JP2011043968A - Batch job processing device and batch job processing system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a batch job processing technique capable of avoiding an abnormal high-load state independently of the contents of batch processing to be input. <P>SOLUTION: In a batch job processing device, a batch job execution part executes only a batch job of a previously determined load level, and until execution of the batch job finishes, the batch job execution part does not execute a next batch job. <P>COPYRIGHT: (C)2011,JPO&amp;INPIT

Description

  The present invention relates to a batch job processing apparatus and a batch job processing system.

  In an information system having a client terminal and a server, the client terminal may request the server to execute batch processing. Generally, batch processing takes time until execution is completed. For this reason, it is difficult to immediately execute the batch process when a batch process execution request is received from the client terminal and immediately return a response result to the client terminal.

  Therefore, there is a case in which, when there is a request for execution of batch processing, only the execution schedule or the like is registered in the server, and the batch processing is executed again on the server side later. A system that links online batch process execution requests and offline batch execution processes in this way is called an online batch system.

  The online batch system can be configured using devices such as a client terminal, a Web server, a database, a batch job scheduling server, and a batch execution server. Under the above configuration, the client terminal transmits the contents of the batch processing and the execution schedule to the Web server, and the Web server stores the information in the database. The batch job scheduling server acquires the contents and execution schedule of batch processing from the database, and schedules job execution according to the execution schedule. The batch execution server executes batch processing according to the schedule.

  According to the online batch system as described above, the online batch processing execution request is completed until information is stored in the database, and the batch processing itself is executed asynchronously with the online processing. Therefore, an offline process can also be executed while a response is immediately returned to an online batch process execution request, so that the online process and the offline process can be linked.

  As known technical documents related to the present invention, there are Patent Document 1, Patent Document 2, and Patent Document 3 below.

JP 07-049837 A Japanese Patent Application Laid-Open No. 07-244653 JP 2000-207482 A

  In general, the load on the batch processing system changes as the contents of the batch job and the data amount change. However, when requesting batch processing, the processing load may not be considered at all, or may be considered only to the extent of an empirical rule.

  Therefore, when a batch processing request with a high load is generated continuously, the high-load batch processing is continuously registered in the job processing queue. In general, other jobs are in a waiting state until the job execution is completed. Therefore, when the above situation occurs, the job execution module is occupied for a long time by high-load batch processing. For example, if such a situation occurs during the peak hours of online work, not only will the work be hindered, but it may also lead to a system down.

  The present invention has been made to solve the above-described problems, and provides a batch job processing technique capable of avoiding an abnormally high load state regardless of the contents of the input batch processing. With the goal.

  In the batch job processing apparatus according to the present invention, the batch job execution unit executes only a batch job at a predetermined load level, and does not execute the next batch job until the execution of the batch job is completed.

  In the batch job processing apparatus according to the present invention, the batch job execution unit executes only one batch job of the load level targeted by itself. As a result, even if high-load batch jobs are continuously submitted, the number of high-load batch jobs that are simultaneously executed can be suppressed within a predetermined limit. Therefore, an abnormally high load state can be avoided regardless of the contents of the batch process to be input.

1 is a configuration diagram of a batch job processing system 1000 according to Embodiment 1. FIG. It is a figure which shows the structure and data example of a schedule table 340. It is a figure which shows the structure of a parameter table 350, and a data example. It is a flow explaining operation | movement of the Web server 200. FIG. It is an operation | movement flow of the batch job acquisition part 310. FIG. 6 is a configuration diagram of a batch job processing system 1000 according to Embodiment 2. FIG. 3 is a schematic diagram showing an internal configuration of a job queue 360. FIG. 10 is an operation flow of the batch job acquisition unit 310 according to the second embodiment. 10 is an operation flow of the batch job execution unit 320 according to the second embodiment. FIG. 10 is a configuration diagram of a batch job processing system 1000 according to a third embodiment. FIG. 10 is a diagram illustrating details of a part corresponding to step S905 in FIG. 9 in the operation flow of the batch job execution unit 320 according to the third embodiment. FIG. 10 is a configuration diagram of a batch job processing system 1000 according to a fifth embodiment.

