JPH03205649A - Automatic preservation system for multifile information medium - Google Patents

Abstract

PURPOSE:To improve the throughput by controlling information mediums so that the medium which will be probably used in future is resident on a drive device for a certain time based on forecast results of times, frequencies, or the like of future requests. CONSTITUTION:This system consists of a host computer 1 which requests input/ output to an information medium automatic preservation depository, a memory 2 where various data are stored, an automatic preservation depository controller 3, recording/reproducing device controllers 4-1 and 4-2, recording/reproducing devices 5-1 to 5-6, medium handling devices 6-1 and 6-2, and a device or medium preservation rack 7. Future processing requests to mediums are forecasted based on data related to the past medium use history or data related to the job execution scheduling of the computer 1. Mediums are controlled based on forecasted results so that the medium which will be probably requested is resident on recording/reproducing devices 5-1 to 5-6 for a certain time. Thus, the response performance and the throughput are improved.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to an operation control method for an automatic storage system for large-capacity information media such as digital VT and optical disks, and reduces the man-hours for mounting and demounting the media. , Concerning a control method for realizing a system with high response performance and throughput.

[Conventional technology]

Online automation of operations is progressing in various fields such as finance, securities, distribution, and manufacturing. As a result, the amount of stored data, including online data, is rapidly increasing. Therefore, we mainly target cartridge MTs with good operability, and are equipped with storage shelves for information media, multiple recording/playback devices, and multiple handling devices.
Information media automatic storage systems that automatically perform mounting or demounting operations are attracting attention.

Cartridge MT has a storage capacity of about 200MB,
Usually, one file is recorded on one medium.

For this reason, all media are stored in storage racks, and when a processing request for a file occurs, the media containing the file is mounted on the recording/playback device using a bundling device, and when the processing is completed, the media is demounted and stored. The company operates by returning the items to the shelves. Therefore, in an information medium automatic storage system for such a cartridge MT, the purpose of control is to shorten the time associated with mounting and demounting the medium in order to achieve high response performance and throughput.

[Problem to be solved by the invention]

By the way, the storage density of information media itself is rapidly improving, and digital VTRs, optical discs, etc. have a storage capacity of several GB to several tens of GB (storage capacity tens to hundreds of times that of cartridge MT). High-capacity information media with high capacity are already in practical use. Since the storage capacity of such large-capacity media is extremely large compared to the file capacity, usually several tens to hundreds of files are recorded on one medium. Therefore, there is a high possibility that multiple computer jobs will use the same medium. For these reasons, if an operation similar to a conventional information medium automatic storage system for cartridge MT is adopted, even if a medium that has been processed by a certain computer job is demounted, another computer job that occurs next will be This is likely to result in inefficient operation, as the same medium must be mounted again in response to a job processing request.

The present invention aims to reduce media mounting time and realize a system with high responsiveness and throughput.
For each medium, the future request time, frequency, etc. are predicted, and based on the results, the medium is
Control it so that it exists on the playback device. [Means for solving the problems] In order to solve the above problems, the following measures are taken.

(1) Means for determining a medium to be mounted on the recording/reproducing apparatus Predicts a medium that is likely to be required in the future, and controls the medium so that it resides on the recording/reproducing apparatus for a certain period of time.

(2) Job execution scheduling means to reduce the number of man-hours required to mount/demount the medium that occurs when the medium containing the requested file is not on the recording/playback device.
Schedule the job execution order so that it is lower.

(3) Means for determining file placement on media Media is arranged so that data that is likely to be used in the future is concentrated on a small number of media, and requests from multiple jobs can be processed in extreme parallelism. Determine file placement in . (4) Temporary storage area management means If the medium requested when writing data is not mounted on the recording/playback device, a storage means prepared in advance,
Alternatively, use another medium that has already been mounted as a temporary storage area, and then write normally to the requested medium.

Further, if the medium requested at the time of data reading is not mounted on the recording/reproducing device, and there is data to be read in the temporary storage area, the data is read from the temporary storage area rather than from the medium.

