JP2014229044A - Program, recording medium and device for increasing speed of data acquisition of data processor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To increase the speed of the data acquisition of a data processor to be performed in accordance with an existing program without adding any correction to the program.SOLUTION: A data processing system 1 includes: processing means 11 for performing data processing in accordance with an application program 111; first storage means 14 as the original storage place of data to be used for processing by the processing means 11; second storage means 15 to be used as a cache memory; and redirect means 16 for redirecting a data request to be output from the processing means 11 to the first storage means 14 to the second storage means 15. Specification means 12 specifies data satisfying a relation indicated by first condition data 121 corresponding to the application program 111 with the data being currently used by the processing means 11 as prefetch object data, and acquisition means 13 acquires the prefetch object data from the first storage means 14, and allows the second storage means 15 to store the prefetch object data.

Description

  The present invention relates to a technique for speeding up data acquisition of a data processing apparatus.

  When a data processing device such as a computer performs processing according to a program, it is necessary to acquire data used for the processing from a storage unit within the own device or an external storage device. If the time required to acquire this data can be shortened, the process completion time can be advanced. Hereinafter, shortening the time required for data acquisition is referred to as “acceleration of data acquisition”. One technique for speeding up data acquisition is a technique called prefetch.

  Prefetch generally refers to a second storage location that requires only a short time to read data predicted to be used in the future from a first storage location that requires a long time to read (for example, an external storage device). (For example, when it is necessary to copy data to the internal cache memory in advance and acquire data used for processing in the data processing apparatus, the second storage location stores the data in the second storage location. This is a technique for speeding up data acquisition by reading out the data from a storage location.

  As a document disclosing a technique using prefetch, for example, there is Patent Document 1. In Patent Document 1, the use efficiency of a cache memory is improved by adjusting the size and number of data to be prefetched according to the nature of an application program that instructs data processing (for example, the number of processing branches). The mechanism is disclosed.

JP-A-7-129464

  In order to speed up data acquisition by software prefetching (hereinafter simply referred to as “prefetch”) in processing in a data processing apparatus, a program that instructs data processing is described to perform prefetching, or prefetching is performed. It is necessary to be described so as to use an external program (for example, a program module included in the OS). Hereinafter, a program described to speed up data acquisition by prefetching is referred to as a “program having a prefetch function”.

  Therefore, if the user feels dissatisfied with the processing speed of the data processing apparatus that operates according to a program that does not have a prefetch function, a new program of the same type having a prefetch function (for example, a new version of the same title program) is obtained. It is necessary to replace the existing program with a new program. However, if the same kind of program having a prefetch function is not provided, the user has no choice but to continue to use the existing program and speed up the processing of the data processing apparatus.

  Even if the same kind of program with a prefetch function is provided, there is a cost to purchase a new program, or inconveniences associated with the use of a new program (user interface changes, data incompatibility) , Performance degradation due to lack of specifications of data processing device, increase in frequency of freeze due to bugs, etc.). Thus, the use of new programs is not always a desirable option for users.

  The present invention has been made in view of the above-described background, and an object of the present invention is to provide means for speeding up data acquisition of a data processing apparatus performed according to a program without modifying an existing program.

  In order to solve the above-described problem, the present invention provides, as one aspect, an acquisition unit that acquires data, a processing unit that performs processing using data acquired by the acquisition unit, and a storage unit that stores data. And acquiring data by the acquisition means according to a first program, storing the data acquired by the acquisition means according to the first program by the storage means, and the processing means according to a second program being one When performing the process using data, if the one data is stored in the storage unit, the one data is acquired from the storage unit by the acquiring unit, and the one data is stored in the storage unit. Is not stored, the second program is stored in a computer that acquires the one data from a location different from the storage unit by the acquiring unit. In accordance with the second program between the process of acquiring metadata, which is data indicating the data used for the process by the processing unit according to the program, and the data indicated by the metadata acquired in the process of acquiring A process of specifying data satisfying a defined relationship as data having a high probability of being used for future processing by the processing means, and a process of acquiring data specified in the specifying process according to the first program Provide a program to be executed.

  In the above program, the processing means performs processing according to each of the plurality of second programs, and the predetermined processing corresponding to each of the plurality of second programs in the specifying process is performed on the computer. Data satisfying the predetermined relationship according to the second program instructing the processing means to use the data indicated by the metadata among the relations for future processing by the processing means. A configuration in which data is specified as data having a high probability of being used may be employed.

