JP2018180959A - Data processing program, data processing apparatus, data processing method and data processing system - Google Patents

Abstract

To provide a data processing program etc. with improved response.
A data processing program estimates a second request to be called after the first request to a computer that receives a first request from a client, and a minimum processable process is performed based on processing of the second request. A process data unit is obtained, the process of the first request is performed in the process data unit, and the process obtained by the first request completed every time the process of the first request in the process data unit is completed The processing for processing the second request is executed on the processing data unit obtained in the above.
[Selected figure] Figure 6

Description

  The present invention relates to a data processing program using a Web Application Program Interface (Web API).

  In recent years, Web API servers have become widespread (Patent Document 1). The Web API server is a server computer that implements the Web API. The Web API is an API that exchanges data based on HTTP (HyperText Transfer Protocol). In a general Web API using HTTP, data exchange is performed by simple request / response.

JP, 2010-146169, A

  Since data exchange is performed by simple request / response, the response is degraded when a plurality of processes are performed consecutively.

  In one aspect, the object of the present invention is to provide a data processing program or the like with improved response.

  The data processing program disclosed in the present application estimates the second request to be called after the first request to the computer that receives the first request from the client, and can process the smallest request based on the processing of the second request. A processing data unit is acquired, the processing of the first request is performed in the processing data unit, and each time the processing of the first request in the processing data unit is completed, the processing is performed by the processing of the completed first request. Based on the processing of the second request on the processing data unit, the minimum process that can be processed is executed.

  According to one aspect of the present invention, the response can be improved.

It is an explanatory view showing composition of a data processing system. It is a block diagram showing the example of composition of Web API server. It is a block diagram showing an example of composition of a client terminal. It is explanatory drawing which shows an example of the record layout of data unit DB. It is explanatory drawing which shows an example of the record layout of log | history DB. It is explanatory drawing which shows the outline of the process which a Web API server performs. It is explanatory drawing which shows the outline of the process which a Web API server performs. It is a flowchart which shows the procedure of the process which a Web API server performs. It is an explanatory view showing an example of temporary data created and updated in processing of a Web API server. It is a flowchart which shows the example of a procedure of an estimation process. It is a block diagram showing an example of a function with which a Web API server is provided.

  Hereinafter, embodiments will be described with reference to the drawings. In the following description, stream processing refers to processing in which a plurality of processes are sequentially performed, and refers to a series of processes or processes included in the series of processes where the process output of the former stage is the process input of the latter stage. Here, as in the case of image data, the input data is divided, each processing is performed on the divided data, and the processing results of the divided data are combined to obtain the same result as processing the input data collectively. It assumes processing that handles data that can be obtained. A combination of such processing is called stream processing.

Embodiment 1
FIG. 1 is an explanatory view showing the configuration of a data processing system. The data processing system includes a Web API server (data processing device) 1, a client terminal 2, and a client terminal 3. The Web API server 1 and the client terminal 2 are communicably connected via the network N1. The Web API server 1 and the client terminal 3 are communicably connected via the network N1 and the wireless network N2. Hereinafter, the client terminal 2 will be described as the client terminal, but the same applies to the client terminal 3.

  FIG. 2 is a block diagram showing a configuration example of the Web API server. The Web API server 1 includes a central processing unit (CPU) 11, a read only memory (ROM) 12, a random access memory (RAM) 13, a large-capacity storage unit 14, a communication unit 15, and a reading unit 16. Each configuration is connected by a bus B.

  The CPU 11 controls each part of the hardware according to the control program 1 P stored in the ROM 12. The RAM 13 is, for example, an SRAM (Static RAM), a DRAM (Dynamic RAM), or a flash memory. The RAM 13 temporarily stores data generated when the CPU 11 executes a program.

  The large-capacity storage unit 14 is, for example, a hard disk or a solid state drive (SSD). The large capacity storage unit 14 stores necessary various data. The large capacity storage unit 14 stores a data unit DB (Data Base) 141 and a history DB 142. Alternatively, the control program 1P may be stored in the large capacity storage unit 14.

