JP2009301298A - System, method and program for estimating response impossibility time of application - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a system, method, and program for estimating the response impossibility time of an application by using a proper threshold by statistically calculating a threshold resulting in the generation of response impossibility in every execution environment when response impossibility is generated. <P>SOLUTION: This system includes: an operation event counting means 202 for counting the number of operation events including at least the number of key strokes from a user accepted in an application; a response impossibility log information DB 130 for, when the generation of an application response impossibility state is detected, storing the number of operation events from an application start time and the execution environment information of the application in association with the application as response impossibility log information by interrupt processing; and a response impossibility estimation means 110 for estimating the estimated number of operation events resulting in the generation of the response impossibility based on the information stored in the response impossibility log information DB 130. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a system, method, and program for estimating when an application cannot respond.

  A state in which the application stops responding to an operation from the keyboard or mouse while the application is being used may occur, and is generally referred to as freezing or hang-up (hereinafter collectively referred to as unresponsiveness). The only way to deal with a failure to respond is to forcefully terminate the application through a special keyboard operation, or restart the computer by pressing the reset switch if it is not accepted. For this reason, it is important for the user to know in advance that a response failure will occur and to save data and prevent the response failure itself.

Therefore, as a method for alerting the user in advance of the inability to respond, when the memory usage amount of the apparatus and the garbage collection processing interval are used and the threshold that is predicted to be inadequate by the user is exceeded, There is a method of warning (for example, Patent Document 1 and Patent Document 2).
JP 2006-285871 A JP 2007-226399 A

  However, according to the above technique, it is determined whether or not a response failure occurs by using a threshold value set in advance by a person, and a warning is given. Therefore, the threshold value is not appropriate when the performance of the apparatus is different or when the application startup environment is different, that is, when the environment where the response failure occurs is different from when the threshold value is set. There are various causes for the inability to respond, but the cause cannot be identified by the memory usage or the garbage collection processing interval. Therefore, the threshold value is an average value in response failure caused by various causes, and may not be appropriate.

  Therefore, in view of the above problems, the present invention estimates the threshold value at which an unresponsiveness occurs for each environment when the unresponsiveness occurs by estimating the occurrence of an unresponsiveness using an appropriate threshold. The present invention is directed to a system, method, and computer program for estimating when an application is unable to respond.

(1) A system for estimating when an application cannot respond,
Application detection means for detecting the activation of the application;
An operation event counting unit that counts the number of operation events including at least the number of keystrokes from the user performed on the application in response to detection of activation by the application detection unit;
Interrupt detection means for detecting occurrence of an unresponsive state of the application;
In response to the detection of the inability to respond to the application by the interrupt detection means, the number of operation events calculated from the start of the application and the information on the execution environment from the start of the application are responded by interrupt processing. Response impossible log information recording means for storing in association with the application as impossible log information;
Based on the information on the execution environment of the application stored by the unresponsive log information recording unit and the number of operation events, the unresponsive log information is classified to estimate the estimated number of operation events to be unresponsive. Means,
A system comprising:

  According to the invention of (1), execution can be performed by estimating the number of operation events in which an application in a predetermined execution environment becomes unresponsive based on the number of operation events calculated from the start of the application and information on the execution environment of the application. It is possible to estimate the number of operation events that become unresponsive for each environment. Thereby, the estimated number of operation events close to the number of operation events that are actually unresponsive can be estimated.

  An operation event is an operation performed on an application. The operation events are roughly divided into two, that is, keystrokes that are events by keyboard operation and events by mouse operation. The present invention includes at least a keystroke that is strongly involved in the inability to respond to an application as an operation event.

  (2) In response to the activation of the application, the estimated number of operation events is determined based on information on an execution environment of the application, and the number of operation events reaches a predetermined threshold that is less than the estimated number of operation events. The system according to (1), further comprising: an informing unit that informs the user of the application to that effect.

