicted execution prediction time is compared with the execution limit time, and the comparison result is presented.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for supporting specification design related to state transition in program design.
In particular, the present invention relates to a program design support technology for supporting a design work related to a processing time of a program required for state transition.

[0002]

2. Description of the Related Art With the rapid development of information processing technology and the expansion of application fields, the scale and functions of programs to be developed are generally rapidly expanding. In response to this situation, various technologies are being developed to improve program development efficiency and quality. For example, in a control system, communication system, or embedded system program, the operation specifications of a program are often defined as a plurality of states and transitions between the states. A method has been developed to support the design specification of a program by describing it in a way that is described in a specific way.

FIG. 10 is a description example of a state transition described in Japanese Patent Application Laid-Open No. 7-78076 “Automatic Program Generator”. This method is a description method widely known as a state transition diagram, and includes a plurality of states, transitions between states, events,
It contains four elements called actions. An event is an event that occurs in the system, and an action is a process performed when a transition from one state to another state occurs due to the occurrence of the event. In this description method, states are circles, transitions between states are arrows, and events and actions are described in the form of “event / action” on both sides of “/” accompanying the transition. In the example of FIG. SN1 and SN2 represent different states, respectively, an arrow from SN1 to SN2 represents a transition from state SN1 to state SN2, en1 represents an event that triggers the transition to occur, and a
n1 represents an action performed at the time of the transition.

For the formal description of the state transition, a description method such as a state chart or a state transition table in addition to a state transition diagram is widely used. The state chart is UML (Un
This is a method for describing state transitions defined in the "form modeling language", and has a feature that, compared with a state transition diagram, hierarchical state transitions and parallel state transitions can be easily described. However, it basically uses a description method using figures similar to state transition diagrams,
The state is described as a rectangle, the transition is described as an arrow between the states, and the event and the action are described as notes to the transition. on the other hand,
The state transition table is a description method in a table format, in which a state is enumerated in a column (horizontal direction), an event is enumerated in a row (vertical direction), and each cell in the table which is an intersection of a row and a column is represented by a corresponding column. As the transition that occurs in response to the event indicated by the corresponding row in the indicated state, the state and action of the transition destination at that time are described in each of these cells. FIG. 11 shows an example of the description of the state transition using the state transition table. In this example, the state SN1
In the transition triggered by the event en1, the action an1 is performed and the state transits to the state SN2.
This description is equivalent to a description example using the state transition diagram shown in FIG.

As described above, since the operation of a control system, a communication system, or an embedded system can often be described by state transitions, in software development in these fields, the state transition described above is performed at the design stage. It is widely used to use formal descriptions of state transitions using diagrams, statecharts, state transition tables, or other methods with equivalent description capabilities. By adopting these formal description methods, it becomes possible to clearly describe the operation specifications of the program and to easily verify its accuracy, thereby improving the development efficiency and quality of the program. Such a formal description method of state transition is used not only in the above-mentioned field but also in, for example, design of a program for realizing a user interface of a device. The operation of the user interface can be represented as a state transition by using the user's operation on the device as an event, and the formal description method of the state transition also improves program development efficiency and quality. That's why.

In a control system, a communication system, or an embedded system, the time required for the processing is limited.
That is, a real-time operation is usually required. Also, with regard to the user interface of the device, in order to realize comfortable operability for the user, there are many cases where restrictions are imposed on the time required from the operation of the user to the response in designing the program.

However, whether or not the developed system can operate while satisfying the time constraint is generally determined only by actually measuring when the program under development starts to operate. In this case, first run the program and measure whether it meets the design time constraints, and if there is a problem, analyze its internal operation and remove the problem that caused the time over, or change the operation logic. The work of shortening the processing time by improving the quality is repeated to finally satisfy the design time constraint, or to minimize the problem. In the development of embedded software, instead of actually measuring the execution time of the program, the execution time of the program is obtained by virtually operating the program using an H / W simulator or the like.
Although the operation time of the program may be estimated by accumulating the execution time of the U instruction, the operation itself for improving the operation time of the program is the same as the operation procedure described above.

For this reason, the achievement status of the constraint on the operation time of the program determined at the design stage, that is, the performance quality is
It cannot be verified unless the source code is created and it is not a downstream process of program development that makes the program operable.If there is a problem, the rework is large and the development process is likely to be delayed. is there. Further, in the verification and improvement work, the measurement work needs to be repeated, so that the number of man-hours is increased, which also tends to cause a delay in the development process.

[0009] In general, it is rare to formally describe constraints on the operation time of a program at the program design stage, and usually, separately from design information on the logic of the program, a comment attached to the program design information is separately provided. Often described in natural language. For example, in any of the formal description methods related to the state transition described above, the description of the execution time permitted for the processing of the action, that is, the description of the allowable execution time (hereinafter, also referred to as execution limit time) is not formalized. For this reason, it is difficult to handle the design information on the state transition operation and the design information on the execution time limit in cooperation with each other, and the design information on the execution time limit is easily overlooked. There is such a problem.

[0010]

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problems of the prior art. In designing a state transition, a time limit (execution time) allowed for processing of the action is considered. Time limit) at the same time,
Program design that estimates the achievement status, that is, performance quality, at the design stage, and presents information for countermeasures, thereby reducing defects that occur in the downstream process of software development and improving program development efficiency and quality. The purpose is to obtain a support system.

[0011]

According to the present invention, there is provided a program design support apparatus for executing a program including a designated processing step designated as a time allowed for execution of a process as an allowed execution time. An execution time prediction comparing unit for estimating a required time to derive an execution prediction time, and comparing the execution prediction time of the designated processing step with the permissible execution time, and a comparison result presenting a comparison result by the execution time prediction comparing unit And an output unit.

The execution time prediction comparing section inputs state transition information indicating a state transition in the program and a processing step executed at the time of the state transition, and inputs the state transition information to the input state transition information. It is characterized in that an estimated execution time of the designated processing step among the indicated processing steps is derived.

The designated processing step includes at least one or more internal processing steps, and the execution time prediction comparing section predicts a time required for executing each internal processing step and calculates an estimated execution time of each internal processing step. It is characterized in that a total value of predicted execution times of the respective internal processing steps is compared with a permissible execution time of the designated processing step.