  The communication unit 15 communicates with the Web API server via the network N1. The reading unit 16 reads a portable storage medium 1 a including a CD (Compact Disc) -ROM and a DVD (Digital Versatile Disc) -ROM. The CPU 11 may read the control program 1 P from the portable storage medium 1 a via the reading unit 16 and store the control program 1 P in the large-capacity storage unit 14. Alternatively, the CPU 11 may download the control program 1 P from another computer via the network N 1 or the like and store the control program 1 P in the large-capacity storage unit 14. Furthermore, the CPU 11 may read the control program 1P from the semiconductor memory 1b.

  FIG. 3 is a block diagram showing a configuration example of the client terminal 2. The client terminal 2 includes a CPU 21, a ROM 22, a RAM 23, a large capacity storage unit 24, a display unit 25, an input unit 26, a communication unit 27 and a reading unit 28. Each configuration is connected by a bus B.

  The CPU 21 controls each part of the hardware according to the control program 2P stored in the ROM 22. The RAM 23 is, for example, an SRAM, a DRAM or a flash memory. The RAM 23 temporarily stores data generated when the CPU 21 executes a program.

  The large capacity storage unit 24 is, for example, a hard disk or an SSD. The large capacity storage unit 24 stores various data necessary for the processing of the CPU 21. The control program 2P may be stored in the large capacity storage unit 24.

  The display unit 25 is a liquid crystal display device. The display unit 25 displays the result of data processing and the like. The input unit 26 is a keyboard or a mouse. Further, the input unit 26 may be a touch panel integrated with the display unit 25.

  The communication unit 27 communicates with the Web API server 1 via the network N1. The reading unit 28 reads the portable storage medium 2 a including a CD-ROM and a DVD-ROM. The CPU 21 may read the control program 2 P from the portable storage medium 2 a via the reading unit 28 and store the control program 2 P in the large-capacity storage unit 24. In addition, the CPU 21 may download the control program 2P from another computer via the network N1 or the like and store the control program 2P in the large-capacity storage unit 24. Furthermore, the CPU 21 may read the control program 2P from the semiconductor memory 2b.

  Next, a DB (Data Base: database) stored by the large-capacity storage unit 24 will be described. FIG. 4 is an explanatory view showing an example of the record layout of the data unit DB 141. As shown in FIG. The data unit DB 141 stores the minimum unit of processing data when executing each processing provided by the Web API server 1 as stream processing. The data unit DB 141 includes an item number column, a processing column, and a minimum data unit column. The item number column stores a unique item number that identifies a record. The item number is also an ID for specifying each process. The processing sequence stores the processing name specifying the processing. The minimum data unit column stores the minimum data unit (processing data unit) when stream processing each processing.

  FIG. 5 is an explanatory view showing an example of the record layout of the history DB 142. As shown in FIG. The history DB 142 stores a history of processing executed by the Web API server in the past. The history DB 142 includes an item number sequence, an access date and time sequence, a processing sequence, an input sequence, and an output sequence. The item number column stores an item number that can uniquely identify the history. The access date and time column stores the date and time when each process was requested from the client terminal 2. The access date and time column may be any other one as long as it indicates the temporal context of each process. The execution date and time of each process, and the date and time when each process returned the process result to the client terminal 2 may be used. The process sequence stores a process name specifying the executed process. The input string stores input data of each process. The output sequence stores output data of each process. The input sequence and the output sequence are assumed to be able to estimate the context of continuous processing. For example, if the output of process A is data 2 and the input of process B is data 2, it can be inferred that process B is performed after process A. The history DB 142 may store an ID for identifying the client terminal 2 or application that has requested each process, such as a Mac (Media Access Control) address or a session ID. As a result, even if a plurality of client terminals 2 and the like request the same processing at the same time, it is possible to accurately grasp the context of the request.