[Effect]

In order to achieve the above-mentioned purpose, it is necessary to predict data that is likely to be used in the future based on past processing results, job execution schedules, etc., and centrally place that data in a small number of media. , the media are controlled to reside on the drive device for a certain period of time. Furthermore, it performs optimal job execution scheduling that minimizes the number of man-hours required for transporting media. As a concrete plan to achieve this, instead of unconditionally demanuminating the media that has been processed as in the past,
One possible method is to leave it on the recording/reproduction snowboard as needed. Here, if we need to demount some of the media on the loader in order to mount new media, we will , demount media in order of decreasing likelihood of future use. In addition, when writing data, if the requested medium is not mounted on the recording/playback device, a storage device prepared in advance or another medium that is already mounted is used as a temporary storage area, and then , performs normal writing to the requested medium. The above method will be described in detail as follows.

(1) Determining the media to be installed in the recording/playback device Based on data related to past media usage history or data related to execution scheduling of computer jobs, it is possible to predict which media will be requested for processing in the future. Based on the results, the media that are likely to be requested are controlled to reside on the recording/playback device for a certain period of time. (2) Job Execution Scheduling Even if media are mounted in advance as in (1), there may be many media in the group of media requested by the job that do not exist on the recording/playback device. In such a case, the job execution order is optimally scheduled in order to minimize the man-hours for mounting and demounting operations.

(3) Determining file placement on media In order to keep a specific group resident on the recording/playback device, data that is likely to be used in the future is concentratedly placed on a small number of media. In addition, even if the man-hours for mounting and demounting work are reduced through (1) and (2) above, if jobs that occur at the same time compete for the same medium, the waiting time will reduce response performance and throughput. decreases.

Therefore, files are placed on each medium so that each job can be processed in parallel as much as possible.

(4) When writing data to the temporary storage area, if the requested medium is not mounted on the recording/playback device, the memory prepared in advance,
A storage device such as a disk, a temporary storage-only medium resident on a recording/playback device, or a free space on another medium that is already mounted is used as a temporary storage area, and then the requested medium is Export normally.

(5) Reading data from the temporary storage area When reading data, if the requested medium is not mounted on the recording/playback device, if there is data to be read in the temporary storage area, it will not be read from the medium but from the temporary storage. Load from area.

〔Example〕

Embodiments of the present invention will be described with reference to FIGS. 1 to 18.

Figure 1 shows the overall configuration of a multi-file type automatic storage system for information media based on the present invention. (a) shows the hardware configuration; (b) and (Q) show the software configuration; however, (b) shows the software configuration when the host computer is assigned the functions for realizing the present invention; (c) shows the software configuration when the software is assigned to the automatic storage controller. In FIG. 1 (a), device 1 is a host computer that makes input/output requests to the automatic storage for information media, and device 2 is a host computer that makes input/output requests to the automatic storage controller. A memory for storing various data, device 3 is an automatic storage controller, device 4-1.4-2 is a recording/playback device controller, devices 5-1 to 5-6 are recording/playback devices, device 6-1 .6-2 is a media handling device, and device 7 is a media storage shelf. In the division of functions shown in FIG. 1(b), the host computer l runs the operating system 10.
l, job execution scheduling unit 102, file management unit 103, medium mount/demount management unit 104,
It consists of a medium file arrangement determining section 105 and a residual medium determining section 106 for the recording/reproducing device. The automatic storage controller 3 only manages the operations of various devices that are the maintenance elements of the automatic storage. On the other hand, in the division of functions shown in FIG. 1(c), the host computer l has only the functions of accepting various jobs and issuing input/output requests to the automatic storage.

The automatic storage controller 3 includes a job execution scheduling section 301, a file management section 302, a medium mount/demount management section 303, a medium file arrangement determination section 304, and a residual medium determination section 30 for the recording/playback device.
5. Consists of a device operation management section 306.

FIG. 2 shows a schematic processing procedure when reading/writing data using the information medium automatic storage system of the present invention assuming the software configuration shown in FIG. 1(b). The host computer performs the "processing". Figure 1 (c
), the automatic media warehouse controller may be in charge of "other processing." The procedure will be explained according to the diagram.

201 (Job reception): The operating system receives a job execution request from a user and registers it in the execution request reception table.

202 (Job schedule instruction): The operating system determines that all the media required for job execution are not in use, and that the recording/playback device for mounting those media is not in progress.
Check. Only jobs that satisfy these two conditions are targeted for the next scheduling. Then, it instructs the job execution scheduling unit to schedule the job so that the number of steps for mounting/demounting the medium is minimized.