The designation processing step includes, as the internal processing step, an execution process of a program part. The program design support apparatus includes: an average execution time information indicating an average execution time of the program part; A program component information database that stores at least one of the maximum execution time information indicated by the maximum execution time of the execution process, wherein the execution time prediction comparison unit stores the program component information database. Based on at least one of the average execution time information and the maximum execution time information, a time required for the execution process of the program component is predicted to derive a predicted execution time of the execution process of the program component. .

[0015] The execution time prediction comparing section determines that the result of the comparison between the predicted execution time of the designated processing step and the permissible execution time indicates that the predicted execution time of the specified processing step exceeds a predetermined level of the permissible execution time. Instructs the comparison result output unit to present warning information about the specified processing step, and based on the instruction from the execution time prediction comparison unit, the comparison result output unit Is presented.

The execution time prediction comparing section derives, for the specified processing step, an occurrence probability that an execution prediction time exceeding a predetermined level of an allowable execution time will occur, and determines that the derived occurrence probability is equal to or greater than a predetermined value. In this case, the comparison result output unit is instructed to present warning information about the designated processing step, and the comparison result output unit is configured to execute the designated process based on the instruction from the execution time prediction comparison unit. It is characterized by presenting warning information on the process.

[0017] The execution time prediction comparing section derives the occurrence probability using at least one of histogram information and probability distribution information.

The designation processing step includes a plurality of internal processing steps, the execution time prediction comparing section performs an ordering among the plurality of internal processing steps, and the comparison result output section includes the warning information together with the warning information. A plurality of internal processing steps are presented according to the ordering.

The designation processing step includes, as the internal processing step, an execution process of a plurality of program parts, and the execution time prediction comparing section orders the execution processes of the plurality of program parts, and The output unit presents the execution process of the plurality of program components in accordance with the order, together with the warning information.

[0020] The execution time prediction comparing section performs an ordering among the execution processes of the plurality of program components in accordance with a magnitude of an influence on an execution prediction time of the designated processing step.

The execution time prediction comparing section compares a predicted execution time of each of the designated processing steps with an allowable execution time for a program including a plurality of designated processing steps. Calculate at least one of the difference from the permissible execution time and the ratio of the predicted execution time to the permissible execution time, and order the designated processing steps based on at least one of the calculated difference and the ratio. The comparison result output unit performs
The designated processing steps are presented in accordance with the ordering by the execution time prediction comparing unit.

A program design support method according to the present invention
For a program including a designated processing step designated as a permissible execution time, a time required for execution of the designated processing step is predicted to derive a predicted execution time, and the execution of the designated processing step is performed. It has an execution time prediction comparison step of comparing the prediction time with the permissible execution time, and a comparison result output step of presenting a comparison result by the execution time prediction comparison step.

In the execution time prediction comparing step, state transition information indicating a state transition in the program and a processing step executed at the time of the state transition is input, and the input state transition information is input to the input state transition information. It is characterized in that an estimated execution time of the designated processing step among the indicated processing steps is derived.

The designated processing step includes at least one or more internal processing steps, and the execution time prediction comparison step predicts a time required for execution of each internal processing step and calculates an estimated execution time of each internal processing step. It is characterized in that a total value of predicted execution times of the respective internal processing steps is compared with a permissible execution time of the designated processing step.

[0025] The designation processing step includes, as the internal processing step, a program part execution process, wherein the program design support method comprises: an average execution time information indicating an average execution time of the program part; A program component information storing step of storing at least one of the maximum execution time information indicating the maximum execution time of the execution process, and the execution time prediction comparing step is stored by the program component information storing step. And estimating a time required for the execution process of the program component based on at least one of the average execution time information and the maximum execution time information to derive a predicted execution time of the execution process of the program component. And

In the execution time prediction comparing step, when it is determined that the predicted execution time of the specified processing step exceeds a predetermined level of the allowable execution time as a result of the comparison between the predicted execution time of the designated processing step and the allowable execution time. Instructing the comparison result output step to present warning information about the designated processing step, wherein the comparison result output step is based on the instruction from the execution time prediction comparison step, Is presented.

The execution time prediction comparing step derives an occurrence probability that an execution prediction time exceeding a predetermined level of an allowable execution time will occur for the specified processing step, and determines that the derived occurrence probability is equal to or more than a predetermined value. In this case, the comparison result output step is instructed to present warning information about the designation processing step, and the comparison result output step is based on the instruction from the execution time prediction comparison step. It is characterized by presenting warning information on the process.

The designation processing step includes a plurality of internal processing steps, the execution time prediction comparison step performs an ordering among the plurality of internal processing steps, and the comparison result output step includes the warning information together with the warning information. A plurality of internal processing steps are presented according to the ordering.

The designation processing step includes, as the internal processing step, execution processing of a plurality of program parts. The execution time prediction comparison step orders the execution processing of the plurality of program parts, and executes the comparison result. The output step presents the execution process of the plurality of program components in accordance with the order, together with the warning information.

A program for causing a computer to execute the program design support method according to the present invention includes a program including a designated processing step in which a time allowed for execution of a process is designated as an allowed execution time. An execution time prediction comparing step of deriving an execution prediction time by estimating a time required for execution and comparing the execution prediction time of the designated processing step with an allowable execution time; and presenting a comparison result by the execution time prediction comparison step. A program for causing a computer to execute a program design support method characterized by including a comparison result output step.

[0031]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiment 1 FIG. 1 is a configuration diagram showing a configuration of a program design support device 10 according to the first embodiment of the present invention. As shown in FIG. 1, a program design support apparatus 10 according to the present embodiment includes a state transition editor unit 1, a state transition information holding unit 2, a program component information database 3, a state transition time prediction comparison unit (execution time prediction comparison unit). 4). In addition, these 1-4
Each of the functions may be a computer program for causing a computer to execute the functions.

In this embodiment, the program developer uses the state transition editor unit 1 to generate a plurality of states, a plurality of events, and the occurrence of an event in each state as information relating to the state transition in the operation specifications of the program. Describes the transitions between the states caused by, and the actions (treatment steps) executed in each transition. In the description of the action, a program flow such as which program component is to be called and in what procedure, and an allowable execution time (also referred to as an execution limit time) allowed for the processing are described as necessary. When describing the program flow,
Mainly for the convenience of the program developer, the program component information database 3 can be accessed, the name of the program component to be used and the program interface can be searched and used to describe the program flow.