  Next, processing performed by the Web API server will be described. FIG. 6 is an explanatory view showing an outline of processing performed by the Web API server. FIG. 6 shows the case where the client terminal 2 requests a plurality of processes sequentially. And the processing result returned to the client terminal 2 with respect to each request is shown as a response. The request contains processing, input data. Request 1 is a request to perform processing 1 on data 1. The request 2 is a request for performing the process 2 on the data 2. The request 3 is a request for performing the process 3 on the data 3. The response contains output data. The output data of response 1 is data 2. The output data of response 2 is data 3. The output data of response 3 is data 4. Requests 1 to 3 shown in FIG. 6 are stream processing in which the output data of the former stage is the input data of the latter stage. API 1 to 3 indicate APIs corresponding to processing. API 1 is an API corresponding to process 1. API 2 is an API corresponding to process 2. API 3 is an API corresponding to process 3. An example of the data format of the request is REST (Representational State Transfer). An example of the data format of the response is XML (Extensible Markup Language) or JSON (JavaScript (registered trademark) Object Notation).

  The example shown in FIG. 6 is an example in which, when the Web API server 1 receives the request 1 which is the first request from the client terminal 2, the processing 2 and the processing 3 are estimated as the post-stage processing. And that is the case when the estimate is true. Therefore, the first input data 1 is divided into the minimum data units, and the processing 1 is performed in the minimum data units. Processing 2 and processing 3 are also processed in divided data units. Further, according to the estimation result, the Web API server 1 starts executing the process 2 before the request 2 is made. Similarly, the Web API server 1 starts executing the process 3 before the request 3 is made. After receiving the request 2, the Web API server 1 returns a response to the process 2 to the client terminal 2. Similarly, after receiving the request 3, the Web API server 1 returns a response to the process 3 to the client terminal 2. Since processing 2 and processing 3 are executed in advance, if processing is completed before request 2 and request 3 are requested, the Web API server 1 may return a response immediately after receiving the request. It is possible.

  FIG. 7 is an explanatory view showing an outline of processing performed by the Web API server 1. Similar to FIG. 6, FIG. 7 shows an outline of the processing from the client terminal 2 to the Web API server 1 and the processing of the Web API server 1 in response to the request. The description format of FIG. 7 is the same as that of FIG.

  The example shown in FIG. 7 is an example in which, when the Web API server 1 receives a request 1 which is the first request from the client terminal 2, the processing 2 and the processing 3 are estimated as the post-processing. However, this is the case when the estimation deviates. Even in this case, the first input data 1 is divided into the minimum data units, and the processing 1 is performed in the minimum data units. The estimated post-stage processing, processing 2 and processing 3 are also processed in divided data units. Further, according to the estimation result, the Web API server 1 starts executing the process 2 before the request 2 is made. The Web API server 1 returns the response 1 to the client terminal 2 after the processing 1 is completed for the divided data 1. On the other hand, the client terminal 2 transmits the request 2 a to the Web API server 1. The input data included in the request 2 a is data 2. Since the input data of request 2 a is data 2, request 2 is a request subsequent to request 1. Further, the process included in the request 2 a is process 4. Then, since the process after the process 1 is the process 2 in the estimation of the Web API server 1, when the request 2a is received, it is determined that the estimation is out.

  Therefore, the Web API server 1 stops the process 2 and the process 3 among the stream processes already started. The Web API server 1 performs the process 4 specified by the request 2 a on the data 2 which is the output of the process 1 and obtains the data 4 as a result of the process 4. The WEbAPI server 1 returns the response 2a including the data 4 as an output to the client terminal 2.