  In the above program, the processing means performs processing using data arranged on a time axis in accordance with the second program, and the computer indicates data indicated by the metadata in the specifying processing. A configuration may be adopted in which data satisfying a predetermined positional relationship on the time axis is specified as data having a high probability of being used for future processing by the processing means.

  Further, in the above program, the second program between the data acquired by the acquisition unit according to the first program and stored in the storage unit, with the data indicated by the metadata. A configuration may be adopted in which processing for instructing the storage means to release data from the storage means satisfying a predetermined relationship according to the above may be employed.

  Further, in the above program, the processing means performs processing according to each of the plurality of second programs, and the processing according to each of the plurality of second programs in the instructing processing to the computer is performed in advance. Among the predetermined relationships, the storage unit is instructed to release data that satisfies the predetermined relationship according to the second program that instructed the processing unit to use the data indicated by the metadata. A configuration of making it possible may be adopted.

  In another aspect, the present invention provides a computer comprising: an acquisition unit that acquires data according to a program; and a processing unit that performs processing using the data acquired by the acquisition unit according to the program. Processing for acquiring data satisfying a predetermined rule corresponding to the program as prefetch data, processing for storing the prefetch data, and acquisition unit acquiring one data When the request is output to one output destination, the acquisition request is intercepted, and when the one data is stored as the prefetch data, the one data stored as the prefetch data is the acquisition request. As a response to the acquisition means, and the one data is transferred to the pre-fetch. If not stored as data provide a program for executing a process of outputting the acquisition request to the one of the destination.

  Moreover, this invention provides the recording medium which records the said program continuously so that reading to a computer as another one aspect | mode.

  In addition, the present invention provides a first acquisition unit that acquires data, a storage unit that stores data acquired by the first acquisition unit, a processing unit that performs processing using data according to a program, and the program When the processing means performs processing using one data, if the one data is stored in the storage means, the one data is acquired from the storage means, and the storage means stores the one data in the storage means. A data processing apparatus comprising: a second acquisition unit that acquires the one data from a location different from the storage unit when the one data is not stored, and delivers the acquired one data to the processing unit Is a device that exchanges data with and acquires metadata that is data indicating data used for processing by the processing means according to the program, and Data satisfying a predetermined relationship according to the program with the data indicated by the data is specified as data having a high probability of being used for future processing by the processing means, and acquisition of the specified data is performed An apparatus including a specifying unit that instructs a first acquiring unit is provided.

  Moreover, this invention is provided with the process means which performs the process using data according to a program as another aspect, The 1st acquisition means which acquires the data used for a process by the said process means, and delivers to the said process means A second acquisition device that transfers data to and from a data processing device, and acquires data that satisfies a predetermined relationship according to the program with data used for processing by the processing means Means for storing the data acquired by the second acquisition means, and when the first acquisition means outputs a data acquisition request to one output destination, the acquisition request is intercepted. When the one data is stored in the storage unit, the first data stored in the storage unit is used as a response to the acquisition request. Passing the resulting unit, in the case where the one of the data is not stored in said storage means to provide a device and a redirection means for outputting the acquisition request to the one of the destination.

  According to the present invention, based on data used for processing performed in accordance with a program that does not have a prefetch function, data that satisfies a predetermined relationship with the data is likely to be used in processing performed in accordance with the program in the future. Then, the specified data is prefetched. Therefore, the data acquisition in the processing according to the program is speeded up without modifying the program without the prefetch function.

It is the figure which showed the hardware constitutions of the data processing system concerning one Embodiment. It is the figure which showed the function structure of the data processing system concerning one Embodiment. It is a figure for demonstrating the example of the conditions used for specification of prefetch object data and release object data in the data processing system concerning one Embodiment. It is the figure which showed the data structure of the cache table used in the data processing system concerning one Embodiment.

[Embodiment]
Hereinafter, a data processing system 1 according to an embodiment of the present invention will be described with reference to the drawings. The data processing system 1 is a system having a mechanism for acquiring data used for processing performed in accordance with an application program not having a prefetch function at a high speed by prefetching in accordance with a plug-in program or the like.