FIGS. 2 and 3 show examples of state transition information described by the state transition editor unit 1 in this embodiment. FIG. 2 is a description example of a state transition by the state transition editor unit 1 according to the present embodiment. In the present embodiment, in the state transition, the state 11 is represented by a circle and the transition 12 between the states is represented by an arrow in the form of a state transition diagram. Also, an event that triggers the transition 12, an action to be executed in the transition 12,
And the action execution time limit is "/"
Is described in connection with the transition 12 in a format such as “event / action / execution time limit”. Here, an action (processing step) in which the execution time limit is specified is referred to as a specified processing step. If there is no need to define the execution time limit, "Event / Action /
− ”Or“ event / action ”. In the description example shown in FIG. 2, there are three states 11, SN1, SN2, and SN3, and the transition 12 from the state SN1 to the state SN2 is "enA / an12 / t12", that is, a transition triggered by the event enA. , Action an12
And the execution time limit at that time is t12.

FIG. 3 is a description example of a program flow of an action by the state transition editor unit 1 in the present embodiment. In the description example shown in FIG. 2, the process S10 of the action an12 is described. The process S10 first executes the program component p1 (S11), and then checks the execution result of p1 (S12). If the result is true (S
12 (YES), the program component p2 is executed, and the process ends (S13). If the execution result of p1 is not true, the process is immediately terminated (NO in S12).

The contents described in the state transition editor unit 1 are stored in the state transition information holding unit 2. Also, the contents held by the state transition information holding unit 2 can be revised again by the program developer using the state transition editor unit 1 and saved again. The state transition time prediction / comparison unit 4 checks the contents of the state transition information held in the state transition information holding unit 2 and determines the program flow of the action for each action (designated processing step) for which the execution time limit is defined. By analyzing, it is possible to know which program part is to be called and in what procedure, and further, by accessing the program part information database 3 and obtaining information on the average execution time of each program part, predicting the time required for action processing. A value (expected execution time) is obtained. If there is an action that exceeds the defined execution time limit, the state transition editor unit 1 is used to inform the program developer of the estimated execution time of the action. A table that warns the display of the corresponding action to indicate that the execution time limit in the design is not met. Instead of the presented. In this case, the state transition editor unit 1 functions as a comparison result output unit that presents a result of comparison between the predicted execution time and the execution time limit by the state transition time prediction comparison unit 4.

A method of estimating the time required for processing an action in the state transition time estimator / comparator 4 in the present embodiment will be described with reference to FIG. FIG. 4 is a flowchart showing a processing procedure for estimating the execution time. In this process, it is assumed that the action program flow is structured. The execution time predicting process S20 first sets a predicted value of a processing target to be obtained as an initialization process to 0 once (S21), and then executes each step from the beginning to the end of the program flow to be processed. The subsequent processing is repeated (S22, S32
while).

In the repetition, first, the processing contents of the target step are checked (S23). If the processing is the calling processing of the program parts (execution processing of the program parts), the program parts information database 3 is searched. The average execution time of the program component is obtained and added to the predicted value of the processing target (S24). If the processing is a basic processing such as an arithmetic operation or an assignment, the execution time of the step corresponding to the processing content is estimated based on the execution time of the CPU required for each of the processings such as the arithmetic operation or the assignment.
It is added to the predicted value to be processed (S25).

On the other hand, if the processing of the target step is a forward branch processing corresponding to an if statement or a switch statement, a partial program flow consisting of a series of steps until a branch merges again for each branch. By performing the execution time prediction process S20 recursively as a target, the execution prediction time for each branch process is obtained (S2
6, S27, S28), the maximum value of the predicted execution time for each branch is regarded as the predicted value of the processing time required from the branch to the merge, and is added to the predicted value of the processing target (S29).
The processing of the target step is a while statement or fo
In the case of the repetition processing corresponding to the r statement, the execution time prediction processing S20 is recursively performed on the partial program flow including all the steps of the internal processing to be repeated to obtain the execution prediction time relating to the repeated internal processing ( S
30), the value is multiplied by the number of repetitions of the internal processing, and the result is added to the predicted value to be processed as the predicted execution time of the entire repetition processing (S31). Then, after performing the processing for each step from the beginning to the end of the program flow to be processed (S32), the predicted value of the processing target is returned as the predicted execution time (S33).

In step S24, if the processing can be stopped inside the program component, it is determined that the processing time cannot be predicted, and the execution time prediction processing S20 is abnormally terminated, or a program developer is inquired of a substitute execution value. By doing so, the process is continued. Similarly, in step S31, if the number of repetitions of the internal processing is not determined, the execution time prediction processing S20 is abnormally terminated assuming that the processing time cannot be predicted, or the processing is continued by inquiring the program developer for a substitute value. Let it.

Since the program design support apparatus according to the present embodiment operates as described above, it is possible to simultaneously describe the state transition and the time limit allowed for the processing of the action when designing the state transition. Further, by estimating the time required for the processing of the action, it is possible to estimate whether the time limit can be achieved. The effect of improving program development efficiency and quality can be obtained.

In the present embodiment, the execution time limit for an action is described as “event / action / execution time limit” associated with a transition. A method may be adopted in which only “event / action” is described, and the execution time limit is described together with the program flow for the action. Further, in the present embodiment, the description of the state transition is performed in accordance with the format of the state transition diagram, but this is performed by adopting another formal description method of the state transition, for example, a form such as a state chart or a state transition table. Has the same effect. Similarly, in the present embodiment, the processing of the action is described as a program flow.
The same effect can be obtained by adopting another graphic description method such as P or a description method in a text format such as a program source code.

In this embodiment, as the average execution time for each program component held in the program component information database 3, a value separately measured at the time of development of each program component can be used. Alternatively, a value predicted by performing the processing time prediction processing S20 in the present embodiment based on the program flow of each program component may be used. Alternatively, when a program component is reused, a value measured during past use may be used. Furthermore, when the operation of the entire program is possible, instead of the value measured for each program component, the measured value of the execution time of each program component obtained when the entire program is operated may be used. In this case, there is an effect that the possibility of predicting a value closer to the actual operation as the predicted execution time for the action is increased, and the accuracy of predicting whether the execution time limit is achievable is increased. Further, by using the maximum execution time instead of the average execution time of the program components, it is possible to compare the execution limit time and the execution prediction time based on a stricter predicted value (worst case prediction). If the program execution performance of the system used to measure the execution time of the program component is not the same as the program execution performance of the system that defines the execution time limit, a comparison between the estimated execution time and the execution time limit is made. The comparison can be performed significantly by considering the difference.

