JPH03180930A - Automatic job starting system - Google Patents

Abstract

PURPOSE:To automatically starting plural jobs whose starting time is determined during the driving of a system by previously applying the starting time of plural jobs to a system with the automatically judgeable format. CONSTITUTION:When start processing starting reference time arrives, an extracting means 7 refers to a starting time registered data unit stored in a storage means 3 and extracts the starting time registered data units to be started at the start processing starting reference time. A starting time output means 9 rearranges the extracted starting time registered data in the starting time order and outputs the rearranged result. A starting means 11 decides whether the starting time deciding time data outputted from a deciding time output means 5 coincide with the starting time registered data outputted from the means 9 or not, and when the time indicated by the starting time registered data arrives, starts the job corresponding to the starting time registered data. Consequently, each job can be automatically started at the arrival of the starting time without using manual operation.

Description

[Detailed Description of the Invention] [Summary] The startup times of a plurality of running jobs whose startup times have been determined are registered in advance, the arrival of each startup time is automatically determined, and the arrival of the startup time is determined. Regarding the job automatic startup method that automatically starts jobs that have been started, the purpose is to automatically start multiple jobs whose startup time is fixed during operation. In a data processing system that requires starting a plurality of jobs, a storage means stores a start time registration data unit for each job, and a judgment time output outputs time data for judgment of start time by referring to the output of a clock. means, and an extraction means for referencing the start time registration data units in the storage means when the reference time for starting the start processing has arrived, and extracting the start time registration data units that are to be started at the reference time for starting the start processing. , startup time output means for rearranging and outputting the extracted startup time registration data in ascending time order; and startup time determination when the time data for startup time determination becomes the time data for startup time determination from the startup time output means. The present invention is configured by providing a starting means for starting a job corresponding to the time data.

[Industrial application field]

The present invention registers in advance the startup times of a plurality of running jobs whose startup times have been determined, automatically determines the arrival of each startup time, and automatically launches the job whose startup time has arrived. Regarding automatic job startup method.

Jobs executed by computer systems can be broadly classified into non-routine jobs and routine jobs.

Here, an atypical job is a job whose startup time is not constant, and a regular job is a job whose startup time is constant. Typical jobs include jobs that perform backup processing to temporarily save processing data, jobs that start and stop the system, and the like.

[Conventional technology]

Since the above-mentioned non-routine job is started at an arbitrary start time, it is normally started manually. or,
Conventionally, routine jobs have been started manually like non-routine jobs.

[Problem to be Solved by the Invention] However, there are a plurality of routine jobs installed in a computer system, and each of the plurality of routine jobs not only has a different start time, but also has a different startup time. They often have different cycles. If you try to start such multiple jobs manually as described above, it will impose a large human burden, and if you forget to start them, it will have an extremely large impact on data processing. , for example, inducing troublesome and labor-intensive work such as necessary data recovery processing. Simply saying that startup is done manually leads to problems like the ones mentioned above in system operation.

The present invention was created in view of such problems, and an object of the present invention is to provide an automatic job starting method that can automatically start a plurality of jobs whose starting times are determined during operation.

[Means for Solving the Problems] FIG. 1 shows a block diagram of the principle of the present invention. As shown in this figure, the present invention has a clock 1 and stores a start time registration data unit for each job in a data processing system that requires starting a plurality of jobs whose start times are fixed during operation. a determination time output means 5 that refers to the output of the clock 1 and outputs time data for determining the startup time; and a determination time output device 5 that outputs startup time determination time data by referring to the output of the clock 1; Extracting means 7 that refers to the units and extracts the startup time registration data unit that is the startup target at the startup processing start reference time, and rearranges the extracted startup time registration data in ascending time order and outputs it. It is configured by providing a start time output means 9 and a start means 11 for starting a job corresponding to the start time registration data when the start time determination time data becomes start time registration data from the start time output means 9. did.

[For production]

When the startup processing start reference time arrives, the extraction means 7 refers to the startup time registration data units in the storage means 3 and extracts the startup time registration data unit that is the startup target at the startup processing start reference time. do. The startup time output means 9 rearranges the extracted startup time registration data in ascending time order and outputs the rearranged data. The starting means 11 determines whether or not the starting time determination time data output by the determining time outputting means 5 is the starting time registration data from the starting time outputting means 9, and determines whether or not the starting time determination time data output by the determining time outputting means 5 is the starting time registration data. When the indicated time arrives, the job corresponding to the start time registration data is started.

Therefore, each job can be completed without any human intervention.
It will start automatically when the start time arrives.