203 (Estimation of mount/demount man-hours): The job execution scheduling unit creates alternative plans for the job execution order and estimates the mount/demount man-hours for each plan.

204 (Job execution order determination): Select the execution order that minimizes the above man-hours, and notify the operating system. 205 (Job execution preparation instruction): The operating system determines the next job to be executed based on the above notified information, and instructs the media mount/demount management unit to prepare the media to be used during execution. Instruct. 206 (acquisition of media usage rights): The media mount/demount management unit examines the files required during job execution and acquires the right to exclusively use the media on which those files are stored for other jobs. .

207 (Medium location check): Check whether the medium used when executing the job is mounted on the recording/playback device or in the storage area.

208: If all the media to be used are mounted on the recording/playback device, proceed to step 215. If not, proceed to step 209 to mount the media still in the storage shelf.

209: If there is an empty recording/playback device on which no media is mounted, the process proceeds to step 213 for mounting the necessary media on that device. If not, proceed to step 210 to demount any already mounted media and empty the device. 210 (Remaining medium determination): For each medium that has already been mounted, predict the possibility that it will be required in the future, and let the medium with the highest value remain in the recording/playback device, and set the medium with the lowest value as the target for demounting.

211 (Demount instruction): Instructs the controller of the information medium automatic storage to execute the demount operation of the medium determined to be demounted above. 212 (demount execution): The controller executes the specified demount operation. When the work is completed, control is returned to the media mount/demount management section.

213 (Mount instruction): Instructs the controller to perform a mounting operation in order to mount a necessary medium in the recording/reproducing device that has become empty due to the above operation.

214 (mount execution): The controller executes the specified mounting operation.
When the work is complete, control is returned to the media mount/demount management section. Now that the job execution preparations are complete, the media mount/demount management section returns control to the operating system.

215 (Job Execution): The operating system executes the job that was in the state of waiting for media preparation. When it is necessary to read/write data to/from the medium during execution, the controller is instructed to read/write.

216 (Data R/W Execution): The controller operates the recording/reproducing device to execute the instructed read/write operation.

217 (Job End): The operating system performs regular job post-execution processing when the job ends. Then, mount the media/
Instructs the demount management section to perform post-processing on the used media. 218 (Release of media usage rights): The media mount/demount management unit releases all usage rights for the used media. However, once the media has been used,
Mount it on the recording/playback device without demounting it. Note that in step 205, if a specific recording/playback device is specified as the recording/playback device that should process the media necessary for job execution, the specification source will receive the following information:
The specified device is assumed to be used virtually, and the selected device is actually used in the following manner.

(1) If the medium to be used is mounted on a recording/playback device: Use that device.

(2) The medium to be used is not mounted on the device,
If there is a vacant device: Use that vacant device. (3) When the medium to be used is not mounted on the device and there is no free device: The device determined to be demounted in step 210 is used.

The following three methods, which are features of the present invention, will be explained below. (1) Method of determining residual medium in recording/playback device.

(2) Job execution scheduling method.

(3) How to determine file placement on the medium.

Method of determining residual media in recording/playback equipment In order to improve the response performance of an automatic information media storage system, first of all,
Predict which media processing requests are likely to occur in the future,
Based on this, it is necessary to ensure that media that are likely to receive requests remain on the recording/playback device. In this embodiment, this prediction is performed using the following two methods.

(1) A prediction method based on the last time of use of the medium.

Generally, data stored on media has the characteristic that the newer the data, the more frequently it is used, and the older the data is, the more rarely it is used. Therefore, for each medium mounted on the recording/playback device, record the time when it was last used. And when deciding which medium to demount,
Select in order of oldest time.

(2) A prediction method based on the frequency of use of the medium.

Instead of using the characteristics of the data in the media as described above, record the usage frequency for each media directly, and when deciding which media to demount, use the most frequently used data. Select from lowest to lowest.

In order to realize the above two methods, as shown in Figure 3, it is necessary to record usage data such as the last usage time and frequency of use for each medium mounted on the recording/playback device. It is necessary. Below, a specific procedure for determining the residual medium will be explained based on these usage record data with reference to FIG. Note that the procedure described here is from step 206 to step 213 in the schematic procedure of FIG.
This is a detailed version of

401 (Reception of file name to be used): Receives the name of the job to be executed and the file name to be used from the operating system.