When the execution time limit is compared with the execution time limit, a warning is not provided when the estimated execution time exceeds the execution time limit, but a value obtained by multiplying the execution time limit by a certain ratio is used. For example, a warning may be presented when the execution time exceeds 80%, and a warning is presented when the execution time exceeds 80% and a warning when the execution time exceeds the execution time. By giving a difference in the presentation method as described above, it is possible to suggest the priority of the countermeasure to the program developer, and it is possible to improve the work efficiency of the countermeasure. Furthermore, for each action for which the execution time limit is specified, calculate the difference or ratio between the execution time limit and the estimated execution time, and use the value to determine the order in which it is difficult for the program developer to achieve the time limit. Can also indicate the priority of the measures. Also, by using both the average execution time and the maximum execution time together to determine the estimated execution time and comparing it with the execution time limit, the degree of achievement of the time limit for each action can be broadly classified. Priority can be suggested.

Generally, in order to improve the processing time of a program, it is effective to take measures in the order of the processing that occupies the main time in the internal processing. For this reason, when an action predicted to be difficult to achieve the time limit is presented to the program developer (when a warning is presented), it has a large effect on the predicted execution time of the action (designated treatment step). By presenting the internal processes (steps that make up the program flow of an action) in order, it is possible to suggest the priority of measures to improve the processing time for a certain action, and to improve the work efficiency of the measures. Can be planned. Similarly, by presenting the program components executed in the processing of the action in the order of the value of the product of the execution time and the number of executions, indicating the priority of a measure for improving the processing time of a certain action. And the work efficiency of the measures can be improved. Similarly, in general, regarding the processing time of a program, when the difference between the maximum execution time and the average execution time is large, it can be estimated that there is much room for improvement in the processing time.
When listing program parts by the above-mentioned method, the value of the maximum execution time and the average execution time, or the information on the difference or ratio between them is also indicated, so that the priority of the countermeasure for the improvement of the processing time is similarly Can be suggested.

Embodiment 2 FIG. 5 is a flowchart for predicting an execution time according to the second embodiment of the present invention. The configuration of the present embodiment and an example of the state transition information described by the state transition editor unit 1 are the same as those shown in FIGS. 1 to 3 shown in the first embodiment, and the description thereof will be omitted.

In the present embodiment, the program part information database 3 holds, as processing time information of each program part, a histogram indicating the distribution of the execution time measured during the development of each program part and the number of occurrences separately. In this embodiment, the state transition time prediction / comparison unit 4 predicts the time required for the processing of the action based on the execution time histogram as the processing time information of each program component, and calculates the predicted value as the histogram. The state transition time prediction comparison unit 4 compares the predicted execution time with the restricted execution time by calculating the probability that the predicted predicted time will exceed the restricted execution time from the distribution of the predicted histogram. In some cases, the state transition editor unit 1 is used to warn the program developer that the predicted execution time of the corresponding action is unlikely to satisfy the design execution time limit. The display is replaced with the display shown.

The method of estimating the execution time in the present embodiment will be described with reference to FIG. In the execution time prediction process S40, the number of executions corresponding to all execution times of the histogram is set to 0 (S41) in order to initialize a prediction histogram to be processed from now on (S41). The subsequent processing is repeated for each step from the beginning to the end of the program flow (between S22 and S32).

In the repetition, first, the processing contents of the target step are checked (S23). If the processing is the calling processing of the program component, the program component information database 3 is searched and the execution time histogram of the corresponding program component is searched. Is calculated, and the prediction histogram to be processed is recalculated (S42). If the processing is a basic processing such as an arithmetic operation or an assignment, the execution time of the step corresponding to the processing content is estimated based on the execution time of the CPU required for each of the processings such as the arithmetic operation or the assignment. Is recalculated (S43).

On the other hand, if the processing of the target step is a forward branch processing corresponding to an if statement or a switch statement, a partial program flow consisting of a series of steps until the branch joins again for each branch is performed. By performing the execution time prediction process S40 recursively, histograms indicating execution prediction times for the respective branch processes are obtained (S26, S44, S28), and merging from the branches is performed using the histogram of the execution prediction time for each branch. The predicted histogram of the execution time is calculated until (S45), and the predicted histogram to be processed is recalculated (S46). If the processing of the target step is a repetition processing corresponding to a while statement or a for statement, the execution time prediction processing S40 is recursively performed on a partial program flow including all steps of the internal processing to be repeated. Then, a histogram indicating the predicted execution time of the internal processing to be repeated is obtained (S47), a prediction histogram of the execution time of the entire repetition processing is obtained from the histogram and the number of repetitions of the internal processing (S48), and the prediction histogram to be processed is recalculated. (S49). Then, after processing is performed for each step from the beginning to the end of the program flow to be processed (S32), the prediction histogram to be processed is returned as a histogram indicating the predicted execution time in the present processing S40 (S50).

As in the first embodiment, if it is determined in step S42 that the processing may stop in the program component, the processing time cannot be predicted, and the execution time prediction processing S40 is abnormally terminated. The process is continued by inquiring the program developer about the substitute execution value. Similarly, in step S47, when the number of repetitions of the internal processing is not determined, the processing time cannot be predicted and the execution time prediction processing S40 is abnormally terminated, or the processing is continued by inquiring the program developer of a substitute value. Let it.

Next, the method of recalculating the prediction histogram to be processed in step S42 will be described with reference to FIG. FIG. 6 is a flowchart showing a method for recalculating the prediction histogram to be processed in step S42.
In principle, this process S60 is performed when the two processes A and B are sequentially performed, and the execution time histogram H of each process is obtained.
Execution time histogram H _{X of the} entire two processes from _A and H _B
Is a process for obtaining In step S42, the predicted histogram to be processed is recalculated by the present process S60, with the predicted histogram to be processed before calculation being H _A , the histogram of the execution time of the program component being H _B , and the predicted histogram after calculation being H _X. .

In the process S60, first, if the histogram H _A is 0, that is, if the execution count for the execution time is all 0 (YES in S61), the histogram H _B is returned as a result of the process S60. (S6
2, S63), also if the histogram H _B is 0, i.e. the initial state (S64 of YES), the histogram H
_A is returned as the result of this processing S60 (S65, S6
3). Next, when both the histograms H _A and H _B are not 0, that is, both are not in the initialized state (NO in S64),
Histograms H _A and H _B are synthesized in the following steps.