〔Example〕

An embodiment of the present invention will be described with reference to FIGS. 2, 3, and 4. FIG. 2 shows a computer system implementing the present invention. In this figure, 8 is a timer, 10 is a central processing unit (CPU), 12 is a main memory, 14 is a disk device, and 16 is a display device. These devices 10
．． 12, 14, and 16 are connected via bus 18. The display device 16 is equipped with a keyboard 17. Figure 3 shows that when the computer system is operated,
An example of how the memory area of the main memory 12 is used is shown. 20 is an area used by O3, 22 is a temporary storage area within area 20, 24 is an area for registers R1, R2, . An area to store a program that executes the processing flow shown in
Reference numerals 26 and 27 are an activation target number counter and an execution counter provided in the area 28. 30 each indicates an area used by the user. The program stored in area 28 is stored in disk device 14, and is called into area 28 prior to performing the processing of the present invention. The same applies to programs used by users. Note that in connection with the present invention, data from the disk device 14 to the main memory 12
The unit of data transfer is, for example, a startup time record registered in a table in the disk device 14 as described later.

This startup time record has startup time data and the like set therein.

In FIGS. 2 and 3, the disk device 14 and main memory 12 correspond to the storage means in FIG.
The program executed by the PU101 main memory 12 and step 312 in FIG. 4 corresponds to the determination time output means 5 in FIG. Register a, register b register c, CPU
101 main memory 12, steps S1 to S1O in FIG.
The program that executes this corresponds to the extracting means 7 in FIG. 1, and the program that executes the register a, the main memory 12 of the CPU 101, and the step Sll of FIG. 4 corresponds to the startup time output means 9 of FIG. CPUIO3 main memory 12,
The program executed by the execution counter 27 and step Sll in FIG. 4 corresponds to the starting means 11 in FIG.

The processing of the present invention in the above system configuration will be described below. Note that registers a, b, and c, which will be described later, are register R.
1, R2, and R3.

Further, it is assumed that the job automatic startup program that executes the processing flow shown in FIG. 4 and each startup time record for each regular job are input into the disk device 14 in advance. The program is input by a loading means (not shown) provided in the computer system, and the startup time record is input after the computer system is activated or while the computer system is in operation before the computer system is activated. , keyboard 17 of display device 16
is operated and each startup time record for each routine job is entered manually. Each of these startup time records is registered in a table format in the startup time record registration area in the disk device 14 (an example of the registration is shown in FIG. 5). In FIG. 5, the first column shows the startup determination conditions, the second column shows the startup time, and the second and subsequent columns show the startup target jobs. Day of the week codes O through 6 set in the first column correspond to Sunday through Saturday, respectively. The day of the week code 7 represents every day.

Loading a job autostart program like this,
After the startup time record is registered, the job automatic startup program is read into the main memory 12 from the disk device 14 and the processing starts at the appropriate time for program processing while the computer system is operating, for example, when the computer system starts operating every week. Then, first read the current time (for example, 093000) from the timer 8 and set it in register a (see 31 in Figure 4), and then read the current time (for example, 8909).
19) is read (see S2 in FIG. 4) and set in register b. This register b data (0) is used to code the day of the week (in the above step, it is Tuesday).
(see S3 in FIG. 4) and set 2 in register C as the day of the week code. In this way, the activation determination condition, which will be described later, is set as Tuesday, September 19th, 9:30900 seconds, as shown in FIG.

Next, each startup time record is read out one by one from the startup time record registration area of the disk device 14 (see 84 in FIG. 4), and it is determined whether the startup target conditions are satisfied ( (See 85 in Figure 4). Here, the activation target conditions include, for example, whether the activation time record is in a regular description format and whether the data set in the first column is a comment definition. This will be explained with reference to the startup time record registration example shown in FIG. 5 as follows.

Transfer of the startup time records from the startup time record registration area in the disk device 14 to the main memory 12 is performed sequentially starting from the first startup time record. Among the startup time records that are sequentially read into the main memory 12, If the activation time record is 1 to 2, the above-mentioned activation target conditions are not satisfied. All of them satisfy the regular description format, but in each startup time record, the data in the first column of the first startup time record ■ is a comment definition (matters that are not subject to startup), and the data in the first column of the first startup time record ■ The startup definition date of the column is Monday, and the first column of the startup time record ■ is
This is because the activation definition for the 20th is placed every day. Then, the determination in step S5 is N for the startup time records to () that do not satisfy the startup target conditions described above.
It is determined whether the read startup time record is the last startup time record (see 36 in FIG. 4).

This final determination of the record is based on the file end mark EOF, which is used when the file is normally filed in the disk device 14.
Use. If the last startup time record has not yet been read (see N in S6 in FIG. 4), reading of the next startup time record begins; however, if it is the last startup time record, the process of step Sll is performed. Proceed to.

On the other hand, activation time records (1) to (2) satisfy the above-mentioned activation target conditions. This means that each startup time record ■ satisfies the regular description format, the first column of startup time records ■ and ■ has a startup definition date of every day, and the startup time record ■
This is because the start definition date in the first column is Tuesday. Therefore, the process proceeds to the process described below.