  Next, the processing procedure will be described with reference to flowcharts of the processes shown in FIGS. FIG. 8 is a flowchart showing the procedure of processing performed by the Web API server 1. The process shown in FIG. 8 is executed each time the Web API server 1 receives a request from the client terminal 2. The CPU 11 of the Web API server 1 acquires the process and the terminal information of the client terminal 2 from the received request (step S1). The terminal information includes, for example, an IP (Internet Protocol) address of the client terminal 2, a MAC address, an authentication ID, a session ID, and the like. The CPU 11 determines whether the request is the first (step S2). The first request is a request that triggers stream processing to start. The determination as to whether the request is the first request is performed as follows, for example. The CPU 11 determines whether the session with the client terminal 2 that transmitted the request is valid or not, whether the reception date and time of the immediately preceding request received from the client terminal 2 is within a predetermined time, or the like. If there is no valid session with the client terminal 2, the CPU 11 determines that it is the first request. If there is a valid session with the client terminal 2, the CPU 11 determines that it is not the first request. Also, if the immediately preceding request received from the client terminal 2 is within a predetermined time, for example, within 15 minutes, it is determined to be the first request, otherwise it is determined not to be the first request.

  When the CPU 11 determines that the received request is the first (YES in step S2), it estimates post-stage processing (step S3). The estimation of the post-stage processing will be described later. The CPU 11 determines the minimum unit of stream processing based on the estimation result (step S4). The CPU 11 determines a stream process including a plurality of processes based on the estimation result. The CPU 11 stores the contents of the stream processing as temporary data in a temporary storage area provided in the RAM 13 or the like. The CPU 11 acquires the minimum unit of each process included in the stream process from the data unit DB 141. The CPU 11 compares the magnitudes of the acquired data units, and sets the largest data unit as the minimum unit of stream processing (processing data unit). The CPU 11 divides the input data according to the determined minimum unit, and starts stream processing with the divided data (step S5). The CPU 11 ends the process.

  If the CPU 11 determines that the received request is not the first one (NO in step S2), it determines whether the process included in the received request is as estimated in step S3 (step S6). For example, the CPU 11 uses the above-described temporary data to determine whether or not it is as estimated. When the CPU 11 determines that the processing is as estimated (YES in step S6), it determines whether stream processing is being processed (step S7). If the CPU 11 determines that the stream processing is not in progress (NO in step S7), the CPU 11 shifts the processing to step S9. If the CPU 11 determines that the stream processing is being processed (YES in step S7), the CPU 11 determines whether the processing corresponding to the request has ended (step S8). When the CPU 11 determines that the process corresponding to the request has not ended (NO in step S8), the CPU 11 repeatedly executes step S8 and waits for the end of the process. When the CPU 11 determines that the process corresponding to the request is completed (YES in step S8), the CPU 11 transmits a response to the client terminal 2 (step S9). The CPU 11 determines whether the stream processing has been interrupted or whether the processing has been completed to the end (step S10). If the CPU 11 determines that the stream processing has been interrupted or determines that the processing has been completed to the end (YES in step S10), the above-described temporary data is deleted (step S11). If the CPU 11 determines that the stream processing has not been interrupted and the processing has not been completed to the end (NO in step S10), the processing ends.

  When the CPU 11 determines that the processing is not as estimated (NO in step S6), it determines whether or not stream processing is being processed (step S12). If the CPU 11 determines that the stream processing is not in progress (NO in step 12), the CPU 11 shifts the processing to step S14. When the CPU 11 determines that the stream processing is being processed (YES in step S12), the stream processing is stopped (step S13). The CPU 11 executes a process corresponding to the request (step S14). The CPU 11 executes step S9 and subsequent steps.

  FIG. 9 is an explanatory view showing an example of temporary data 231 created and updated in the processing of the Web API server. The temporary data 231 includes a client ID 231a and a processing table 231b. The processing table 231 b includes an item number column, a processing column, an input column, an output column, a status column, and a response column. The item number column stores a number indicating the order of processing. The processing sequence stores the processing name specifying the processing. Not the process name but the process ID may be used. The input string stores input data. The output string stores output data. The status column stores the status of each process. The status is, for example, unprocessed, under processing, or processed. The response string stores whether or not a response has been returned to the client terminal 2. For example, the response string stores "done" if a response is returned. The response string stores "not yet" if no response is returned.