[0053] In the synthesis of the histogram, for initializing the histogram H _X is a work area for obtaining the first calculation result and sets the execution count corresponding to all the execution time of the histogram H _X to 0 (S66). Then the histogram H _A valid time range T _{A, min} ~T _A, each of the time values t _A of _max, the histogram H _B valid time range T
_{B, min} ~T _B, for each of all combinations of time values t _B of _max, repeats the following processing (S67, S71
while). Inside the iteration, the number of times of execution when execution time t _A in the histogram _{H A H A (t A)} , the number of times of execution when execution time t _B in the histogram H _B H
_B (t _B ) and the execution time t in the histogram H _X
As H _X (t) the number of executions when a H _A (t _A)
The value multiplied by H _B (t _B ) is added to the current H _x (t _A + t _B ), and that value is set as a new H _x (t _A + t _B ), that is, the following calculation is performed ( S69). H _x (t _A + t _B ) + = _HA (t _A ) × H _B (t _B ) After performing the above calculation for all combinations of time ranges effective in the histograms H _A and H _B (S 70, 71) returns the histogram H _x as a histogram showing the execution prediction time in the process S60 (S63).

Next, the method of recalculating the prediction histogram to be processed in step S43 will be described with reference to FIG. FIG. 7 is a flowchart showing a method of recalculating the prediction histogram to be processed in step S43.
In the process S80, the execution time is, in principle, the histogram H
Following treatment A given as _A and the execution time is a fixed length T _B
When performing the process B to the sequential it is the entire two processing execution time a histogram H _X process. In step S43, the predicted histogram to be processed is recalculated by the present process S80, with the predicted histogram to be processed before calculation being H _A , the estimated execution time of the basic processing being T _B , and the predicted histogram after calculation being H _X. .

In the process S80, first, the histogram H_ABut
0, that is, the number of executions for the execution time is all 0
If it is in the initialized state (YES in S61), execute
Time T_BThis processing S
Returned as result of 80 (S81, S63), otherwise
If (NO in S61), the histogram H_ARun against
Time T_BA minute shift process is performed. In shift processing, first
Histogram H, which is a work area for obtaining a calculation result
_XTo initialize the histogram H_XAt all runtimes of
The number of times of execution corresponding thereto is set to 0 (S66). next
Histogram H_AValid time range T_{A, min}~ T_{A, max}of
Each time value t_ARepeat the following process for
(Between S67 and S71). Histog inside the repetition
Ram H_AExecution time t in_AThe number of executions at
_A(T_A), Histogram H_XAt execution time t in
Execute count H_XAs (t), H_A(T_A) To H_x(T
_A+ T_B), That is, the following processing is performed (S8)
2). H_x(T_A+ T_B) = H_A(T_A) Histogram H_AFor all combinations of time ranges valid in
After performing the above calculation (S71), the histogram
H_xIs a history indicating the predicted execution time in the main processing S80.
It is returned as a gram (S63).

Next, a method of calculating a prediction histogram of the execution time from the branch to the merge in step S45 will be described with reference to FIG. FIG. 8 is a flowchart showing a method for calculating a prediction histogram of the execution time from the branch to the merge in step S45.

[0057] In this processing S90, the histogram H _X is a work area for obtaining the first calculation result to initialize the execution count corresponding to all the execution time of the histogram H _X is set to 0 (S66), Next, the following processing is repeated for each branch (between S91 and S95). Repeat S91
-S95, the following processing is further performed for each time value t _i in the effective time range T _{i, min to} T _{i, max} in the prediction histogram H _i of the execution time from the branch to the merge for the branch _i. Repeat (between S92 and S94). Inside between repeating _{S92-S94, H i (t} i) the number of execution times when the execution time t _i in the histogram H _i, the total number of branches n, the number of times of execution when execution time t in the histogram H _X H _{As X} (t), the value of H _i (t _i ) is divided by n and then added to Hx (t _i ), that is, the following processing is performed (S93). H _X (t _i ) + = H _i (t _i ) / n The above calculation is performed for all combinations of valid time ranges in the histogram H _i (S94), and the process is repeated for all branches (S95). returns the histogram H _x as a histogram showing the execution prediction time in the process S90 (S63).

Next, a method of calculating a prediction histogram of the execution time of the entire iterative process in step S48 will be described with reference to FIG. FIG. 9 is a flowchart showing a method for calculating a prediction histogram of the execution time of the entire iterative process in step S48.

[0059] In this process 100, the histogram H _X is a work area for obtaining the first calculation result to initialize the execution count corresponding to all the execution time of the histogram H _X is set to 0 (S66), Next, the processing for the number of repetitions or less is repeated (between S101 and S103). Inside the repetition, the histogram H _X is used by using the process S60.
If, from the prediction histogram H _A running time of the internal processing part repeats found at step S47 to perform the process S40 as a preceding step S48, the re-computes the value of new histogram, sets it to the histogram H _X ( S102). Repeated step S102 is finished (S103), the processing histograms H _x S100
(S63).

Finally, a method of recalculating the prediction histogram to be processed in steps S46 and S49 will be described. The method of recalculating the prediction histogram to be processed in both steps is the same as the method of recalculating the prediction histogram to be processed in step S42, and the calculation is performed using the processing S60.

Since the program design support apparatus according to the present embodiment operates as described above, it is possible to simultaneously describe the time limit allowed for the processing of the action together with the state transition when designing the state transition. Further, by estimating the time required for processing the action as a histogram, the probability of achieving the time limit can be estimated, thereby reducing the possibility of occurrence of a problem relating to the possibility of achieving the time limit in a downstream process of software development. As a result, effects such as improvement in program development efficiency and quality can be obtained.

In the present embodiment, the program component information database holds, as processing time information of each program component, a histogram indicating the distribution of the execution time measured during the development of each program component and the number of occurrences separately. However, for this, a histogram of the execution time predicted by executing the processing time prediction processing S40 in the present embodiment based on the program flow of each program component may be used. Alternatively, when a program component is reused, a histogram based on measurements performed in the past may be used. Furthermore, if the operation of the entire program is possible, instead of using a value measured for each program component, a histogram based on the measured value of the execution time of each program component obtained when the entire program is operated is used. In this case, the distribution of the prediction histogram of the execution time for the action is more likely to be closer to the distribution in the actual operation, and the effect of increasing the accuracy of predicting whether or not the execution time limit can be achieved is obtained.