  According to the invention of (2), the user's operation event is monitored, and when the predetermined threshold value obtained from the estimated operation event number at which the application becomes unresponsive is approached, the user is notified. The threshold value smaller than the estimated operation event number is, for example, a value of 80% of the estimated operation event number. Moreover, alerting | reporting is displaying a message or sounding an alarm, for example. As a result, the user knows in advance that an unresponsiveness occurs, and the influence when the unresponsiveness occurs can be minimized by storing data.

  (3) The system according to (2), wherein an execution status of another application is used as an element for determining the estimated number of operation events together with information on an execution environment of the application.

  According to the invention of (3), the system uses an execution status of another application in combination with an element for determining the estimated number of operation events. As a result, when the estimated number of operation events is determined, the system improves the accuracy of the determination by determining the execution status of other applications that are considered to have an effect on the occurrence of a response failure. Can do.

  (4) The system according to any one of (1) to (3), wherein the response failure estimation unit classifies the response failure log information by a support vector machine using the estimated operation event number as a support vector.

  According to the invention of (4), by classifying non-response log information by the support vector machine, it is possible to classify non-response cases showing the same tendency even if the execution environment is different, and the estimated operation event The number is obtained as a support vector. Thereby, it is possible to estimate an appropriate number of estimated operation events for an unknown case.

  A support vector machine is a learning system that uses a hypothesis space of a linear function in a high-dimensional feature space. Specifically, in 1995, AT & T's V.C. It is a learning machine proposed by Vapnik in the framework of statistical learning theory, and is a learning system that uses a hypothesis space of a linear function in a high-dimensional feature space.

(5) Furthermore, an operation event number storage unit that stores the number of operation events calculated from the start of the application by an interrupt process in response to detection of an inability to respond to the application by the interrupt detection unit; ,
The number of operation events stored in the operation event number storage means, and the number of operation event transmission means for transmitting information on the execution environment from the time of starting the application,
The system according to any one of (1) to (4).

  According to the invention of (5), the operation event number storage means responds by storing the operation event number calculated from the start of the application by interrupt processing in response to detecting that the application cannot respond. The number of operation events can be saved even when it is impossible. Further, by providing the operation event number transmission means, the number of operation events and the like can be integrated into one, and statistics when response is impossible can be obtained.

  (6) The system according to any one of (1) to (5), wherein the interrupt detection unit determines that the response is disabled when the application is stopped or when the processing of the application is delayed. .

  According to the invention of (6), the present invention can also be applied to a state in which the application is substantially stopped by including the case where the processing of the application is stagnant in the inability to respond.

  (7) The system according to any one of (1) to (6), wherein the information on the execution environment includes performance of a device in which the application is activated and information on an application that is simultaneously activated.

  According to the invention of (7), the estimated number of operation events is calculated based on the performance of the device in which the application is activated as the application information, and information on other applications activated in parallel and the number of operation events. Is done. Thereby, it is possible to reflect in the calculation of the estimated number of operation events a factor that is considered to have an influence on the occurrence of an unresponsiveness of the application.

(8) A method for estimating when an application cannot respond using a computer,
Detecting activation of the application;
Counting the number of operation events including at least the number of keystrokes from the user performed on the application;
Detecting the occurrence of an unresponsive state of the application;
In response to the detection of the inability to respond to the application, the number of operation events calculated from the start of the application by interrupt processing and the information on the execution environment from the start of the application are used as the response impossible log information. Associating and storing, and
Classifying the unresponsive log information based on information on the execution environment of the application and the number of operation events, and estimating an estimated number of operation events that result in unresponsiveness;
Including methods.

  By causing the computer to execute the method (8), the same actions and effects as in (1) can be expected.

(9) A program for causing a computer to estimate when an application cannot respond,
Detecting activation of the application;
Counting the number of operation events including at least the number of keystrokes from the user performed on the application;
Detecting the occurrence of an unresponsive state of the application;
In response to the detection of the inability to respond to the application, the number of operation events calculated from the start of the application by interrupt processing and the information on the execution environment from the start of the application are used as the response impossible log information. Associating and storing, and
Classifying the unresponsive log information based on information on the execution environment of the application and the number of operation events, and estimating an estimated number of operation events that result in unresponsiveness;
A program that executes

  By introducing the program (9) into a computer and executing each step, the same actions and effects as in (1) can be expected.