<Embodiment 1>
FIG. 1 is a configuration diagram of a batch job processing system 1000 according to Embodiment 1 of the present invention. The batch job processing system 1000 is an online batch system that receives a batch job execution request and executes the batch job, and includes a client terminal 100, a Web server 200, and a batch job processing apparatus 300.

  The client terminal 100 is an operation terminal used for a user to submit a batch job processing request to the batch job processing apparatus 300. The user accesses the Web server 200 using, for example, a Web browser, inputs job contents, load level, execution schedule, and the like on the Web page, and transmits them to the Web server 200.

  The Web server 200 includes a batch job reception unit 210. The batch job reception unit 210 receives a batch job processing request from the client terminal 100 and stores it in a database 330 described later. The batch job reception unit 210 can be configured as a Web application using a technique such as CGI (Common Gateway Interface) or Servlet.

  The batch job processing device 300 is a device that executes a batch job processing request submitted by a user using the client terminal 100, and includes a batch job acquisition unit 310, a batch job execution unit 320, and a database 330.

  The batch job acquisition unit 310 acquires a batch job execution schedule from a schedule table 340 described later stored in the database 330, and instructs the batch job execution unit 320 to execute the batch job according to the execution schedule. .

  The batch job execution unit 320 is an integrated body of three batch job execution units 321, 322, and 323. Hereinafter, these are collectively referred to as a batch job execution unit 320. The batch job execution unit 320 receives an instruction from the batch job acquisition unit 310 to execute the batch job, acquires the contents of the batch job from a parameter table 350 (to be described later) stored in the database 330, and executes the batch job. To do.

  The database 330 stores a schedule table 340 and a parameter table 350 described later with reference to FIGS.

  The batch job acquisition unit 310 and the batch job execution unit 320 can be configured by using hardware such as a circuit device that realizes these functions, or an arithmetic device such as a CPU (Central Processing Unit) and its operation. It can also be configured using prescribed software. In the latter case, each unit can be implemented in the form of a software module or the like.

  The database 330 can be configured by using a data file that records data of each table and a writable storage device such as an HDD (Hard Disk Drive).

  FIG. 2 is a diagram illustrating a configuration of the schedule table 340 and data examples. The schedule table 340 is a table that holds an execution schedule of batch job processing requests received from the client terminal 100. Each record holds an execution schedule of an individual batch job.

  The schedule table 340 includes a “No” column 341, an “execution date / time” column 342, a “type” column 343, a “load level” column 344, a “batch ID” column 345, a “result output destination” column 346, and a “status” column. 347, has a “valid” column 348.

  The “No” column 341 is a serial number for identifying each record in the table.

  The “execution date and time” column 342 holds the scheduled execution date and time of the batch job.

  The “type” column 343 indicates the type of batch job. When the value of this column is “normal”, the batch job is reserved and executed as an offline batch. When the value of this column is “immediate”, the batch job is immediately executed. In addition, it may be possible to register a type such as “periodic” that is periodically activated at the same time.

  The “load level” column 344 indicates the level of processing load of the batch job. A higher value in this column means a higher processing load. In the first embodiment, “load level = 3” has the highest processing load, and “load level = 1” has the lowest processing load. “Load level = 0” means that the value of the load level is undefined.

  The “batch ID” column 345 holds an ID for identifying the batch job.

  The “result output destination” column 346 holds a file name to which the execution result of the batch job is output. This column may be empty for batch jobs that do not require output of execution results.

  The “status” column 347 indicates the execution result of the batch job. This column can take values such as “success”, “failure”, “unexecuted”, “running”, and “waiting”. For records with “load level = 0”, the value in this column is set to “standby”.

  The “valid” column 348 indicates whether or not the batch job is valid. If the value in this column is “valid”, the batch job is executed at the date and time specified by the value in the “execution date and time” column 342. When canceling the execution of a job once reserved, the user instructs the value in this column to be “invalid” via the Web server 200.