  FIG. 1 is a diagram schematically illustrating a hardware configuration of the data processing system 1. The data processing system 1 includes a computer 10 that is a general-purpose computer, an operation device 21 such as a keyboard and a mouse that allow the user to input data to the computer 10 by outputting a signal corresponding to the user's operation to the computer 10, a computer 10, a display device 22 such as a liquid crystal display that displays an image according to image data output from the computer 10, a storage device 23 </ b> A such as an HDD or an optical drive connected to the computer 10 without going through the network, and data to the computer 10 through the network. Data communication between the computer 10 and the storage device 23B, such as a storage device 23B such as a data distribution server device (hereinafter referred to as “storage device 23” if they are not distinguished from each other) A network that provides a route It is equipped with a click 24. Note that the number of storage devices 23A and 23B is arbitrary.

  The computer 10 is a general computer having communication means, and includes a processor 101, a memory 102, an input / output I / F (Interface) 103, and a communication I / F 104. These can exchange data with each other via the bus 109.

  The processor 101 is a CPU, for example, and performs various data processing and controls other components of the computer 10 according to various programs stored in the memory 102. The memory 102 includes, for example, a nonvolatile memory 1021, an SSD 1022, and an HDD 1023, but these configurations can be arbitrarily changed. The memory 102 stores various programs executed by the processor 101 and various data used in processing by the processor 101, and is also used as a work area for storing data temporarily generated in processing by the processor 101.

  The input / output I / F 103 is connected to an external device by, for example, a wired connection through a cable such as a USB terminal or an HDMI (registered trademark) terminal and a wireless connection according to a communication standard such as Bluetooth (registered trademark). Input and output various data between them. The communication I / F 104, for example, transmits various data to and from an external device via the network 24 by wired connection according to a communication standard such as IEEE802.3 or wireless connection according to a communication standard such as IEEE802.11. Send and receive.

  FIG. 2 is a diagram schematically illustrating a functional configuration of the data processing system 1. That is, when the processor 101 of the computer 10 shown in FIG. 1 performs processing according to various programs stored in the memory 102, the data processing system 1 having the hardware configuration shown in FIG. It will function as a system with the components shown.

  Hereinafter, the functional configuration of the data processing system 1 and the operation of each functional component will be described with reference to FIG. The data processing system 1 is used for processing by processing means 11 that performs data processing according to an application program, identification means 12 that performs data processing according to a first plug-in program, acquisition means 13 that performs data processing according to a service program, and processing means 11 First storage means 14 and second storage means 15 for storing data, redirection means 16 for performing data processing according to the driver program, and instruction means 17 for performing data processing according to the second plug-in program are provided.

  The processing means 11 performs various data processing according to each of n (where n is an arbitrary natural number) application programs (application programs 111-1 to 111-n). Hereinafter, when the application programs 111-1 to 111-n are not distinguished, they are referred to as “application programs 111”.

  The type of the application program 111 may be any type, such as a moving image editing application, a music editing application, or a word processor application. However, for convenience of the following description, it is assumed that the application program 111-1 is a moving image editing application.

  The specifying unit 12 outputs a metadata request for requesting metadata, which is data indicating data currently used by the processing unit 11, to the processing unit 11, for example, every elapse of a predetermined time. Metadata is acquired from the processing means 11 as a response to the metadata request.

  For example, the metadata identifies the data used by the processing means 11 and uses a file name indicating an attribute such as the type of the data, a URI indicating the storage location (original storage location) of the data, and the data. The application name that identifies the application program 111 instructing the processing to be performed is included.

  Note that the storage location (original storage location) identified by the URI included in the metadata is any location of the nonvolatile memory 1021, the SSD 1022 and the HDD 1023, the storage device 23A, or the storage device 23B constituting the memory 102. However, generally, the HDD 1023, the storage device 23A, or the storage device 23B, which has a low access speed but a large storage capacity, is used as the original storage location.

  The specifying unit 12 is a processing unit based on the metadata acquired from the processing unit 11 according to the first condition data 121-1 to 121-n stored in advance according to each of the application programs 111-1 to 111-n. 11 specifies data that has a high probability of being used for processing in the near future. Hereinafter, when the first condition data 121-1 to 121-n are not distinguished, they are referred to as “first condition data 121”. Further, data that is specified by the specifying unit 12 and that has a high probability of being used for processing in the near future by the processing unit 11 is hereinafter referred to as “prefetch target data”. The data currently used by the processing means 11 indicated by the metadata is hereinafter referred to as “in-use data”.