  According to the present invention, for each environment when an unresponsiveness occurs, an application that can estimate the unresponsiveness occurrence time using an appropriate threshold value by statistically obtaining a threshold value at which unresponsiveness occurs. It is possible to provide a system, a method, and a computer program for estimating when a user cannot respond.

  Hereinafter, the best mode for carrying out the present invention will be described with reference to the drawings. This is merely an example, and the technical scope of the present invention is not limited to this.

[Overall configuration of estimation system when unable to respond]
FIG. 1 is a diagram illustrating an overall configuration of a response impossible estimation system 1 according to the present embodiment. The client terminal 200 and the server apparatus 100 in the response impossible time estimation system 1 are connected via a communication line of a network represented by the Internet, an intranet, or the like. In addition, the client terminal 200 and the server apparatus 100 in this structure are not necessarily one unit, and may be composed of a plurality of units. In addition, the response impossible time estimation system 1 may be provided as a device. However, in order to obtain a threshold value at which the response impossible time occurs, it is preferable to use many events, and thus a system in which a plurality of client terminals 200 are connected. Is desirable.

[Functional configuration of the client terminal of the response impossible estimation system]
As shown in FIG. 1, the client terminal 200 includes a communication unit 206 and a control unit 210. The control unit 210 includes an application detection unit 201, an operation event count unit 202, an interrupt detection unit 203, an operation event number storage unit 204, an operation event number transmission unit 205, an application monitoring unit 207, and a notification unit 208.

  The application detection unit 201 detects that an application has been started on the client terminal 200. The operation event counting means 202 counts the number of operation events performed by the user with a keyboard or mouse for each application. However, the operation event includes at least a keystroke. The reason will be described later.

  The interrupt detection unit 203 detects the occurrence of an inability to respond to an application, and instructs the operation event number storage unit 204 to store the number of operation events of the application that has become unresponsive.

  In response to an instruction from the interrupt detection unit 203, the operation event number storage unit 204 acquires the number of operation events at that time from the operation event count unit 202 and executes an interrupt routine that temporarily stores the operation event number. Further, the operation event count storage unit 204 clears the count of the operation event count unit 202 when temporarily storing the operation event count. When the response failure occurs, the processing being executed by the CPU is interrupted, and the interrupt routine executed by the operation event number storage unit 204 performs the processing with priority.

  The operation event number transmission unit 205 acquires the operation event number temporarily stored in the operation event number storage unit 204, the application information that has become unresponsive, and the execution environment information of the application, and transmits the information to the server device 100. Information is prepared and passed to the communication means 206. Information on the execution environment of the application includes the performance of the client terminal 200 such as the CPU (Central Processing Unit) of the client terminal 200 and the specifications of the virtual memory, the date and time when the response becomes impossible, the information of the application that is started in parallel, etc. It is.

  When the application detection unit 201 detects that the application has been started, the application monitoring unit 207 acquires the estimated operation event number estimated to be unresponsive to the application from the server device 100, and the estimated operation event number It is monitored whether or not the number of operation events acquired from the operation event counting means 202 exceeds the threshold obtained from the above. The application monitoring unit 207 issues an instruction to the notification unit 208 when the number of operation events exceeds the operation event threshold. Here, the threshold value is a value arbitrarily set by a system administrator or the like based on the estimated operation event number, and for example, a value of 80% of the estimated operation event number can be set as the threshold value. Note that the number of operation events is sequentially acquired from the operation event counting unit 202.

  When receiving the instruction from the application monitoring unit 207, the notification unit 208 gives a warning to the user at an output unit (not shown). The control unit 210 clears the count of the operation event counting unit 202 when the application is finished or when the user performs an operation according to the warning performed by the notification unit 208.