  FIG. 3 is a diagram illustrating the configuration of the parameter table 350 and data examples. The parameter table 350 is a table that holds information for specifying the contents of a batch job. Each record holds the contents of an individual batch job. FIG. 3 exemplifies a column configuration for designating the contents of a batch job that aggregates the sales status of products in a designated period. For batch jobs that execute other processing contents, a parameter table 350 having a different column configuration may be provided separately.

  The parameter table 350 illustrated in FIG. 3 includes a “No” column 351, a “Totalization start date” column 352, a “Totalization end date” column 353, a “Product ID” column 354, a “Department name” column 355, and a “Section name”. A column 356 is provided.

  The “No” column 351 is a foreign key that refers to the “No” column 341 in the schedule table 340. Records whose values in this column match the “No” column 341 hold information about the same batch job.

  The “aggregation start date” column 352 holds the start date of the period for which sales status is aggregated.

  The “aggregation end date” column 353 holds the end date of the period for which the sales status is aggregated.

  The “product ID” column 354 holds an ID for identifying a target product for which the sales situation is tabulated.

  The “Department name” column 355 and the “Section name” column 356 hold department names that handle the target product.

  The configuration of the batch job processing system 1000 has been described above. Next, a method in which the batch job execution unit 320 suppresses the system load will be described.

  Each batch job execution unit 321 to 323 has a predetermined load level of a batch job to be executed. Here, the load level in charge of the batch job execution unit 321 is 1, the batch job execution unit 322 is 2, and the batch job execution unit 323 is 3. The batch job execution unit 321 executes only the load level 1 batch job, the batch job execution unit 322 executes only the load level 1-2 batch job, and the batch job execution unit 323 executes the load level 1-3 batch job. It is assumed that everything is set to be executed.

  For example, it is assumed that the same user has submitted a batch job with a load level 3 to the batch job processing system 1000 in a short time. When the batch job execution units 321 to 323 are configured as described above, only one batch job at the load level 3 can be executed simultaneously by the batch job execution unit 323. Even if it is done, there is no fear that the operation resources of the batch job processing system 1000 will be occupied by these batch jobs of load level 3.

  That is, the batch job execution unit 323 executes the load level 3 batch jobs one by one in order, and the other batch job execution units 321 and 322 do not execute the load level 3 batch job. Therefore, the number of batch jobs at load level 3 is always 1 or less, and the system load can be stabilized.

  The above-described method for limiting the number of batch jobs to be executed simultaneously and the load level thereof is particularly effective in a system in which a user can submit a batch job of any load level at any time, such as an online batch system. . Since the offline batch system executes a batch process planned in advance at a scheduled time, it is possible to prepare a necessary and sufficient amount of computing resources of the system. On the other hand, in an online batch system, it is difficult to predict what kind of batch job will be submitted, so a load level stabilization method different from that in an offline batch system is desired. However, it goes without saying that the technique of the present invention is also useful in an offline batch system.

  In the foregoing, the method in which the batch job execution unit 320 suppresses the system load has been described. Hereinafter, the operation of the batch job processing system 1000 will be described.

  FIG. 4 is a flowchart for explaining the operation of the Web server 200. Hereinafter, each step of FIG. 4 will be described.

(FIG. 4: Step S400)
Using the client terminal 100, the user requests the Web server 200 to transmit a screen for registering a batch job processing request. The client terminal 100 transmits the request to the Web server 200.

(FIG. 4: Step S401)
The batch job reception unit 210 of the Web server 200 receives a request from the client terminal 100 to transmit a batch job processing request registration screen. The batch job reception unit 210 transmits a batch job processing request registration screen to the client terminal 100.

(FIG. 4: Step S402)
On the batch processing request registration screen displayed by the client terminal 100, the user inputs a batch job processing request such as the content of the batch job, the load level, and the execution schedule. The client terminal 100 transmits the input content to the Web server 200. The batch job reception unit 210 of the Web server 200 receives a batch job processing request.