  The first condition data 121 is data indicating a relationship between in-use data and prefetch target data. For example, the first condition data 121-1 corresponding to the application program 111-1 (moving image editing application) includes “the time for reproducing the moving image based on the image data currently used in the processing according to the application program 111-1. Processing m pieces of image data (m is an arbitrary natural number) adjacent in the positive direction on the axis (a series of m pieces of image data reproduced following the currently used image data) in the near future Data indicating a condition “specify as data having a high probability of being used” is included.

  FIG. 3 is a diagram for explaining the conditions exemplified above (and examples of conditions used for specifying release target data described later). “1021.bmp”, “1022.bmp”,... Shown in FIG. 3 are file names of image data representing still images constituting one moving image. The extension “.bmp” of these file names indicates an attribute that the data is data representing a bitmap format still image, and a numerical value such as “1021” indicates the order on the time axis in the reproduction of a moving image. Show.

  In the example of FIG. 3, the processing unit 11 is currently performing processing using image data “1023.bmp” constituting one moving image data in accordance with the application program 111-1. The moving image data is composed of a total of 500 pieces of image data “1001.bmp” to “1500.bmp”, and these image data are URI “http://xyz.com/footage/video/3256”. It can be obtained from the memory location specified by

  In the case of the example in FIG. 3, the specifying unit 12 first uses the metadata acquired from the processing unit 11 to use the data currently used by the processing unit 11 in the processing according to the application program 111-1 (the in-use data related to the application program 111-1. ) Is “1023.bmp”. Subsequently, a total of m series of data such as “1024.bmp”, “1025.bmp”,... Is specified as data satisfying the relationship indicated by the above-described condition with the specified in-use data. . The data specified in this way is data (prefetch target data related to the application program 111-1) that has a high probability of being used in the near future by the processing means 11 in the processing according to the application program 111-1.

  When specifying the prefetch target data, the specifying unit 12 outputs an acquisition instruction for instructing acquisition of the specified prefetch target data to the acquiring unit 13. The acquisition instruction includes, for example, a file name for identifying prefetch target data, a URI for identifying the storage location of the prefetch target data, and an application name for identifying the corresponding application program 111.

  When receiving the acquisition instruction from the specifying unit 12, the acquisition unit 13 outputs (or transmits) a data request to the first storage unit 14 according to the URI included in the acquisition instruction. The data request includes a URI and a file name included in the acquisition instruction. It should be noted that the URI included in the request data is converted into an address indicating a specific data storage location at an appropriate timing as necessary, and is transferred to the first storage unit 14 according to the address obtained by the conversion. .

  When the first storage unit 14 receives a data request from the acquisition unit 13, the first storage unit 14 reads out the requested data according to the data request and outputs it to the acquisition unit 13. When the acquisition unit 13 acquires data from the first storage unit 14 as a response to the data request, the acquisition unit 13 outputs an instruction to store the data to the second storage unit 15. The storage instruction includes data acquired by the acquisition unit 13 from the first storage unit 14 and metadata of the data (file name, URI, corresponding application name, etc.).

  The second storage unit 15 is a storage unit that can read and write data faster than the first storage unit 14. The second storage unit 15 may be realized by using any one of the nonvolatile memory 1021, the SSD 1022 and the HDD 1023, the storage device 23A, or the storage device 23B constituting the memory 102, but generally has a small storage capacity. However, the nonvolatile memory 1021 or the SSD 1022 having a high access speed is used to implement the second storage unit 15.

  When the second storage unit 15 receives the storage instruction from the specifying unit 12, the second storage unit 15 stores the data instructed according to the storage instruction as cache data. The second storage unit 15 stores cache tables 151-1 to 151-n corresponding to the application programs 111-1 to 111-n, respectively. Hereinafter, when the cache tables 151-1 to 151-n are not distinguished, they are referred to as “cache tables 151”. When the second storage unit 15 newly stores the cache data, the second storage unit 15 registers the data related to the cache data in the cache table 151 corresponding to the application program 111 corresponding to the cache data.

  FIG. 4 is a diagram schematically showing the data structure of the cache table 151. The cache table 151 is a collection of data records corresponding to each of the data stored as cache data in the second storage means 15, and each data record has data fields "URI", "file name", and "address". Is provided. The data field “URI” stores a URI (URI indicating the original storage location of data) included in the storage instruction. The data field “file name” stores the file name included in the storage instruction. In the data field “address”, an address (for example, the head address of data) for identifying the storage location of data in the storage area provided in the second storage unit 15 is stored.