In the present embodiment, among the prediction processing S40 of the time required for the action processing in the state transition time prediction comparison section 4, in step S43, the predicted value of the execution time of the basic processing such as arithmetic operation and substitution is Although a fixed value is used, the execution time of the CPU required for each processing such as arithmetic operation and substitution is treated as a histogram, and the prediction of the execution time of the basic processing is calculated in the same manner as in the processing S40. May be treated as a histogram. In this case, the recalculation of the prediction histogram to be processed in step S43 is performed in step S80.
This is performed by adopting the program flow shown in step S60 instead of the program flow shown in FIG.

Further, in the present embodiment, in the prediction process S40 of the time required for the processing of the action in the state transition time prediction comparison section 4, the method S90 for calculating the prediction histogram of the execution time from the branch to the merge in step S45. Then, the probability of each branch occurring is equal 1 / n (n is the number of branches)
In step S93, the following calculation method is used: H _X (t _i ) + = H _i (t _i ) / n. If the probability of each branch occurring is known,
Using the probability p (i) of occurrence of each branch i, the above calculation method may be replaced by the following calculation method: H _X (t _i ) + = H _i (t _i ) × p (i), in this case The accuracy of predicting the execution time from the branch to the merge in step S45 can be improved.

Further, similarly to the first embodiment, in the present embodiment, the state transition time prediction comparing section 4 compares the execution restriction time with the execution restriction time by multiplying the execution restriction time by a constant ratio. For example, the comparison with the predicted execution time may be performed using, for example, 80% of the restricted execution time as a guide, and the presentation method to the program developer is changed according to the probability that the predicted predicted execution time exceeds the restricted execution time. Is also good. For example, be careful if the probability that the predicted execution time exceeds 80% of the execution time limit exceeds 5%, and be cautious if the probability that the execution time exceeds 80% exceeds 20%.
Furthermore, when the probability of exceeding the execution time limit of 100% exceeds 5%, a warning and a method of presenting it to the program developer are differentiated, so that the priority of the measure can be suggested to the program developer. In addition, the work efficiency of the measures can be improved.

As in the first embodiment, also in this embodiment, when presenting an action predicted to be difficult to achieve the time limit to the program developer,
Accompanying this, the program components executed in the processing of the action are listed in descending order of the effect on the distribution of the predicted execution time. Further, when listing the program components, the execution time of the program component is By indicating the information on the distribution of the histogram, it is possible to suggest the priority of the measure for improving the processing time for the action, and to improve the work efficiency of the measure.

The degree of influence of each program component on the distribution of the predicted execution time of an action can be measured by several methods. For example, it can be considered that the influence on the predicted execution time is greater as the value of the product of the average value of the processing times of the program components and the number of executions increases. Alternatively, a correlation between a histogram representing the processing time corresponding to the number of executions of each program component and a distribution of the execution time prediction histogram may be obtained, and when the correlation is high, it may be considered that the influence on the distribution of the execution prediction time is large. . Furthermore, other methods may be employed, or some methods may be used in combination. In any case, the priority of a measure for improving the processing time for an action can be suggested.

Further, in the present embodiment, when the histogram of the execution time or the predicted time is handled, the histogram is simply handled by the distribution of the execution time-the number of executions. For example, the total execution number is always 100. May be normalized. The normalization of the histogram so that the total number of executions is always 100 can be performed as follows. In this case, the number of executions corresponding to each execution time of the histogram after the normalization matches a value indicating the possibility that the actual time will be each execution time as a percentage.

[0069]

(Equation 1)

As described above, when the histogram is calculated, the processing is performed so as to always re-normalize, so that it is possible to avoid an operation error during the calculation of the histogram as compared with a case where normalization is not performed, and to stabilize the calculation processing. The effect that the property is improved is obtained.

In the above, the distribution of the execution time or the predicted time is treated as a histogram relating to the execution time and the number of executions. However, a more general probability distribution may be used, and the same effect as in the case of using the histogram is obtained. Is obtained. In the present embodiment, when using the probability distribution instead of the histogram, first, the probability distribution is held in the program component information database 3 as the processing time information of each program component. The transition time prediction comparison unit 4 predicts the time required for the processing of the action based on the probability distribution of the execution time as the processing time information of each program component, and calculates the predicted value as the probability distribution. Also, the state transition time prediction comparison unit 4
In the comparison between the execution time limit and the execution time limit, the probability that the predicted execution time exceeds the execution time limit is determined from the predicted probability distribution, and if there is an action whose value exceeds a certain value, the state transition editor unit 1 is used. In order to indicate to the program developer that the predicted execution time of the corresponding action is unlikely to satisfy the design execution time limit, the display of the corresponding action is changed to a display indicating a warning and presented. Each process S6 for further calculating a histogram
0, S80, S90, and S100 are replaced by the following processes.

The process S60 is basically composed of two processes A,
When B is performed sequentially, the execution time histograms H _X of the entire two processes are obtained from the execution time histograms H _A and H _{B of the} respective processes. For this reason, in the case where the probability distribution is used instead of the histogram, when two processes A and B are sequentially performed, two processes A and B are used to calculate the execution time probability distributions P _A and P _{B of} each process. This is a process for obtaining the probability distribution P _X of the execution time of the entire process. This processing is represented by the following equation.

[0073]

(Equation 2)

[0074] process S80, when the execution time in principle is followed by the execution time to process A given as histograms H _A went sequential processing B is a fixed length T _B, 2
One execution of the overall processing time is a process of obtaining a histogram H _X. Therefore, in the case of using the probability distribution instead of the histogram, the two processes A and B are performed sequentially, and the probability distribution P _A of the execution time of the process _A and the execution time T of the process B a process of obtaining the probability distribution P _X for the entire two processing execution time from _B. This processing is represented by the following equation. _{P X (t) = P A} (t-T B) where P _A (t) ... probability indicating the probability that the execution time is t processing A distribution T _B ... processing execution time of the B (fixed value) P _X ( t) Probability distribution indicating the probability that the time at which the processes A and B are sequentially executed becomes t

[0075] processing S90 is processing having a plurality of branches, based on the predicted histogram H _i of execution time to merge from a branch for each branch i, the process for obtaining the execution time histogram H _X leading to the merging from the branch is there. For this reason, the equivalent process in the case of using the probability distribution instead of the histogram is based on the predicted probability distribution P _i of the execution time from the branch to the junction for each branch i, and the probability distribution P of the execution time from the branch to the junction. _This is the process for obtaining _X. This processing is represented by the following equation, assuming that the probability of each branch occurring is equal.