(FIG. 4: Step S403)
The batch job reception unit 210 registers the batch job load level and execution schedule included in the received batch job processing request in the schedule table 340. In addition, the contents of the batch job are registered in the parameter table 350.

  FIG. 5 is an operation flow of the batch job acquisition unit 310. Hereinafter, each step of FIG. 5 will be described.

(FIG. 5: Step S501)
The batch job acquisition unit 310 acquires all records held in the schedule table 340.

(FIG. 5: Step S502)
The batch job acquisition unit 310 repeats the following steps S503 to S504 for the number of records acquired in step S501.

(FIG. 5: Step S503)
The batch job acquisition unit 310 sets the value of the “valid” column 348 to “valid” and the value of the “status” column 347 to “unexecuted” or “standby” for one of the records acquired in step S501. And whether the value in the “execution date and time” column 342 is earlier than the current time. If the record satisfies all these three conditions, the process proceeds to step S504, and if any one condition or more is not satisfied, the loop proceeds to the next record.

(FIG. 5: Step S504)
The batch job acquisition unit 310 instructs the batch job execution unit 320 corresponding to the value in the “load level” column 344 to execute the batch job. The batch job execution unit 320 (any one of 321, 322, and 323) checks the load level of the received batch job, and executes the batch job if it is a load level that can be executed by itself. If there is a batch job being executed first, the received batch job is executed after the execution of the batch job is completed.

(FIG. 5: Step S505)
The batch job acquisition unit 310 sleeps (waits without doing anything) for 15 seconds.

(FIG. 5: Step S506)
When the process of the batch job acquisition unit 310 is finished, the operation flow is finished. When the process is continued, the process returns to step S501 and the same process is repeated.

  The operation of the batch job processing system 1000 has been described above. In FIG. 5, all records are acquired in step S501 and the condition is checked in step S503. However, only records satisfying the condition may be acquired at the time of step S501.

  As described above, according to the first embodiment, the batch job execution unit 320 executes only one batch job at a preset load level. If there is a batch job being executed first, the received batch job is executed after the execution of the batch job is completed. Thus, for example, even when the user continuously submits a batch job with a high load level in a short time, it can be avoided that the batch job processing apparatus 300 is occupied by the batch job with the high load level.

  Further, according to the first embodiment, the user does not accurately grasp the load level at the time of submitting the batch job, or seems to have entered the batch job intentionally specifying an inaccurate load level. Even in this case, the load level of batch jobs executed simultaneously by the batch job processing apparatus 300 can be suppressed within a certain range by the action of the batch job execution unit 320. Accordingly, the batch job processing system 1000 can be stably operated regardless of whether the load level can be predicted and whether the load level of the batch job to be input is high or low.

  In the first embodiment, the batch job processing apparatus 300 includes a database 330 that stores a batch job execution schedule. The batch job acquisition unit 310 acquires the execution schedule from the database 330 and delivers the batch job to the batch job execution unit 320. As a result, the database 330 provides a bridge between online batch job processing requests and offline batch job execution, so that online processing and offline processing can be linked. A similar configuration can be employed in a complete offline batch system, but it is particularly useful to employ the same configuration in an online batch system.

<Embodiment 2>
FIG. 6 is a configuration diagram of a batch job processing system 1000 according to the second embodiment of the present invention. In the second embodiment, the job queue 360 is introduced between the batch job acquisition unit 310 and the batch job execution unit 320. Since other configurations are substantially the same as those of the first embodiment, the following description will focus on the differences.

  The job queue 360 uses a storage area in a storage device such as a memory device (not shown) provided in the batch job processing apparatus 300. The job queue 360 receives batch jobs in order from the batch job acquisition unit 310 and holds them in the order received. When the batch job execution unit 320 instructs to deliver a batch job, the batch job is delivered and the batch job is deleted from the job queue 360.

  The job queue 360 can be configured using hardware such as a circuit device that implements the function, or can be configured using an arithmetic device such as a CPU and software that defines the operation thereof. In the latter case, the job queue 360 can be configured using a software module such as a message queuing system.