  The processing unit 11 outputs (or transmits) a data request for requesting the data to the first storage unit 14 when performing processing using data that has not yet been acquired in accordance with the application program 111. In the data request, a URI indicating the original storage location of the data specified by the processing unit 11 by processing according to the application program 111, the file name of the data, and processing using the data are instructed. The application name of the application program 111 is included.

  The redirect unit 16 intercepts the data request output from the processing unit 11 to the first storage unit 14. That is, the data request output to the first storage unit 14 is intercepted by the redirect unit 16 and does not reach the first storage unit 14. The redirect unit 16 outputs the intercepted data request to the second storage unit 15. The second storage unit 15 searches the cache table 151 corresponding to the application name included in the data request received from the redirect unit 16 for a data record storing the URI and file name included in the data request.

  If the search of the data record is successful, the second storage means 15 reads the data from the address indicated in the data field “address” of the searched data record and outputs it to the redirect means 16. On the other hand, if the search of the data record fails, the second storage unit 15 outputs an error notification indicating that the requested data is not stored as cache data in the second storage unit 15 to the redirect unit 16.

  When the redirect unit 16 receives data from the second storage unit 15 as a response to the data request output to the second storage unit 15, the redirect unit 16 intercepts the data request (output from the processing unit 11 to the first storage unit 14). The data acquired from the second storage unit 15 is output to the processing unit 11 as a response to the response to the processing unit 11. At that time, the redirect unit 16 pretends that the output source of the data is not the redirect unit 16 but the first storage unit 14 that is the original output destination of the data request, if necessary. Accordingly, the processing unit 11 acquires the data output from the redirect unit 16 as the data output (or transmitted) from the first storage unit 14.

  On the other hand, when the redirect unit 16 receives an error notification from the second storage unit 15 as a response to the data request output to the second storage unit 15, the redirect unit 16 first intercepts the data request (from the processing unit 11 to the first storage unit 14. Is output (or transmitted) to the first storage means 14. At that time, the redirect unit 16 pretends that the output source of the data request is not the redirect unit 16 but the processing unit 11 that is the original output source of the data request, as necessary. Accordingly, the first storage unit 14 acquires the data output from the redirect unit 16 as a data request output (or transmitted) from the processing unit 11.

  When the first storage unit 14 receives the data request output (or transmitted) from the redirect unit 16, the first storage unit 14 reads the data in response to the data request and outputs (or transmits) the data to the processing unit 11 as a response to the data request. ) The processing unit 11 acquires data output (or transmitted) from the first storage unit 14.

  As described above, the processing unit 11 responds to the data request output (or transmitted) to the first storage unit 14 from the redirect unit 16 or the first storage unit 14 disguised as the first storage unit 14. When data is acquired, processing according to the application program 111 is executed using the data.

  The instruction unit 17 stores the data stored in the second storage unit 15 so that the storage capacity is not insufficient when new data is stored as cache data in accordance with the storage instruction received by the second storage unit 15 from the acquisition unit 13. Among them, the second storage unit 15 is instructed to release data having a low probability of being used in the near future by the processing unit 11. That is, the instructing unit 17 uses a storage area used for storing data with a low probability of being used in the near future out of the storage areas of the second storage unit 15 as a storage area that may be used for storing other data. specify.

  For this purpose, for example, the instruction unit 17 outputs a metadata request to the processing unit 11 every time a predetermined time elapses, for example, and the metadata (currently used) is sent from the processing unit 11 as a response thereto. Show data). In addition, the instruction unit 17 refers to the cache table 151 stored in the second storage unit 15 every time a predetermined time elapses, for example, in advance according to each of the application programs 111-1 to 111-n. According to the stored second condition data 171-1 to 171-n, data having a low probability of being used for processing in the near future by the processing unit 11 is specified among the data stored in the second storage unit 15. Hereinafter, when the second condition data 171-1 to 171-n are not distinguished, these are referred to as “second condition data 171”. Further, data specified by the instruction unit 17 and having a low probability of being used for processing by the processing unit 11 in the near future is hereinafter referred to as “release target data”.

  The second condition data 171 is data indicating a relationship between in-use data and release target data. For example, the second condition data 171-1 corresponding to the application program 111-1 (moving image editing application) includes “time for reproducing a moving image based on image data currently used in processing according to the application program 111-1. Processing p pieces of image data (p is an arbitrary natural number) adjacent in the negative direction on the axis (a series of p pieces of image data reproduced prior to the currently used image data) in the near future The data indicating the condition of “specify as data with a low probability of being used in” is included.