  In the example shown in FIG. 9, the process for the first request is process 1. The processes estimated as the subsequent stages are processes 2 and 3. With regard to processing 1 and processing 2, processing for all divided data has been completed. Only process 3 is in process. The response to the request including process 1 has been completed. In this situation, when the request including the process 2 is received, since the process 2 is completed, the CPU 11 can immediately return a response. When a request including the process 3 is received, the process waits for the completion of the process 3 and returns a response. Furthermore, in this situation, when a request including the process 4 is received, the stream process is stopped, the process 2 is changed, and the process 4 is executed.

  Subsequently, estimation processing will be described. The estimation process is a process performed in step S3 of FIG. FIG. 10 is a flowchart showing an example of the procedure of the estimation process. The CPU 11 sets the process included in the first request as the search target process (step S21). The CPU 11 searches the history DB 142 for the same process as the search target process (step S22). The CPU 11 selects the history of hits in the search (step S23). The CPU 11 uses the output data of the process of the selected history and searches the history DB 142 for a process (following candidate process) whose access date and time is close to the process of the history (step S24). The CPU 11 temporarily stores the search result (step S25). The CPU 11 determines whether or not there is an unprocessed history (step S26). If the CPU 11 determines that there is an unprocessed history (YES in step S26), the process returns to step S23. If the CPU 11 determines that there is no unprocessed history (NO in step S26), it determines whether there is a subsequent candidate process (step S27). That is, it is determined whether one or more records hit as a result of repeating step S24. If the CPU 11 determines that there is a candidate process (YES in step S27), one of the candidate processes is determined as an estimated subsequent process (step S28). The CPU 11 sets the determined process as a search target process (step S29). The CPU 11 returns the process to step S22. If the CPU 11 determines that there is no candidate process (NO in step S27), the process ends, assuming that there is no subsequent process.

  The contents of the estimation process will be described again using an example. It is assumed that the process included in the first request is "image decoding". The CPU 11 sets the search target processing to "video decoding" (step S21). The CPU 11 searches the history DB 142 for a record whose processing sequence is "image decoding" (step S22). The CPU 11 acquires item numbers 1, 5, 7 and other histories as search results. The CPU 11 sequentially selects item numbers 1, 5, 7 and other records (step S23), and searches for candidate processes that can be subsequent (step S24). For example, for the record of item No. 1, the record of item No. 2 is hit. For the item No. 5 record, the item No. 6 record is hit. For the item No. 7 record, the item No. 8 record is hit. As a search result, the CPU 11 temporarily stores the records of the item numbers 2, 6 and 8 (step S25). If there is an unprocessed record among the records hit in step S22 (YES in step S26), the CPU 11 repeats step S23 and subsequent steps. If all the records have been processed (NO in step S26), the CPU 11 determines the presence or absence of candidate processing (step S27). Here, at least the records of item numbers 2, 6 and 8 are stored as the candidate process (YES in step S27), the CPU 11 determines which item number the process included in is to be the subsequent process ( Step S28). Here, since only the item numbers 1, 5 and 7 indicate the history of “image decoding”, it is difficult to determine the subsequent processing, but it is assumed that there are many other histories. Therefore, there are a large number of subsequent candidate processes for all the "video decoding" histories. For example, the CPU 11 obtains the frequency of the subsequent candidate process, and determines the one with the largest frequency as the estimated subsequent process. For example, the CPU 11 determines “noise removal” as the estimation subsequent processing. The CPU 11 sets “noise removal” as the search handling process (step S29). The CPU 11 executes step S22. As a result of step S29, history records such as item numbers 2 and 8 are hit. The CPU 11 performs step S23 and subsequent steps, and further estimates the subsequent processing. When the candidate process is not found (NO in step S27), the CPU 11 ends the estimation process. If the number of subsequent processes is large, it may be cut off halfway. For example, the number of subsequent processes is 10 at maximum, and the estimation process ends when the number of subsequent processes reaches 10.