  FIG. 7 is a schematic diagram showing the internal configuration of the job queue 360. Stored in the job queue 360 are messages (361 to 363 in FIG. 7) describing the contents and execution schedules of individual batch jobs. As a description format of each message 361 to 363, for example, a general-purpose description format such as XML (extensible Markup Language) is used for convenience of processing.

  FIG. 8 is an operation flow of the batch job acquisition unit 310 according to the second embodiment. Hereinafter, each step of FIG. 8 will be described.

(FIG. 8: Steps S801 to S803)
These steps are the same as steps S501 to S503 in FIG.

(FIG. 8: Step S804)
The batch job acquisition unit 310 inputs a batch job that satisfies the condition of step S803 to the job queue 360. The job queue 360 stores the contents of the batch job as a message described in the format described with reference to FIG.

(FIG. 8: Steps S805 to S806)
These steps are the same as steps S505 to S506 in FIG.

  FIG. 9 is an operation flow of the batch job execution unit 320 according to the second embodiment. Hereinafter, each step of FIG. 9 will be described. The operation flow shown in FIG. 9 is common to the individual batch job execution units 321 to 323. Each batch job execution unit executes the same operation flow in parallel.

(FIG. 9: Step S901)
The batch job execution unit 320 acquires a batch job (message describing the contents of the batch job) stored in the job queue 360. At this time, the job queue 360 may be locked so that other batch job execution units 320 do not access the job queue 360 at the same time.

(FIG. 9: Step S902)
The batch job execution unit 320 repeats steps S903 to S906 as many times as the number of batch jobs acquired in step S901.

(FIG. 9: Step S903)
The batch job execution unit 320 confirms whether one of the batch jobs acquired in step S901 has a load level equal to or lower than the load level that the batch job execution unit 320 is responsible for. If it is below the load level that it is in charge of, the process proceeds to step S904, and if not, the loop is advanced to the next batch job.

(FIG. 9: Step S904)
The batch job execution unit 320 takes out one batch job satisfying the condition of step S903 from the job queue 360. The extracted batch job is deleted from the job queue 360.

(FIG. 9: Step S905)
The batch job execution unit 320 executes the batch job extracted from the job queue 360 in step S904. Further, the “status” column 347 of the record of the batch job held in the schedule table 340 is updated to a value such as “success” or “failure”.

(FIG. 9: Step S906)
The batch job execution unit 320 sleeps (waits without doing anything) for 15 seconds.

(FIG. 9: Step S907)
When the process of the batch job execution unit 320 is finished, this operation flow is finished, and when the process is continued, the process returns to step S901 and the same process is repeated.

  The operation of the batch job processing system 1000 according to the second embodiment has been described above.

  As described above, according to the second embodiment, the job queue 360 is provided between the batch job acquisition unit 310 and the batch job execution unit 320, and the batch job execution unit 320 acquires the batch job from the job queue 360. It was. As a result, the batch job execution unit 320 only needs to access the job queue 360 after the previously executed batch job is completed, so that the batch job execution unit 320 itself does not need to hold the job execution waiting list. Therefore, the batch job execution unit 320 can concentrate exclusively on the execution of the batch job.

  Further, according to the second embodiment, since the batch job acquisition unit 310 and the batch job execution unit 320 are loosely coupled by the job queue 360, the batch job acquisition unit 310 or the batch job execution unit 320 is replaced with a new module. It becomes easy.

  In the second embodiment, since the job queue 360 stores a message describing the contents of the batch job and the like, it is easy to access the batch job from outside the job queue 360. For example, if the contents of a batch job are described in the XML format as illustrated in FIG. 7, the batch job can be processed in a format that is easy to handle for a computing device such as a computer, which is convenient for processing.