[Functional configuration of server device of estimation system when unable to respond]
As shown in FIG. 1, the server device 100 includes a communication unit 101, a response impossible log information recording unit 102, a response impossible estimation unit 110, an estimated response impossible information acquisition unit 120, a response impossible log information database (hereinafter, for simplification). DB) 130, and a response impossible estimation DB 131. The response impossible log information recording unit 102 receives application information indicating that the response is disabled, the number of operation events when the response is disabled in the application, and information on the execution environment of the application from the client terminal 200 via the communication units 206 and 101. Acquired and stored in the response impossible log information DB 130.

  The unresponsiveness estimation means 110 can determine the estimated number of operation events by classifying the unresponsiveness log information. Here, the unresponsiveness estimation unit 110 is a learning machine that performs classification, and includes, for example, a support vector machine (SVM) and naive bayes. In this embodiment, the case where the unresponsiveness estimation unit 110 is a support vector machine will be described.

  In an example of the present embodiment, a reliable means called a support vector machine is used. Here, the learning result of the support vector machine includes an identification surface and a support vector for classifying the non-response log information stored in the non-response log information DB 130 into classes, and is stored in the non-response estimation DB 131. When classifying into multiple classes, combine multiple support vector machines.

  The estimated response impossibility information acquisition unit 120 acquires the estimated number of operation events from the response impossibility estimation DB 131 based on the newly started application information acquired from the client terminal 200 and the execution environment information of the application, and the client terminal 200. Send to.

  The non-response log information DB 130 stores the non-response log information, that is, information on the number of operation events of the application and the execution environment when the response failure occurs in the client terminal 200. Details will be described later. In the unresponsiveness estimation DB 131, the learning result in the unresponsiveness estimation unit 110 and the estimated number of operation events are stored for each class into which unresponsiveness log information is classified by the unresponsiveness estimation unit 110.

  The hardware configurations of the client terminal 200 and the server device 100 will be described later.

[Response impossible estimation means]
FIG. 2 is a diagram showing details of the response impossible estimation means 110. The unresponsiveness estimation unit 110 includes a class generation unit 111 and an estimated operation event number determination unit 112. The class generation unit 111 performs feature analysis of the non-response log information stored in the non-response log information DB 130 and classifies the non-response log information. The estimated operation event number determination unit 112 calculates the estimated number of operation events that cause a response failure for each class created by the class generation unit 111.

  By the way, the reason why the response failure is estimated based on the number of operation events is that it is possible to predict the occurrence of a memory leak that causes the response failure based on the number of operation events. The memory leak is a state in which the memory area occupied by the application software for processing is not released for some reason. As the number of operation events increases, the memory area that is not released may increase. However, it is difficult to detect the occurrence of the memory leak by monitoring the memory usage. Because it cannot distinguish between what the program reserves memory for future use and what it keeps occupying even though there is no plan for such future use, This is because just because the memory usage is large, it cannot be said that a memory leak has occurred. Therefore, as a means that is more effective than the memory monitoring, in the present invention, the response failure time is estimated based on the number of operation events.

  As described above, the operation event includes an event by a mouse operation in addition to a keystroke which is an event by a keyboard operation. In the present invention, attention is paid to a keystroke as an event that is strongly involved in the inability to respond to an application. This is because there are various events such as button presses, URL clicks, etc., because the relationship between the memory area reservation and release is not always clear, but the keystrokes are input of character characters, Kana-Kanji conversion, etc. This is because the relationship between securing and releasing the memory area is stable and suitable for estimating the inability to respond to an application.

[Estimated response impossible information acquisition means]
FIG. 3 is a diagram showing details of the estimated response impossibility information acquisition means 120. The estimated response impossibility information acquisition unit 120 includes a class determination unit 121 and an estimated operation event number extraction unit 122. Based on the application information (including at least the application ID) of the client terminal 200 that has made the estimated operation event acquisition request and the execution environment information 2 of the application, the class determination unit 121 determines the class to which they belong. The estimated operation event number extraction unit 122 acquires the estimated operation event number from the response impossible estimation DB 131 based on the class determined by the class determination unit 121.