  In the case of the example of FIG. 3, the instruction unit 17 first specifies that the data in use related to the application program 111-1 is “1023.bmp” based on the metadata acquired from the processing unit 11. Subsequently, the instruction means 17 has the data field “URI” of the cache table 151-1 as “http://xyz.com/footage/video/3256”, and the data field “file name” indicates the specified use. A data record indicating a file name that satisfies the relationship indicated by the above-described condition with the middle data is extracted. As a result, for example, data records relating to a total of p series of data such as..., “1021.bmp”, “1022.bmp” are extracted. Data corresponding to the data record extracted in this way is release target data related to the application program 111-1.

  When the instruction unit 17 specifies the release target data, the instruction unit 17 outputs a release instruction for instructing the release of the specified release target data to the second storage unit 15. The release instruction includes the contents of the data record previously extracted from the cache table 151.

  When receiving the release instruction from the instruction unit 17, the second storage unit 15 deletes the data record included in the release instruction from the cache table 151. Thus, when the second storage unit 15 stores new data in accordance with the storage instruction acquired from the acquisition unit 13, the second storage unit 15 refers to the cache table 151 and is used for any cache data registered in the cache table 151. New data is stored in a storage area that does not exist.

  The above is the description of the configuration and operation of the data processing system 1. According to the data processing system 1 described above, in accordance with the application program 111, data that is likely to be used for processing by the processing unit 11 in the near future is prefetched to the second storage unit 15 in advance, and the first storage is performed when the data is used. Data is transferred from the second storage means 15 capable of reading data at a higher speed than the means 14 to the processing means 11. As a result, data acquisition is speeded up with a higher probability than when no prefetching is performed.

  What should be noted here is that data acquisition can be speeded up without modification of the application program 111. That is, in the data processing system 1, the prefetch target data and the release target data are not specified by the processing unit 11 but are specified by the specifying unit 12 or the instruction unit 17 that operates independently of the processing unit 11. Therefore, the processing unit 11 that performs processing according to the application program 111 does not require any change in operation as long as it can output metadata to the specifying unit 12 or the instruction unit 17 in response to a metadata request.

  In addition, the effect of speeding up data acquisition by prefetching depends greatly on the conditions under which prefetch target data and release target data are specified. In general, different application programs 111 have different conditions that should be adopted in order to obtain a high effect. In the data processing system 1, in accordance with the first condition data 121 corresponding to each of the plurality of application programs 111, data satisfying a predetermined relationship with the data in use is used as prefetch target data, and a plurality of application programs In accordance with the second condition data 171 corresponding to each of the data 111, data satisfying a predetermined relationship with the data in use is specified as release target data. Therefore, the prefetching effect for each of the various application programs 111 can be enhanced only by appropriately preparing the first condition data 121 and the second condition data 171 for each application program 111.

[Modification]
The above-described embodiments can be variously modified within the scope of the technical idea of the present invention. Examples of these modifications are shown below. These modifications may be combined as appropriate.

  In the above-described embodiment, the identification unit 12, the acquisition unit 13, the redirect unit 16, and the instruction unit 17 have been described as individual components, but two or more of these may be configured as one component. . For example, the specifying unit 12 and the instruction unit 17 may be configured as a single component, or the specifying unit 12 and the acquisition unit 13 may be configured as a single component. Further, any one of the specifying unit 12, the acquiring unit 13, the redirecting unit 16, and the instruction unit 17 may be configured by two components that operate in cooperation with each other. For example, the instruction unit 17 may be configured by a configuration unit that specifies release target data and a configuration unit that instructs the second storage unit 15 to release the release target data specified by the configuration unit. That is, the distinction between the specifying unit 12, the acquiring unit 13, the redirecting unit 16, and the instruction unit 17 is an example, and the processing performed by these constituent units may be modified to be shared by the constituent units of other configurations. Therefore, the program used to implement the specifying unit 12, the acquiring unit 13, the redirecting unit 16, and the instruction unit 17 may not be configured by four programs corresponding to these components, and the configuration may be variously modified. Is possible.

  In the above-described embodiment, the first condition data corresponding to each application program 111 is used by one specifying unit 12 that is not distinguished for each application program 111, and the second condition data is not distinguished for each application program 111 1 Although one instruction unit 17 is used, a configuration in which both or one of the specifying unit 12 and the instruction unit 17 is provided for each application program 111 may be employed.