  In the second embodiment, the job queue 360 may be configured such that a batch job once stored cannot be deleted other than being taken out for execution. Similarly, the order may be changed. In this case, once a batch job is stored in the job queue 360, the batch job is surely delivered to the batch job execution unit 320. Therefore, an unexecuted batch job remains stored in the database 330 indefinitely. There is an effect of suppressing such a situation from occurring as much as possible.

<Embodiment 3>
FIG. 10 is a configuration diagram of a batch job processing system 1000 according to Embodiment 3 of the present invention. In the third embodiment, a load level prediction unit 370 that operates in conjunction with the batch job execution unit 320 is newly introduced. Since the other configuration is substantially the same as that of any one of the first and second embodiments, the following description will focus on the differences. 10 illustrates a configuration in which a load level prediction unit 370 is added to the configuration of the second embodiment.

  The load level prediction unit 370 predicts the load level of a batch job using an appropriate method such as the technique described in Patent Document 2, for example. The load level prediction unit 370 can be configured using hardware such as a circuit device that implements the function, or can be configured using an arithmetic device such as a CPU and software that defines the operation thereof.

  FIG. 11 is a diagram illustrating details of a portion corresponding to step S905 in FIG. 9 in the operation flow of the batch job execution unit 320 according to the third embodiment. Hereinafter, each step of FIG. 11 will be described.

(FIG. 11: Step S1101)
The batch job execution unit 320 confirms the load level of the batch job extracted from the job queue 360 in step S904. If the load level is 0 (undefined), the process proceeds to step S1104; otherwise, the process proceeds to step S1102.

(FIG. 11: Step S1102)
The batch job execution unit 320 executes the batch job extracted from the job queue 360 in step S904.

(FIG. 11: Step S1103)
The batch job execution unit 320 updates the “status” column 347 of the record of the batch job held in the schedule table 340 to a value such as “success” or “failure”.

(FIG. 11: Step S1104)
The batch job execution unit 320 instructs the load level prediction unit 370 to predict the load level of the batch job extracted from the job queue 360 in step S904. The load level prediction unit 370 predicts the load level of the batch job and notifies the batch job execution unit 320 of the prediction result.

(FIG. 11: Step S1105)
The batch job execution unit 320 updates the “load level” column 344 of the record of the batch job held in the schedule table 340 with the value of the prediction result received from the load level prediction unit 370.

  The operation of the batch job processing system 1000 according to the third embodiment has been described above. When the load level prediction unit 370 is provided in addition to the configuration described in the first embodiment, the batch job execution unit 320 executes the same process as in FIG. 11 after step S504 in FIG.

  As described above, according to the third embodiment, even when the load level of the batch job is unknown at the stage where the user inputs the batch job using the client terminal 100, the load level prediction unit 370 prompts the user. Instead, predict load levels. For this reason, when the user transmits a batch job processing request, it is not always necessary to specify the load level of the batch job. Accordingly, it is possible to reduce a situation in which the user specifies an inaccurate load level for a batch job whose load level is unknown and the load distribution in the batch job processing apparatus 300 is disturbed.

<Embodiment 4>
In the first to third embodiments, each record held in the schedule table 340 and the parameter table 350 may be deleted after a predetermined period (for example, 30 days) has elapsed. At this time, only the batch job whose value in the “status” column 347 is “successful” may be deleted, or the batch job after the predetermined period may be deleted unconditionally.

<Embodiment 5>
FIG. 12 is a configuration diagram of a batch job processing system 1000 according to the fifth embodiment of the present invention. In the fifth embodiment, a batch job execution server 400 is newly provided separately from the batch job processing apparatus 300, and the processing entity that executes the batch job is cut out from the batch job execution unit 320. Since the other configuration is the same as that of any of Embodiments 1 to 4, the following description will focus on the differences.

  The batch job execution server 400 includes a batch job processing unit 410. The batch job processing unit 410 is a functional unit that handles the processing main body of the batch job executed by the batch job execution unit 320. In the fifth embodiment, the batch job execution unit 320 essentially serves as an activation unit that activates the batch job processing unit 410.