402 (Acquisition of media use rights): Investigate the media storing used files and acquire the right to use those media exclusively for other jobs. 403 (Medium position check): Check whether each medium to be used is mounted on the recording/playback device or is still in the storage cabinet.

404: If some of the media to be used have not been mounted yet, proceed to step 405. If all media are mounted, the process ends.

405: If there is an empty device in which no medium has been mounted yet in the recording/playback device, the process proceeds to step 409 for mounting the medium in that device. If there is no empty device, the process advances to step 406.

406 (Determine residual media): Refer to the device-mounted media management table shown in Figure 3, check the last use time of each media mounted on the recording/playback device, and select the media that will not be used in this job and , select the medium with the oldest time among them. 407 (Demount instruction): Instructs the controller to execute the demount operation of the selected medium. 408: Wait for the demount operation to finish. 409 (Mount instruction): Select one medium that is to be used in this job but has not yet been mounted, and instruct the controller to mount that medium.

410: Wait for the mounting process to finish.

41l (updating the device-mounted medium management table) 2. Along with the above-described demounting and mounting operations, information on the newly mounted medium is registered in the device-mounted medium management table shown in FIG. Then, the process advances to step 404. Next, a specific procedure for registering the last use time and use frequency in the device-mounted media management table will be explained with reference to FIG. Note that this procedure is a detailed version of steps 217 to 218 in the general failure order in FIG.

501 (Reception of completed job information): Receives the name of the completed job, the name of the files used, and the name of the medium in which these files are stored from the operating system.

502 (Registration of last use time): Search for the medium name accepted above in the device-mounted medium management table, and update the value of the last use time of the record related to the medium to the current time.

503 (registration of usage frequency): Increase the usage frequency value of the same record by 1.

504 (Release of media usage rights): Release rights to all media used.

Job Execution Scheduling Method> Now, even if the remaining media are determined by the method described above, there may be many media among the media requested by the job that do not exist on the recording/playback device. In such cases, the job execution order is optimally scheduled in order to minimize the number of steps required for mounting/demounting. However, the scheduling targets are jobs that satisfy the following two conditions from among the jobs waiting for execution that are managed by the operating system.

A job with the highest priority, or a job whose execution order can be reversed.

The total number of jobs to be scheduled falls below a predetermined upper limit.

In this embodiment, scheduling is performed using the following two methods.

(1) Method of prioritizing the job with the least number of man-hours For each job to be scheduled, estimate the number of mount/demount operations that will occur as the job is executed. Then, the job with the lowest value is executed with the highest priority. (2) Method for determining the order that minimizes the total number of work steps.

Permutation of the execution order of n jobs to be scheduled n! Identify all the items and estimate the total number of mounting and demounting operations for each item. Then, jobs are executed in the execution order that minimizes the value.

In order to realize the above two methods, as shown in Figure 6, for each job to be scheduled,
It is necessary to record the file name used by the job, its storage medium, and the medium location.

First, according to FIG. 7, a scheduling procedure using a method that gives priority to the job with the least number of man-hours will be explained. Note that the procedure described here is step 2 in the outline procedure in Figure 2.
This is a detailed version of steps 03 to 204.

701 (Initial registration of job to be scheduled): The job with the highest priority is selected from the group of jobs waiting for execution managed by the operating system and registered in the job to be scheduled table shown in FIG. Let the number of registered jobs n be 1. Or a job whose execution order can be reversed. 702: Among the jobs waiting to be executed, the job with the next highest priority (
(Priority order n + 1st job) is selected. Then, it is checked whether the execution order of that job can be reversed with the jobs already registered in the scheduled job table. If it is OK to reverse the process, proceed to step 703; otherwise, finish job registration and proceed to step 705.
Proceed to.

7o3: If the total number of jobs already registered in the scheduling target job table is less than or equal to the predetermined value 11l}N, the process proceeds to step 704; otherwise, job registration is ended and the process proceeds to step 705.