As mentioned above, since the activation time records (1) to (2) are the activation targets (see 35 Y in FIG. 4), each time each of these activation time records is read, the activation target number counter 26 in the area 28 is set to 1. (see S7 in FIG. 4) and writes the read startup time record into the main memory 12 (see S8 in FIG. 4). The startup time record written in the main memory 12 is shown in FIG. Then, the next startup time record reading process begins (see 39 in Figure 4), but it is determined whether or not the target startup time record read at that time is the last startup time record. If this is not the case (see SIO N in Figure 4), the next startup time is determined and the 3 code reading process begins (see 34 in Figure 4).The above-mentioned startup time record reading process is repeated. Finally, when all the startup time records registered in the disk device 14 have been read out (see S6 and SIO Y in FIG. 4), the startup time record that is the target of startup is changed to the time A process of generating a startup time record string in ascending order for each time and temporarily storing it in a storage area in the main memory 12 begins (see 311 in FIG. 4). The startup time record written in the main memory 2 is shown in FIG.

Upon completion of this process, the job starting process corresponding to the start time data to be started starts.

The timer 8 changes the current time to a predetermined time (for example,
1 minute) and sets it in register a (see 312 in FIG. 4). Each time, the sequence of startup time records temporarily stored in main memory 2 in ascending order is read out (see 313 in Figure 4), and each startup time record is combined with the current time set in register 4 data a. For comparison, 314 in Figure 4
(see 315 in FIG. 4), if they do not match (see 314 N in FIG. 4), a waiting process is entered (see 315 in FIG. 4). This waiting process is a process of requesting the supervisor of the computer system to start a job for the comparison target time when the comparison target time in the register a has arrived.

Also, if a match is obtained in the activation from the supervisor or in the comparison (see Y at 314 in FIG. 4)
, starts a job for the comparison target time of the register a (see 316 in FIG. 4). For example, if the current time read from the timer 8 is 093100, the above-mentioned waiting process is entered, and the timer 8 starts at the startup time indicated by the startup time record (3) shown in FIG. 7 at 10:00:00. , the job specified by the supervisor's activation time record (3) is activated.

Each time such a job is started, 1 is added to the value of the execution counter 27 in the area 28 (see 316 in FIG. 4). The value of the execution counter 27 and the number of activation targets counter 26
If they do not match, the current time reading process begins (see 318N in FIG. 4). If they match (see Y at 318 in FIG. 4), the activation process for that day ends (see 319 in FIG. 4).

In addition, in the above embodiment, an example was explained in which the activation target condition is related to the regular description format and whether the data set in the first column is a comment definition, but if the number of activation time records is If there are many
For example, you may set conditions to narrow down the list to only those that should be activated every day.

Further, the job may be started by an interrupt from a job automatic start program instead of the above-mentioned start request to the supervisor.

The present invention can also be implemented in a system environment in which multiple processors operate in parallel.

■ [Effects of the invention] As is clear from the above, according to the present invention,
Since the start times of a plurality of jobs are given to the system in advance in a format that can be automatically determined by the system, it is possible to automatically start a plurality of jobs whose start times are fixed during operation. Therefore, when there are a large number of jobs, the hassle of manual work is eliminated, and the prevention of forgetting to start them is also far removed. There is no longer any problem with system operation caused by forgetting to start the system.

[Brief explanation of drawings]

Fig. 1 is a block diagram of the principle of the present invention, Fig. 2 is a diagram showing a computer system implementing the present invention, Fig. 3 is a diagram showing how the main memory is used, and Fig. 4 is a diagram showing the processing flow of the present invention. Figure 5 is a diagram showing an example of registration of startup time records in the table, Figure 6 is a diagram showing only the startup time records of the startup targets extracted from the startup time records in the table, and Figure 7. 2 is a diagram showing each of the boot time records obtained by rearranging the boot time records of the extracted boot targets in ascending order of time. FIG. 1 to 3, 1 is a clock, 3 is a storage means (disk device 14, main memory 12), and 5 is a judgment time output means (register a, CPU101 main memory 12, step S12 in FIG. 4). 7 is an extraction means (register a, register b, register c,
CPU101 main memory 12, a program that executes steps S1 to SIO in FIG. 4), 9 a startup time output means (register a, CPU101 main memory 12, a program that executes step SLL in FIG. 4), 11 a startup device (CPUIO 1 main memory 12, real 8 row counter 27, program executed in step Sl I in FIG. 4). Principle of infallible θH Fig. 1, 77 Fig. 1

Claims (1)

[Claims] (1) In a data processing system that has a clock (1) and requires the startup of a plurality of jobs whose startup times are fixed during operation, a storage means (3) that stores startup time registration data units for each job. and judgment time output means (5) for outputting time data for judgment of starting time by referring to the output of the clock (1), and when the start processing start reference time arrives, the starting time of the storage means (3) is determined. an extracting means (7) that refers to the registered data units and extracts a startup time registration data unit that is a startup target at the startup processing start reference time, and arranges the extracted startup time registration data in ascending time order; a start time output means (9) for outputting the start time data instead of the start time data, and a start means for starting a job corresponding to the start time data when the start time determination time data becomes the start time data from the start time output means (9). (11) A job automatic starting method characterized by providing the following.