  The batch job processing unit 410 can be configured using hardware such as a circuit device that implements the function, or can be configured using an arithmetic device such as a CPU and software that defines the operation thereof.

  Note that some of the functional units included in the batch job processing apparatus 300 such as the load level prediction unit 370 may be arranged in the batch job execution server 400.

  As described above, according to the fifth embodiment, part of the functions of the batch job processing apparatus 300 is cut out to the batch job execution server 400, so that the processing burden on the batch job processing apparatus 300 can be reduced. .

<Embodiment 6>
In the first to fifth embodiments, the configuration including the three batch job execution units 321 to 323 has been described. In addition, each batch job execution unit executes batch jobs of load levels 1, 1-2, and 1-3, respectively. The configuration of the batch job execution unit 320 is not limited to this, and the configuration can be appropriately changed according to the nature and number of batch jobs to be executed.

  For example, when the method of the present invention is introduced in a system that executes many batch jobs with a high load level, a larger number of batch job execution units 323 may be provided. Also, the total number of batch job execution units 320 may be four or more, or two or less.

  The batch job execution unit 320 may execute only a single load level. For example, each batch job execution unit executes batch jobs of load levels 1, 2, and 3, respectively.

  By adjusting the number of batch job execution units 320 and the target load level, the processing performance and parallelism of the batch job processing system 1000 can be adjusted according to the specifications of the target job.

  DESCRIPTION OF SYMBOLS 100: Client terminal, 200: Web server, 210: Batch job reception part, 300: Batch job processing apparatus, 310: Batch job acquisition part, 320-323: Batch job execution part, 330: Database, 340: Schedule table, 350 : Parameter table, 360: job queue, 370: load level prediction unit, 400: batch job execution server, 410: batch job processing unit, 1000: batch job processing system.

Claims (8)

A batch job acquisition unit for acquiring a batch job and its execution schedule;
A batch job execution unit for acquiring the load level of the batch job and executing the batch job;
With
The batch job execution unit
A batch job processing apparatus that executes only the batch job of a predetermined load level and does not execute the next batch job until the execution of the batch job is completed. A storage unit for storing the execution schedule of the batch job;
The batch job acquisition unit acquires the execution schedule from the storage unit,
The batch job execution unit
The batch job processing apparatus according to claim 1, wherein the batch job is executed according to the execution schedule acquired by the batch job acquisition unit. A job queue that temporarily holds the batch job acquired by the batch job acquisition unit and outputs it according to an output request,
The batch job acquisition unit
Store the batch job acquired from the storage unit in the job queue,
The batch job execution unit
The batch job processing apparatus according to claim 2, wherein the batch job is acquired from the job queue and executed. The batch job processing apparatus according to claim 3, wherein the batch job stored in the job queue cannot be deleted from the job queue except when executed by the batch job execution unit. A load level prediction unit for predicting the load level of the batch job;
The storage unit
Storing the execution schedule of the batch job for which the load level is not specified;
The load level prediction unit
The batch according to any one of claims 2 to 4, wherein a load level of the batch job is predicted before the batch job execution unit executes the batch job for which the load level is not specified. Job processing unit. A load level prediction unit for predicting the load level of the batch job;
The storage unit
Storing the execution schedule of the batch job for which the load level is not specified;
The batch job acquisition unit
Storing the batch job in which the load level is not specified in the job queue;
The load level prediction unit
The load level of the batch job is predicted before the batch job execution unit executes the batch job stored in the job queue, the load level of which is not specified. The batch job processing apparatus according to claim 3 or 5. The batch job processing apparatus according to any one of claims 1 to 6,
A receiving server that receives the content of the batch job and an instruction of an execution schedule and outputs the received instruction to the batch job processing apparatus;
A batch job processing system comprising: A batch job processing apparatus according to any one of claims 2 to 6,
A receiving server for receiving the content of the batch job and an instruction of an execution schedule;
Have
The receiving server is
Store the contents and execution schedule of the batch job in the storage unit,
The batch job execution unit
The batch job processing system, wherein the batch job is executed according to the execution schedule.

Images