704 (Additional registration of scheduling target job): Add the n-th priority job waiting for execution to the scheduling target job table. Then, the number n of registered jobs is increased by 1. Then, the process returns to step 702. 705 (Calculation of number of mount/demount operations for each job): For each of the n jobs registered in the table, calculate the number L of mount/demount operations associated with job execution. The calculation formula is as follows. L = (Total number of demounts of unused media) + (Total number of mounts of used media) = (Number R of media in the storage column among used media) x 2 706 (Determination of execution job): Mount the above / Select the job with the minimum value of the number of demount operations L, and notify the operating system of the job.

Next, according to FIG. 8, a scheduling procedure using a method for determining the order that minimizes the total number of work steps will be explained. Step 8
Steps 01 to 804 are the registration process for a job to be scheduled, and steps 701 to 70 described above
The procedure is exactly the same as up to 4. Therefore, steps 805 and subsequent steps will be explained. 805 (Extraction of permutation of job execution order): Extracting permutation n of the execution order of n jobs registered in the scheduling target job table. Create a piece and register it in the worktable. For example, the sixth
As shown in FIG. 9, the execution order of the 34m jobs Jl, J2, and J3 registered in the scheduled job table shown in the figure is 3! =6 pieces can be created.

806 (calculation of number of mounts/demounts): For each permutation registered in the work table, calculate the total number of mounts/demounts T when the job is executed in that order. This number of times is calculated by sequentially checking which media are mounted on the recording/reproducing device and which media are mounted as the job is executed.

For example, the jobs to be scheduled are Jl, J
Consider a case where there are 3 recording/reproducing devices, 2 and J3, and media A, B, C, D, and E are mounted on each device in the initial state. In order to execute jobs in the order of J1→J2→J3 as shown in the first column of FIG. 9, the medium on the recording/playback port is replaced as shown in the figure.
However, it is assumed that the media will be demounted in the order of the oldest last use time. At this time, a total of 10 operations are required: 5 mounting operations and 5 demounting operations. 807 (Determine job execution order): Select the execution order that minimizes the above-mentioned total number of mounts/demounts T, and notify the operating system of the execution order.

Method for determining file placement on storage media> By the way, even if the number of steps for mounting and demounting work is reduced by optimally determining residual media and job execution scheduling, if jobs that occur at the same time compete for the same media, The response performance deteriorates due to the waiting time. Therefore, in order to allow each job to be processed in parallel as much as possible,
Determine the placement of files on media. Specifically, we monitor whether there is a medium that is likely to receive conflicting requests from multiple jobs, and if there is such a medium, we will monitor whether some or all of the files stored on the medium are rearrange it on another medium.

In this example, if there is a medium that is likely to conflict, a method will be described in which a part of the files stored there are transferred to a new medium ill that does not store anything. In order to realize this method, as shown in Figure 10, a medium that indicates how many times usage requests from multiple jobs have conflicted in the past for all media stored in the information media automatic storage system is required. A management table is required. Further, detailed information on the cause of the conflict between the media, ie, as shown in FIG. 11, information on which file and how many times requests to use which file have occurred simultaneously is also required.

First, the procedure for recording the history in the medium management table of FIG. 10 and the conflict file recording table of FIG. 11 when there is a conflict of requests to use a medium will be explained with reference to FIG. Note that this procedure is similar to step 206 in the schematic procedure in FIG. 2, or step 401 and step 4 in FIG.
This is a more detailed explanation of 02. 1201 (Reception of file name to be used): The name of the job to be executed and the file name to be used are input from the operating system, and the file name is registered in the work table.

1202: Select the first file name from among the files registered in the table above.

1203: Check whether the right to use the medium storing the file selected above has been acquired by another job. If usage rights have been acquired, proceed to step 1204 to record historical media conflict data. If it has not been acquired, the process advances to step 1208 to perform similar processing on another file registered in the work table.

1204 (update of medium management table): In the medium management table shown in FIG. 10, the value of the number of conflicts of the medium in which a conflict with another job has occurred is increased by 1.

1205 (Investigation of file name used by competing job) Check which file stored on the competing medium is being used by another competing job.

1206 (Conflict file record table update): 11th
In the conflicting file record table shown in the figure, the value at the intersection of the file name of interest and the file name used by the other job is increased by 1.

1207 (Waiting for Competitor Job to End): Waits until the Competitor Job finishes and the right to use the required medium is released. Upon receiving the notification that the other party's job has ended, the process returns to step 1201.