[0076]

(Equation 3)

If the probability of the occurrence of each branch is known,
It is represented by the following equation using the probability p (i) of occurrence of each branch i.

[0078]

(Equation 4)

In the process S100, when the repetition of the internal process A is repeated N times, the execution time histogram H _{X of the} entire repetition process is obtained from the execution time histogram H _A of the internal process.
Is a process for obtaining Thus, equivalent processing when using the probability distribution instead of a histogram, when the internal processing A of the repetition is repeated n times, the probability distribution iteration overall execution time from the probability distribution P _A in the execution time of the internal processing This is a process for obtaining P _X. This processing is represented by the following equation.

[0080]

(Equation 5)

As described above, even when a general probability distribution is used instead of a histogram, the same effect as that obtained when a histogram is used can be obtained. Further, for example the process A, when executing continue the process B, if the correlation information on the distribution of execution times t _B execution time t _A and the processing B of the process A has been obtained, continued processing A, the B and when obtaining the probability distribution of the execution times of the time, the processing B of the probability distribution can be used in accordance with the execution time t _a of the probability distribution P _B to the processing a of the execution time, execution prediction time for this action Is more likely to be closer to the distribution of execution times in an actual operation, and the effect of increasing the accuracy of predicting whether or not the execution time limit can be achieved is obtained. In addition, by selecting an appropriate mathematical expression as the probability distribution to be used, the operation of calculating the probability distribution can be easily handled using statistical knowledge, and the processing for predicting the execution time can be speeded up. The effect of improving work efficiency is obtained.

In the first and second embodiments described above, the program design support apparatus according to the present invention has been described. However, a program design support method according to the present invention can be realized by a similar processing procedure.

The features of the present invention described above are summarized as follows. The program design support method according to the present invention includes, as program specifications, a plurality of states, a plurality of events, transitions between states caused by occurrence of an event in each state, and an action executed in each transition. A state transition editor unit for describing state transition information including, a state transition information holding unit for storing state transition information described by the state transition editor unit, and processing time information for each program component used for creating a program. It consists of a held program part information database and a state transition time prediction unit that predicts the time required for processing an action based on the state transition information and the processing time information of the program part. In the description, you can define the program A program flow can be described, and a time limit (execution time limit) allowed for the processing of an arbitrary action can be described. Based on this information, the time required for processing the action is predicted using the processing time information of the program components included in the program component information database (execution predicted time), and the execution predicted time and the state transition editor section are described. It is characterized in that it can be presented in comparison with the execution time limit for the action.

In the above program design support system, the program component information database holds the average value and / or the maximum value of the processing as the processing time information of each program component. When calculating, make a prediction based on the average or maximum value for each program component stored in the program component information database, and if the value exceeds or close to the action execution time limit, a warning should be presented. Features.

In the above program design support method, when the state transition time prediction section gives a warning, the program parts used in the program flow of the corresponding action are listed in ascending order of the processing time. It is characterized by.

In the above-mentioned program design support system, the program component information database holds processing time information of each program component as a statistical value such as a histogram or a probability distribution instead of a specific fixed value.
The state transition time prediction unit calculates the predicted execution time of the action by performing a prediction based on statistical processing based on the processing time information of each program component held in the program component information database, thereby executing the prediction. It is characterized in that the probability that the predicted time exceeds or approaches the action execution time limit is obtained, and if the probability exceeds a certain value, a warning is presented.

In the above program design support system, when the state transition time prediction section gives a warning, the program component used in the program flow of the corresponding action has a high effect on the predicted execution time. They are listed in order.

[0088]

According to the present invention, when designing a state transition, it is possible to simultaneously describe the time limit allowed for the processing of the action together with the state transition, and to predict the time required for the processing of the action. , It is possible to estimate whether or not the time limit can be achieved, thereby reducing the possibility of occurrence of a problem related to whether or not the time limit can be achieved in a downstream process of software development, and improving program development efficiency and quality. Is obtained.

According to the present invention, when designing a state transition, the time limit allowed for the processing of the action can be described together with the state transition, and the time required for the processing of the action can be predicted as a histogram or a probability distribution. By doing so, it is possible to estimate the probability of achieving the time limit, thereby reducing the possibility of occurrence of a problem related to the achievement of the time limit in the downstream process of software development, and to improve the program development efficiency and quality. The effect of improvement is obtained.

[Brief description of the drawings]

FIG. 1 is a diagram showing a configuration of a program design support device.

FIG. 2 is a diagram showing an example of state transition information.

FIG. 3 is a diagram showing an example of an action program flow.

FIG. 4 is a flowchart showing an execution time prediction process according to the first embodiment;

FIG. 5 is a flowchart showing an execution time prediction process according to the second embodiment.

FIG. 6 is a flowchart showing a prediction histogram recalculation method according to the second embodiment.

FIG. 7 is a flowchart showing a prediction histogram recalculation method according to the second embodiment;

FIG. 8 is a flowchart showing a method for calculating a prediction histogram according to the second embodiment.

FIG. 9 is a flowchart showing a method for calculating a prediction histogram according to the second embodiment.

FIG. 10 is a diagram showing a state transition diagram according to the related art.

FIG. 11 is a diagram showing a state transition table according to the related art.

[Explanation of symbols]

1 state transition editor section, 2 state transition information holding section, 3
Program part information database, 4 state transition time prediction comparison section, 10 program design support device, 11 states, 12 transitions.