[Non-response log information DB]
FIG. 4 is a diagram showing a response impossible log information table of the response impossible log information DB 130. The non-response log information table includes an application ID (hereinafter referred to as an application ID for simplification), an application name (hereinafter referred to as an application name for simplification), the number of operation events when the response is impossible, and an unresponsive time and date. In addition, it includes items including information 2 on the execution environment of the application such as CPU, memory capacity, and virtual disk procedures. In addition, the item shown here is an illustration, and may not have all and may have another item. However, it is desirable that the information be considered to be related to a factor that causes the inability to respond.

  The application ID stores a symbol that uniquely identifies the application. The application ID may be arbitrarily determined by a server administrator or the like, or a commonly used application ID may be used. The application name stores the name of the application in which the response failure occurred. However, when the application name is not included in the information acquired from the client terminal 200, it is necessary to convert the application ID into the application name. Therefore, a DB that stores the correspondence between application IDs and application names is required.

  In the number of operation events when the response is disabled, the number of operation events executed until the response is disabled in the application where the response is disabled is stored. The date and time when the response failure occurred is stored as the response impossible date and time, and this date and time is Japan Standard Time or World Standard Time. The CPU, memory capacity, and virtual disk capacity store information about the CPU, memory, and virtual disk installed in the client terminal 200 where the response failure occurred. In addition to the capacity, the CPU name, manufacturer name, and the like may be included. By storing the situation when the response failure occurs, the time when the response failure occurs can be estimated statistically.

[Unresponsiveness estimation DB]
FIG. 5 is a diagram showing a response impossible estimation table of the response impossible estimation DB 131. The unresponsiveness estimation table includes items such as a class ID and an estimated number of operation events. In addition, the item shown here is an illustration, and may not have all and may have another item. The class ID is a symbol that uniquely identifies a class for classifying the unresponsive log information in the unresponsiveness estimating unit 110. The estimated number of operation events is the number of operation events estimated to cause a response failure determined by the response failure estimation unit 110. By storing the number of operation events estimated to cause inability to respond, when a new application is started, the number of operation events estimated to cause inability to respond to the application is easily presented. be able to.

[Classification of unresponsive log information]
FIG. 6 is a diagram illustrating an example of machine learning using the SVM algorithm performed by the response impossible estimation unit 110. The unresponsiveness estimation unit 110 uses unresponsiveness log information stored in the unresponsiveness log information DB 130 as learning data, and uses the SVM algorithm to learn data that is closest to the identification plane that classifies these learning data into two classes. And an identification surface that maximizes the distance between the identification surface and the identification surface. Then, the learning data closest to the identification surface is used as a support vector, and the number of estimated operation events in each class is set.

For example, the case of classifying into class 1 and class 2 will be described. In that case, as shown below, some unresponsive log information classified into class 1 and class 2 is used as learning data.
Class 1: Application: XX word, CPU: 1.5 GHz, Virtual disk capacity: 3 GB, ...
Class 2: Application: XX sheet, CPU: 1.5 GHz, Virtual disk capacity: 3 GB, ...

  Then, the unresponsiveness estimation means 110 calculates an identification surface 30 that maximizes the distance (margin) between the identification surfaces 30 that identify the class 1 learning data and the class 2 learning data. Class 1 learning data 11 and class 2 learning data 21 that are closest to the identification surface 30 are used as support vectors. In this way, the response impossible estimation means 110 uses the class 1 response impossible log information and the class 2 response impossible log information as learning data, and the class 1 response impossible log information group 10 and the class 2 response impossible log. An identification plane which is identification data for distinguishing the information group 20 is generated. The learning data 11 gives the number of estimated operation events of class 1, and the learning data 21 gives the number of estimated operation events of class 2. As described above, by using the support vector machine, it is possible to classify the unresponsive log information and obtain the estimated number of operation events of the classified class. As described above, classification of two or more classes can also be performed using a plurality of support vector machines.