1208: When the right to use the medium on which the file of interest is stored has not been acquired by another job. Check to see if there are any other files in the worktable. If there is, proceed to step 1209; if not, proceed to step 1210, 1209: Select the next file name in the work table and proceed to step 12
Return to 03.

1210 (Acquisition of media usage rights): Acquires the usage rights of the media in which all files in the work table are stored. Next, we will use the media management table in Figure 10 and the conflict file record table in Figure 11 to find out which media are likely to have conflicting usage requests, and to remove some of the files from other files in order to resolve the conflict. The procedure for rearranging the data on the medium will be explained with reference to FIG.

1301 (Detection of conflicting medium): Refers to the medium management table and finds a medium for which the number of occurrences of usage request conflict is greater than a predetermined upper limit. For example, if the upper limit is set to 20, then in FIG. 10 it is determined that medium A is a medium that is likely to compete.

1302: Check whether there is a medium that satisfies the above conditions. If there is, proceed to step 1303; otherwise, the process ends. 1303 (Detection of conflicting file pair): Select one medium that satisfies the conditions of step 130l, and refer to the conflict file record table regarding that medium. Then, check whether there is a file pair for which the number of conflicts is greater than the limit value. For example, if the upper limit is 5, then in Fig. 11, f
It is determined that three file pairs, l and f2, f2 and f3, and f3 and f4, are likely to conflict. 1304: Check whether there is a file pair that satisfies the above conditions. If there is, proceed to step 1305; if not, proceed to step 13
The process returns to 01 and processes another competing medium.

1305 (Registration of conflicting files): Registers the names of files that are to be avoided in the pair of files that are likely to conflict, detected in step 1303, in the work table. Then, let F be the total number of registered files. For example, in the example shown in Figure 11. Register fl, f2, f3, f4 in the work table. The number of registered files F is 4. 1306 (Creation of alternative file migration plan): Create an alternative plan for moving part of the files registered in the work table to one new medium within the range of F/2.

In the example of FIG. 11, the following three types of alternatives can be created, and the total number of alternatives is 11.

Plan not to move a file... 1 Plan to move one file... 4 Plan to move 2 files... 6 Figure 14 shows the specific contents of each alternative plan, 1307
(Evaluation of alternatives): Calculate how much the number of conflicts will be reduced for each alternative. The calculation method is as follows. As a result of file movement,
When a conflicting file pair is distributed over two media, the number of conflicts for that file pair is O.

If a conflicting file pair coexists on one medium as a result of file movement, the number of conflicts for that file pair remains the same. Section 1311 shows evaluation examples of each alternative. Next, we will explain how to use another medium on the device or another storage means as a temporary storage area when the requested medium is not on the drive device when writing data. First, Figure 15 shows the procedure when data requested to be written from a host computer is temporarily stored in another medium already mounted on the recording/playback device instead of being stored in the specified medium. The following is explained below. Here, Figure 16 is a temporary storage area management table for showing which files are stored in media for temporary storage (hereinafter referred to as temporary storage area), and Figure 17 shows which files are stored in each medium. This is a media management table that shows whether files are stored. The same applies when using storage means such as memory or disks as a temporary storage area. 1501
(Reception of file name to be used): Receives the name of the job to be executed and the file name to be used from the operating system.

1502 (Acquisition of media usage rights): Investigate the media that store used files and acquire the right to use those media exclusively for other jobs. 1503 (Medium position check): Check whether each medium to be used is mounted on the recording/playback device or is still in the storage rack.

1504: If there is a file to be used whose storage medium has not yet been mounted, select one file from among them and proceed to step 1505. If not, the process ends.

P5 0 5: Check whether the selected file is used for data reading or writing. In the case of reading, the process proceeds to step 1506, and in the case of writing, the process proceeds to step 1510.

1506: Check whether the file is stored in the temporary storage area by referring to the media management table shown in FIG. If it is stored, proceed to step 1507; otherwise, proceed to step 1512. 1507: Obtain the right to use the temporary storage area. 1508: The read source of the file is set to a predetermined address in the temporary storage area instead of its storage medium.