Claims (20)

[Claims] 1. A program including a designated processing step designated as a time allowed for execution of a process as a permissible execution time, estimating a time required for executing the designated processing step to derive an estimated execution time, A program design support device comprising: an execution time prediction comparison unit that compares an execution prediction time of a designated processing step with an allowable execution time; and a comparison result output unit that presents a comparison result by the execution time prediction comparison unit. . 2. The execution time prediction / comparison unit inputs state transition information indicating a state transition in the program and a processing step executed at the time of the state transition, and the input state transition 2. The program design support apparatus according to claim 1, wherein an estimated execution time of the designated processing step among the processing steps indicated in the information is derived. 3. The designated processing step includes at least one or more internal processing steps, and the execution time prediction comparing unit predicts a time required for executing each of the internal processing steps and performs execution prediction of each of the internal processing steps. 2. The program design support apparatus according to claim 1, wherein a time is derived, and a total value of predicted execution times of the respective internal processing steps is compared with a permissible execution time of the designated processing step. 4. The designated processing step includes, as the internal processing step, an execution process of a program component, wherein the program design support device includes: an average execution time information indicating an average execution time of the program component; A program component information database that stores at least one of maximum execution time information indicating the maximum time of the execution process of the program component, wherein the execution time prediction comparison unit is stored in the program component information database Based on at least one of the average execution time information and the maximum execution time information,
4. The program design support apparatus according to claim 3, wherein a time required for the execution process of the program component is predicted to derive an estimated execution time of the execution process of the program component. 5. The execution time prediction comparing section, as a result of a comparison between an execution prediction time of the designated processing step and an allowable execution time,
When determining that the predicted execution time of the designated processing step exceeds a predetermined level of the permissible execution time, instructing the comparison result output unit to present warning information about the designated processing step; 2. The program design support device according to claim 1, wherein the output unit presents warning information about the designated processing step based on an instruction from the execution time prediction comparison unit. 6. The execution time prediction comparison unit derives an occurrence probability that an execution prediction time exceeding a predetermined level of an allowable execution time occurs for the specified processing step, and the derived occurrence probability is equal to or more than a certain value. When it is determined, the comparison result output unit is instructed to present warning information about the designated processing step, the comparison result output unit, based on the instruction from the execution time prediction comparison unit, 2. The program design support apparatus according to claim 1, wherein warning information about the designated processing step is presented. 7. The program design support device according to claim 6, wherein the execution time prediction comparison unit derives the occurrence probability using at least one of histogram information and probability distribution information. . 8. The designated processing step includes a plurality of internal processing steps, the execution time prediction comparing section performs an ordering among the plurality of internal processing steps, and the comparison result output section includes a plurality of processing steps together with the warning information. 7. The program design support device according to claim 5, wherein the plurality of internal processing steps are presented in accordance with the ordering. 9. The specification processing step includes, as the internal processing step, an execution process of a plurality of program components, the execution time prediction comparing unit performs an ordering among the execution processes of the plurality of program components, The program design support apparatus according to claim 8, wherein the comparison result output unit presents the execution processing of the plurality of program components in accordance with the ordering, together with the warning information. 10. The execution time prediction comparing section orders the execution processing of the plurality of program components according to a magnitude of an influence on an execution prediction time of the designated processing step. A program design support device according to item 1. 11. The execution time prediction comparing section, for a program including a plurality of designated processing steps, as a result of comparing the predicted execution time of each designated processing step with the permissible execution time, Calculate at least one of the difference between the time and the execution allowable time and the ratio of the execution prediction time to the execution allowable time, and calculate the difference between the designated processing steps based on at least one of the calculated difference and the ratio. 2. The program design support device according to claim 1, wherein the ordering is performed, and the comparison result output unit presents each of the designated processing steps according to the ordering performed by the execution time prediction comparing unit. 3. 12. For a program including a designated processing step designated as a permissible execution time of a process, a predicted execution time is derived by estimating a time required to execute the designated processing step. A program design support method comprising: an execution time prediction comparison step of comparing an execution prediction time of a designated processing step with an allowable execution time; and a comparison result output step of presenting a comparison result by the execution time prediction comparison step. 13. The execution time prediction comparing step includes inputting state transition information indicating a state transition in the program and a processing step executed at the time of the state transition, and the input state transition 13. The program design support method according to claim 12, wherein an estimated execution time of the designated processing step among the processing steps indicated in the information is derived. 14. The specified processing step includes at least one or more internal processing steps, and the execution time prediction comparison step predicts the time required for execution of each internal processing step and predicts the execution of each internal processing step. 13. The program design support method according to claim 12, wherein a time is derived, and a total value of predicted execution times of the respective internal processing steps is compared with an allowable execution time of the designated processing step. 15. The method according to claim 15, wherein the designating process includes a program component execution process as the internal processing process, wherein the program design support method includes: an average execution time information indicating an average time of the program component execution process; A program component information storing step of storing at least one of maximum execution time information indicating the maximum time of the execution process of the program component, wherein the execution time prediction comparing step is performed by the program component information storing step. Estimating a time required for the execution process of the program component based on at least one of the stored average execution time information and maximum execution time information to derive an estimated execution time of the execution process of the program component; The program design support method according to claim 14, wherein: 16. The execution time prediction comparing step determines that the estimated execution time of the specified processing step exceeds a predetermined level of the allowable execution time as a result of the comparison between the estimated execution time of the designated processing step and the allowable execution time. In this case, the comparison result output step is instructed to present warning information about the designation processing step, and the comparison result output step is based on the instruction from the execution time prediction comparison step. 13. The method according to claim 12, wherein warning information about the process is presented. 17. The execution time prediction comparing step derives a probability of occurrence of a predicted execution time exceeding a predetermined level of an allowable execution time for the specified processing step, and the derived probability is a predetermined value or more. When it is determined, the comparison result output step is instructed to present warning information about the designated processing step, the comparison result output step is based on an instruction from the execution time prediction comparison step, 13. The program design support method according to claim 12, wherein warning information about the designated processing step is presented. 18. The designated processing step includes a plurality of internal processing steps, the execution time prediction comparing step performs an ordering among the plurality of internal processing steps, and the comparing result output step includes a step of outputting together with the warning information. ,
18. The method according to claim 16, wherein the plurality of internal processing steps are presented according to the order. 19. The specification processing step includes, as the internal processing step, an execution process of a plurality of program components, the execution time prediction comparing step orders the execution processes of the plurality of program components, The comparison result output step includes, together with the warning information,
19. The program design support method according to claim 18, wherein the execution processes of the plurality of program components are presented according to the ordering. 20. For a program including a designated processing step designated as a permitted time for execution of a process as a permitted execution time, predicting a time required for executing the designated processing step to derive an estimated execution time, A program design support method, comprising: an execution time prediction comparison step of comparing an execution prediction time of a designated processing step with an allowable execution time; and a comparison result output step of presenting a comparison result by the execution time prediction comparison step. A program for causing a computer to execute.