  The effects of the present embodiment will be described. Conventionally, even if the Web API server performs continuous processing according to a series of requests, the Web API server is in a waiting state for a subsequent request after returning a response to the request. That is, if the processing of the previous stage is completed and the next request does not arrive, the processing of the subsequent stage is not executed. Therefore, for example, in the case of image processing that often performs a plurality of continuous processes, the throughput is significantly reduced. A reduction in throughput leads to a delay in response.

  When receiving the first request from the client terminal 2, the Web API server 1 of the present embodiment estimates the subsequent processing. The Web API server 1 processes the request as stream processing based on the estimation result. Each process included in the stream process is executed prior to the subsequent request. If the process is completed before the subsequent request is made, the response time for the subsequent request is only the communication time. As a result, it is possible to increase the throughput and improve the response speed to the client. The stream processing is effective particularly when the number of continuous processing is large or when the data size is large.

  FIG. 11 is a block diagram showing an example of the function of the Web API server. The Web API server 1 includes a request reception unit 11a, an estimation unit 11b, an acquisition unit 11c, a unit execution unit 11d, and a return unit 11e. These function units are realized by the CPU 11 operating based on the control program 1P.

  The request receiving unit 11a receives the first request and the second request. The estimation unit 11 b estimates a second request to be called after the first request. The acquisition unit 11c acquires the smallest processable data unit that can be processed, based on the processing of the second request. The unit execution unit 11d executes the processing of the first request in units of processing data. Furthermore, each time the processing of the first request in units of processing data is completed, the unit execution unit 11 d executes the processing of the second request estimated on the unit of processing data obtained in the processing of the completed first request. The return unit 11e returns the processing result of the first request by the unit execution unit 11d.

The technical features (component requirements) described in the respective embodiments can be combined with each other, and by combining, new technical features can be formed.
It should be understood that the embodiments disclosed herein are illustrative in all respects and not restrictive. The scope of the present invention is indicated not by the meaning described above but by the claims, and is intended to include all the modifications within the meaning and scope equivalent to the claims.

  Further, the following appendices will be disclosed regarding the above embodiment.

(Supplementary Note 1)
On the computer receiving the first request from the client,
Estimating a second request to be called after the first request,
Based on the processing of the second request, obtain the smallest processing data unit that can be processed,
Processing the first request in units of the processing data;
A data processing program for executing processing of the second request on the processing data unit obtained by processing of the completed first request, every time processing of the first request in the processing data unit is completed. .

(Supplementary Note 2)
The data processing program according to Appendix 1, wherein a processing data unit associated with the estimated second request is acquired from a storage unit that stores the second request and the processing data unit in association with each other.

(Supplementary Note 3)
If the request called by the client following the first request is different from the estimated second request, the processing of the estimated second request is aborted, and the processing of the called request is executed. Or the data processing program as described in Supplementary Note 2.

(Supplementary Note 4)
The data processing program according to any one of appendices 1 to 3, wherein estimation of the second request is performed using a history storage unit that stores requests called in the past.

(Supplementary Note 5)
The history storage unit stores the called time, input data and output data in association with the called request,
Determine before and after the time when the two requests were called,
If the output data of the previous request is the input data of the later request, it generates pair data in which the first request is the first request and the second request is the second request,
The frequency of the second request corresponding to each first request is calculated from the generated pair data,
The data processing program according to Appendix 4, wherein the second request with the highest frequency is the second request estimated for the first request.

(Supplementary Note 6)
The data processing program according to any one of appendices 1 to 5, wherein the processing result of the first request is returned to the client when the processing of the first request is completed for all of the data processing units.