[Estimation process when unable to respond]
FIG. 7 is a flowchart of the response impossible estimation process. When the application detection unit 201 detects that a new application has been started on the client terminal 200, the process starts.

  S1: When detecting that the application has been activated, the application detection unit 201 acquires at least an application ID for specifying the activated application, and makes a request for counting the number of operation events to the operation event counting unit 202.

  S2: The operation event counting means 202 that has received the count request counts the number of operation events in the application. When a plurality of applications are activated, the number of operation events for each application is counted. This is because the unresponsiveness occurs for each application, and the cause is due to an operation event performed on the application.

  S3: The interrupt detection unit 203 determines whether or not a response failure has occurred in the application. When detecting that a response failure has occurred, the interrupt detection unit 203 stores the number of operation events when a response failure has occurred in the operation event number storage unit 204. Make a save request.

  S4: On the other hand, when the interrupt detection unit 203 detects that the response failure is not detected and the user ends the application, the control unit 210 resets the operation event count of the operation event count unit 202. At this time, only the number of operation events of the terminated application is reset. This is because the number of operation events is counted for each application.

  S5: When response failure occurs in the application in step S4 and an operation event number storage request is made, the operation event number storage unit 204 executes an interrupt routine, and when the response cannot be obtained from the operation event count unit 202 Is temporarily stored in a memory or the like.

  S6: The operation event number transmission unit 205 acquires the application ID acquired in step S1, etc., the operation event number when response is temporarily stored in step S5, the performance of the client terminal 200, etc., and the communication unit. It transmits to the server apparatus 100 via 206. In some cases, the client terminal 200 is restarted in order to eliminate the inability to respond. In this case, if the client terminal 200 is restarted before the information is transmitted to the server apparatus 100 in step S6, there is a problem that the information that can be acquired by the server apparatus 100 is reduced. For this reason, when the restart of the client terminal 200 is input by the user, it is desirable that the restart of the client terminal 200 is executed by the computer after the end of step S6.

  S7: The response impossible log information recording unit 102 transmits the number of operation events when the response is impossible temporarily stored in step S5 and the performance of the client terminal 200 via the communication unit 101, such as the application ID transmitted in step S6. And is recorded in the response impossible log information DB 130. In this way, the server apparatus 100 can collect the information indicating that the response cannot be generated at the client terminal 200 connected via the network.

  S8: The class generation unit 111 of the response impossible estimation unit 110 uses the SVM algorithm described above to obtain the identification plane and the support vector using the response impossible log information stored in the response impossible log information DB 130 as learning data. . Note that the timing at which step S8 is executed may be arbitrary. For example, it may be executed every predetermined period, or may be arbitrarily executed by the server apparatus administrator. However, since the estimated number of operation events is obtained statistically, no change occurs when one or two pieces of information increase. Therefore, it is desirable to execute at a timing when information is accumulated to some extent. Note that step S8 may be executed automatically or manually by an administrator of the server device 100 or the like.

  S9: The estimated operation event number determination means 112 of the unresponsiveness estimation means 110 obtains the estimated operation event number from the support vector of each class obtained in step S8.

  S10: The response impossible estimation unit 110 stores the estimated operation event number obtained in step S9 in the response impossible estimation DB 131. At this time, the estimated number of operation events is stored in association with the class ID.

[Non-response notification process]
FIG. 8 is a flowchart of the response impossible notification process. When the application detection unit 201 detects that a new application has been activated on the client terminal 200, a process for notifying that the application cannot respond is started.

  S11: The estimated response impossible information acquisition unit 120 acquires at least the application ID and the performance of the client terminal 200 from the client terminal 200 via the communication units 206 and 101. The application ID and the like are acquired by the application detection unit 201, and the performance of the client terminal 200 is acquired by the application monitoring unit 207.