  In the above description of the embodiment, the cache table 151 is managed by the second storage unit 15, but the cache table 151 may be managed by other components such as the redirect unit 16. In addition, an independent configuration unit that manages the cache table 151 may be provided separately.

  In the description of the above-described embodiment, a moving image editing application that frequently uses time-series data in one direction on the time axis in order is used as an example of the application program 111 in the description. As described above, the moving image editing application is an example of the application program 111, and various other types of application programs can be adopted as the application program 111.

  Therefore, the contents of the first condition data 121 and the second condition data 171 used according to each of the application programs 111 also prefetch data that satisfies a predetermined positional relationship on the time axis with the data in use. It is not limited to data specified as target data or release target data.

  That is, in a certain application program, if the probability that the second attribute data is used is statistically significantly higher after the first attribute data is used, the first attribute data is used data. In this case, data indicating a condition for specifying the data having the second attribute as the prefetch target data can be used as the first condition data corresponding to the application program. Similarly, in a certain application program, after the first attribute data is used, if the probability that the second attribute data is used is statistically significantly low, the first attribute data is used data. In some cases, data indicating a condition for specifying the data having the second attribute stored in the second storage unit 15 as the release target data can be used as the second condition data corresponding to the application program.

  In addition, regarding the processing performed by the processing unit 11 in the past according to the application program 111, the data processing system 1 uses the first attribute data according to each of the application programs 111, for example, before a predetermined time elapses. In addition, condition updating means for updating a condition indicated by one or both of the first condition data and the second condition data based on the probability that the data of the second attribute is used may be provided.

  Further, when specifying prefetch target data according to the first condition data 121 or specifying release target data according to the second condition data 171, the priority of each of a plurality of candidate data is calculated according to a predetermined rule. These data specifying methods can be variously modified, such as specifying prefetch target data and release target data in descending order of priority. For example, consider assigning high points to conditions of high importance, assigning low points to conditions of low importance, and using the total value of points assigned to satisfying conditions as an index of priority. It is done.

  In the description of the above-described embodiment, the first condition data related to the moving image editing application is exemplified as a condition for specifying data adjacent in use in the positive direction on the time axis as prefetch target data. In addition, when the data in use is at the end on the time axis (for example, still image data indicating the last frame of the moving image data), for example, a predetermined number of data located at the beginning on the time axis is to be prefetched Conditions specified as data may be indicated. When processing is performed according to an application program that handles time-series data, such as editing video data, when the data used for processing reaches the end on the time axis, the data is processed again sequentially from the beginning on the time axis. This is because there are many cases.

  In the above-described embodiment, the specifying unit 12 and the instructing unit 17 are configured to acquire the metadata indicating in-use data from the processing unit 11. A configuration in which data is acquired by another method may be employed. For example, the data indicated by the data request that the processing unit 11 outputs to the first storage unit 14 is highly likely to be data that the processing unit 11 uses from now. Therefore, the specifying unit 12 or the instruction unit 17 obtains the data request output from the processing unit 11 directly or via the redirect unit 16 and the like, and regards the data requested by the data request as in-use data. (That is, the URI and file name included in the data request are used as metadata).

  In the present application, when “data in use” or “data used for processing” is referred to, the data does not necessarily have to be strictly used at that time. Data that is actually used within the period, data that is scheduled to be used (data that can be used by the processing means), and the like.

  In the above-described embodiment, the specifying unit 12, the acquiring unit 13, the redirecting unit 16, and the instruction unit 17 are realized as functional components by the processor 101 performing processing according to the program. One or more of these components may be realized by so-called hardware. In this case, various types of data are transferred between a device including a configuration unit realized by hardware and a computer including a processing unit 11 that performs processing according to at least the application program 111.

  In the above-described embodiment, the programs (first plug-in program, service program, driver program, second plug-in program) used to realize the specifying unit 12, the acquiring unit 13, the redirecting unit 16, and the instruction unit 17 are used. At least a part of the program is provided in the form of being downloaded to the computer 10 via a network, and is distributed in the form of a computer-readable recording medium for continuously recording these programs. It may be provided in read form.