(Appendix 7)
A receiver for receiving a first request from a client;
An estimation unit configured to estimate a second request to be called after the first request;
An acquisition unit for acquiring a minimum processable data processing unit based on the processing of the second request;
A unit execution unit that executes the processing of the first request in units of the processing data;
The unit execution unit executes the processing of the second request on the processing data unit obtained by the processing of the completed first request, every time the processing of the first request in the processing data unit is completed. Data processing unit.

(Supplementary Note 8)
The computer is
Receive the first request from the client,
Estimating a second request to be called after the first request,
Based on the processing of the second request, obtain the smallest processing data unit that can be processed,
Execute processing of the first request in units of the processing data;
Data processing of executing processing of the second request on the processing data unit obtained by processing of the completed first request, every time processing of the first request in the processing data unit is completed Method.

(Appendix 9)
A transmitter for transmitting the first request and the second request;
A client having a result receiving unit that receives the processing result of the first request; and a request receiving unit that receives the first request and the second request;
An estimation unit configured to estimate a second request to be called after the first request;
An acquisition unit for acquiring a minimum processable data processing unit based on the processing of the second request;
A unit execution unit that executes processing of the first request in units of the processing data;
A data processing apparatus comprising: a return unit for returning the processing result of the first request by the unit execution unit;
The unit execution unit executes the processing of the second request on the processing data unit obtained by the processing of the completed first request, every time the processing of the first request in the processing data unit is completed. ,
The data processing system, wherein the transmitting unit transmits the second request after the result receiving unit receives the processing result of the first request.

1 server 11 CPU
11a request reception unit 11b estimation unit 11c acquisition unit 11d unit execution unit 11e return unit 12 ROM
13 RAM
14 Large-capacity storage unit 141 Data unit DB
142 History DB
15 communication unit 16 reading unit 1P control program 1a portable storage medium 1b semiconductor memory 2, 3 client terminal

Claims (6)

On the computer receiving the first request from the client,
Estimating a second request to be called after the first request,
Based on the processing of the second request, obtain the smallest processing data unit that can be processed,
Processing the first request in units of the processing data;
A data processing program for executing processing of the second request on the processing data unit obtained by processing of the completed first request, every time processing of the first request in the processing data unit is completed. . The data processing program according to claim 1, wherein a processing data unit associated with the estimated second request is acquired from a storage unit that stores the second request and the processing data unit in association with each other. The data processing program according to claim 1 or 2, wherein estimation of the second request is performed using a history storage unit that stores a request called in the past. A receiver for receiving a first request from a client;
An estimation unit configured to estimate a second request to be called after the first request;
An acquisition unit for acquiring a minimum processable data processing unit based on the processing of the second request;
A unit execution unit that executes the processing of the first request in units of the processing data;
The unit execution unit executes the processing of the second request on the processing data unit obtained by the processing of the completed first request, every time the processing of the first request in the processing data unit is completed. Data processing unit. The computer is
Receive the first request from the client,
Estimating a second request to be called after the first request,
Based on the processing of the second request, obtain the smallest processing data unit that can be processed,
Execute processing of the first request in units of the processing data;
Data processing of executing processing of the second request on the processing data unit obtained by processing of the completed first request, every time processing of the first request in the processing data unit is completed Method. A transmitter for transmitting the first request and the second request;
A client having a result receiving unit that receives the processing result of the first request; and a request receiving unit that receives the first request and the second request;
An estimation unit configured to estimate a second request to be called after the first request;
An acquisition unit for acquiring a minimum processable data processing unit based on the processing of the second request;
A unit execution unit that executes processing of the first request in units of the processing data;
A data processing apparatus comprising: a return unit for returning the processing result of the first request by the unit execution unit;
The unit execution unit performs processing of the second request estimated for the processing data unit obtained by processing of the completed first request every time processing of the first request in the processing data unit is completed. Run
The data processing system, wherein the transmitting unit transmits the second request after the result receiving unit receives the processing result of the first request.

Images