  S12: Based on the information acquired in step S11, the class determination unit 121 of the estimated response impossibility information acquisition unit 120 cannot respond to a newly started application from the identification plane created in step S8 in FIG. In this case, the class to which the response impossible log information belongs is obtained, and the class ID is specified. The class ID is specified in order to search the response impossible estimation DB 131. Note that the estimated response impossibility information acquisition unit 120 may also acquire the execution status of other applications running on the client terminal 200 in step S11, and use them to determine the class.

S13: The estimated operation event number extraction unit 122 of the estimated response impossibility information acquisition unit 120 searches the response impossibility estimation DB 131 based on the class ID specified in step S12, and acquires the estimated operation event number.
S14: The estimated response impossibility information acquisition unit 120 transmits the estimated operation event number acquired in step S13 to the application monitoring unit 207 of the client terminal 200.

  S15: The application monitoring unit 207 monitors whether or not the number of operation events in the application has reached a threshold value. The number of operation events in the application is acquired from the operation event counting means 202. As described above, the threshold is a value arbitrarily set by the system administrator or the like based on the estimated number of operation events, and is a value smaller than the estimated number of operation events. The reason why the threshold value is reached instead of the estimated number of operation events is that there is a possibility that a response failure may occur if the estimated operation event is reached.

  S16: The notification unit 208 warns the user when the number of operation events reaches a threshold value. For example, if it is a warning in a document creation application, a message “Please save” is displayed or announced. A save button may be displayed together with the message. It is not always necessary to give a warning, and the computer may automatically save without giving a warning.

[Hardware configuration diagram]
FIG. 9 is a diagram illustrating an information processing apparatus 500 as a hardware configuration example of the client terminal 200 and the server apparatus 100 that configure the response impossible time estimation system according to the present embodiment. Hereinafter, an example of the hardware configuration of the information processing apparatus 500 will be described. The information processing apparatus 500 includes a CPU (Central Processing Unit) 310 (a plurality of CPUs such as a CPU 320 may be added in a multiprocessor configuration), a bus line 420, a communication I / F (I / F: An interface) 330, a main memory 340, a BIOS (Basic Input Output System) 350, an I / O controller 360, a hard disk 370, an optical disk drive 380, and a semiconductor memory 390. The hard disk 370, the optical disk drive 380, and the semiconductor memory 390 are collectively referred to as a storage device 410.

  The control unit 300 is a part that controls the information processing apparatus 500 in an integrated manner. By appropriately reading and executing various programs stored in the hard disk 370 (described later), the control unit 300 cooperates with the hardware described above. Various functions related to are realized.

  The communication I / F 330 is a network adapter when the information processing apparatus 500 transmits and receives information via the Internet. The communication I / F 330 may include a modem, a cable modem, and an Ethernet (registered trademark) adapter.

  The BIOS 350 records a boot program executed by the CPU 310 when the information processing apparatus 500 is started, a program on which the feature word selection apparatus 1 depends on hardware, and the like.

  A storage device 410 such as a hard disk 370, an optical disk drive 380, and a semiconductor memory 390 can be connected to the I / O controller 360.

  The hard disk 370 stores various programs for causing the hardware to function as the information processing apparatus 500, programs for executing the functions of the present invention, and tables and records to be described later. The information processing apparatus 500 can also use a hard disk (not shown) separately provided as an external storage device.

  As the optical disk drive 380, for example, a DVD-ROM drive, a CD-ROM drive, a DVD-RAM drive, or a CD-RAM drive can be used. In this case, the optical disk 400 corresponding to each drive is used. A program or data can be read from the optical disk 400 by the optical disk drive 380 and provided to the main memory 340 or the hard disk 370 via the I / O controller 360.

  Note that the computer in the present invention refers to an information processing device including a storage device, a control unit, and the like, and the information processing device 500 includes an information processing device including a storage device 410, a control unit 300, and the like. The information processing apparatus is included in the concept of the computer of the present invention.