DESCRIPTION OF SYMBOLS 1 ... Data processing system, 10 ... Computer, 11 ... Processing means, 12 ... Identification means, 13 ... Acquisition means, 14 ... First storage means, 15 ... Second storage means, 16 ... Redirection means, 17 ... Instruction means, 21 DESCRIPTION OF SYMBOLS Operation device 22 ... Display device 23 ... Storage device 24 ... Network 101 ... Processor 102 ... Memory 103 ... Input / output I / F 104 ... Communication I / F 109 ... Bus 111 ... Application program 121 ... first condition data 151 ... cache table 171 ... second condition data 1021 ... nonvolatile memory 1022 ... SSD 1023 ... HDD

Claims (9)

An acquisition means for acquiring data;
Processing means for performing processing using the data acquired by the acquisition means;
Storage means for storing data,
Acquire data by the acquisition means according to the first program,
Storing the data acquired by the acquisition means according to the first program by the storage means;
When the processing means performs processing using one data according to the second program, when the one data is stored by the storage means, the one data is stored from the storage means by the obtaining means. If the one data is not stored in the storage means, the computer acquires the one data from a location different from the storage means by the acquisition means.
Processing for obtaining metadata that is data indicating data used for processing by the processing means according to the second program;
Data that satisfies a predetermined relationship according to the second program with data indicated by the metadata acquired in the processing to be acquired is data that has a high probability of being used for future processing by the processing means. Process to identify,
A program for executing a process of acquiring data specified in the specified process according to the first program. The processing means performs processing according to each of the plurality of second programs,
In the computer,
In the specifying process, the second instruction instructing the processing means to use the data indicated by the metadata among the predetermined relationship corresponding to each of the plurality of the second programs. The program according to claim 1, wherein data satisfying the predetermined relationship according to the program is specified as data having a high probability of being used for future processing by the processing means. The processing means performs processing using data arranged on a time axis according to the second program,
In the computer,
In the specifying process, data satisfying a predetermined positional relationship on the time axis with the data indicated by the metadata is specified as data having a high probability of being used for future processing by the processing means. The program according to claim 1 or 2. In the computer,
Of the data acquired by the acquisition unit according to the first program and stored in the storage unit, the relationship predetermined by the second program is satisfied with the data indicated by the metadata The program according to any one of claims 1 to 3, wherein a program for instructing the storage means to release data from the storage means is executed. The processing means performs processing according to each of the plurality of second programs,
In the computer,
In the instructing process, out of the predetermined relationships corresponding to each of the plurality of the second programs, the second instructing the processing means to use the data indicated by the metadata. The program according to claim 4, wherein the storage unit is instructed to release data satisfying the predetermined relationship according to the program. Acquisition means for acquiring data according to a program;
In a computer comprising processing means for performing processing using data acquired by the acquisition means according to the program,
A process of acquiring data satisfying a rule determined in advance according to the program as data to be used for processing by the processing means as prefetch data;
Processing to store the prefetch data;
When the acquisition means outputs an acquisition request for one data to one output destination, the acquisition request is intercepted, and when the one data is stored as the prefetch data, it is stored as the prefetch data. And processing to output the acquisition request to the one output destination when the one data is not stored as the prefetch data. Program to make.   The recording medium which records the program of any one of Claim 1 thru | or 6 so that a computer can read continuously. First acquisition means for acquiring data;
Storage means for storing data acquired by the first acquisition means;
Processing means for performing processing using data according to a program;
When the processing means performs processing using one data according to the program, if the one data is stored by the storage means, the one data is acquired from the storage means, and the storage means The second acquisition means for acquiring the one data from a location different from the storage means when the one data is not stored in the storage means and delivering the acquired one data to the processing means A device that exchanges data with a processing device,
Acquiring metadata that is data indicating data used for processing by the processing means according to the program, and satisfying a predetermined relationship according to the program with the data indicated by the metadata, An apparatus comprising: specifying means for specifying data having a high probability of being used for future processing by a processing means and instructing the first acquisition means to acquire the specified data. Processing means for performing processing using data according to a program;
An apparatus for transferring data to and from a data processing apparatus comprising: a first acquisition means for acquiring data used for processing by the processing means and delivering the data to the processing means;
Second acquisition means for acquiring data satisfying a predetermined relationship according to the program with data used for processing by the processing means;
Storage means for storing data acquired by the second acquisition means;
When the first acquisition means outputs an acquisition request for one data to one output destination, the acquisition request is intercepted, and when the one data is stored in the storage means, the storage means The stored one data is handed over to the first acquisition means as a response to the acquisition request, and if the one data is not stored in the storage means, the acquisition request is sent to the one output destination. A redirecting means for outputting.