It is a figure which shows the whole structure of the estimation system at the time of an impossible response which concerns on this embodiment. It is a figure which shows the detail of an unresponsiveness estimation means. It is a figure which shows the detail of an estimated response impossibility information acquisition means. It is a figure which shows the response impossible log information table of response impossible log information DB. It is a figure which shows the response impossible estimation table of response impossible estimation DB. It is a figure which shows an example of the machine learning using the algorithm of SVM by an unresponsiveness estimation means. It is a flowchart of the estimation process at the time of a response impossible. It is a flowchart of a response impossible notification process. It is the figure which showed the information processing apparatus as a hardware structural example of the client terminal and server apparatus which comprise the response impossible time estimation system which concerns on this embodiment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Response impossible estimation system 100 Server apparatus 101 Communication means 102 Response impossible log information recording means 110 Response impossible estimation means 120 Estimated response impossible information acquisition means 130 Response impossible log information DB
131 Response impossible estimation DB
DESCRIPTION OF SYMBOLS 200 Client terminal 201 Application detection means 202 Operation event count means 203 Interrupt detection means 204 Operation event number storage means 205 Operation event number transmission means 206 Communication means 207 Application monitoring means 208 Notification means 210 Control means

Claims (9)

A system that estimates when an application is unresponsive,
Application detection means for detecting the activation of the application;
An operation event counting unit that counts the number of operation events including at least the number of keystrokes from the user performed on the application in response to detection of activation by the application detection unit;
Interrupt detection means for detecting occurrence of an unresponsive state of the application;
In response to the detection of the inability to respond to the application by the interrupt detection means, the number of operation events calculated from the start of the application and the information on the execution environment from the start of the application are responded by interrupt processing. Response impossible log information recording means for storing in association with the application as impossible log information;
Based on the information on the execution environment of the application stored by the unresponsive log information recording unit and the number of operation events, the unresponsive log information is classified to estimate the estimated number of operation events to be unresponsive. Means,
A system comprising:   In response to the activation of the application, the number of the estimated operation events is determined based on the execution environment information of the application, and the number of the operation events has reached a predetermined threshold that is smaller than the number of the estimated operation events. The system according to claim 1, further comprising notification means for notifying a user of the application to that effect.   The system according to claim 2, wherein an execution state of another application is used as an element for determining the estimated operation event number together with information on an execution environment of the application.   The system according to any one of claims 1 to 3, wherein the unresponsiveness estimation unit classifies the unresponsiveness log information by a support vector machine using the estimated number of operation events as a support vector. Furthermore, in response to the detection of the inability to respond to the application detected by the interrupt detection means, an operation event number storage means for storing the operation event number calculated from the start of the application by interrupt processing;
The number of operation events stored in the operation event number storage means, and the number of operation event transmission means for transmitting information on the execution environment from the time of starting the application,
The system according to claim 1, comprising:   The system according to any one of claims 1 to 5, wherein the interrupt detection unit determines that the response is disabled when the application is stopped or when the processing of the application is delayed.   The system according to any one of claims 1 to 6, wherein the information on the execution environment includes information on performance of an apparatus in which the application is activated and information on an application that is activated at the same time. A method for estimating when an application is unresponsive by a computer,
Detecting activation of the application;
Counting the number of operation events including at least the number of keystrokes from the user performed on the application;
Detecting the occurrence of an unresponsive state of the application;
In response to the detection of the inability to respond to the application, the number of operation events calculated from the start of the application by interrupt processing and the information on the execution environment from the start of the application are used as the response impossible log information. Associating and storing, and
Classifying the unresponsive log information based on information on the execution environment of the application and the number of operation events, and estimating an estimated number of operation events that result in unresponsiveness;
Including methods. A program that causes a computer to estimate when an application cannot respond,
Detecting activation of the application;
Counting the number of operation events including at least the number of keystrokes from the user performed on the application;
Detecting the occurrence of an unresponsive state of the application;
In response to the detection of the inability to respond to the application, the number of operation events calculated from the start of the application by interrupt processing and the information on the execution environment from the start of the application are used as the response impossible log information. Associating and storing, and
Classifying the unresponsive log information based on information on the execution environment of the application and the number of operation events, and estimating an estimated number of operation events that result in unresponsiveness;
A program that executes

Images