This address is also obtained by referring to the media management table. Then, the process advances to step 1504. 1509: Check whether there is free space in the temporary storage area to write the requested data by referring to the temporary storage area management table shown in FIG. If there is, step 151
If not, proceed to step 1512.

1510: Obtain the right to use the temporary storage area.

1511: Set the write destination of the file to the address of the free space in the temporary storage area, not the storage medium. This address is also determined by referring to the temporary storage area management table. Then, the process advances to step 1504. Step 1512 to Step 15
18 is step 405 in FIG. 4, as shown below.
This is the same as step 411. 1512: If there is an empty device in which no medium has been mounted yet in the recording/playback device, proceed to step 15l6 for mounting the medium in that device. If there is no empty device, proceed to step 1513.

1513 (Determining residual media): Refer to the device mounted media management table shown in Figure 3, check the last use time of each media mounted on the recording/playback device, and select the media that will not be used in this job and , select the medium with the oldest time.

1514 (Demount instruction): Instructs the controller to execute the demount operation of the selected medium.

1515. Wait for the demounting process to finish.

1516 (Mount instruction): Selects one medium that is to be used in this job but has not yet been mounted, and instructs the controller to mount that medium. 1517: Wait until the mounting operation is completed.

1518 (Update of device-mounted media management table): Along with the above-mentioned demounting and mounting operations, information on the newly mounted medium is registered in the device-mounted management table of FIG. 3. Then, the process advances to step 1504.

After making all the files necessary for job execution accessible using the steps above, execute the job. When the job is completed, the data regarding the storage file name and its location in the media management table is updated. It also updates the data regarding the storage file in the temporary storage area management table.

Furthermore, the procedure for writing a file stored in the temporary storage area to a regular medium will be explained with reference to FIG.

Here, we will explain the case where regular writing is performed according to the user's specifications, but the same applies to cases where it is performed based on the results of monitoring the media request frequency or job scheduling data. 1801: Upon receiving the user's specification, the temporary storage area management table is referred to to check whether there is a file in the temporary storage area that has not been properly written. If there is, proceed to step 1802, and if not, the process ends.

1802: Select one file that has not been exported yet.

1803: Check whether or not a medium for storing the selected file exists on the recording/playback device by referring to the medium management table mounted on the recording/playback device. If there is, proceed to step 1804; otherwise, proceed to step 1805. 1804: Wait until the recording/playback device becomes ready. Then, proceed to step l807. 1805: Instructs to mount the medium. If there is no vacant recording/playback device, select from 1512 to 15 in Figure 15.
After creating a free device using the same procedure as step 15, instruct it to mount.

1806: Wait for the mount to finish.

1807: Instructs to write the contents of the file in the temporary storage area to a predetermined location on the mounted medium.

1808: Wait for the export to finish.

1809: Sets a regular write completion flag in the column of the file in the temporary storage area management table. Then, the process returns to step 1801.

〔Effect of the invention〕

According to the present invention, it is possible to minimize the number of man-hours required for mounting/demounting media in an information media automatic storage system. It is also possible to reduce competition for the same medium between multiple computer jobs.

With the above, a system with high responsiveness and throughput can be realized.

[Brief explanation of drawings]

Fig. 1 is an overall configuration diagram of the present invention, Fig. 2 is a schematic processing procedure of the system accompanying job execution, Fig. 3 is an example of a device-installed media management table, and Fig. 4 is a diagram showing how to manage residual media in the recording/playback device. Determination procedure, Figure 5 is the procedure for registering usage record data in the device-mounted media management table, Figure 6 is an example of the job table to be scheduled, Figures 7 and 8 are the job execution scheduling procedure, and Figure 9 is the mounting procedure. /Explanatory diagram of how to calculate the number of demounts, Figure 10 is an example of a media management table,
Figure 1l is an example of a conflict file recording table, Figure 12 is a procedure for recording history data of media conflicts, Figure 13 is a procedure for detecting media that is likely to cause conflicts and resolving conflicts, and Figure 14 is an alternative plan for moving files. For example, Fig. 15 shows the procedure for writing data using the temporary storage area, Fig. 16 shows an example of the temporary storage area management table, Fig. 17 shows an example of the media management table, and Fig. 18 shows how to write data from the temporary storage area to the regular medium. data book 8 3 (2) 5th figure me6 mek3 figure 6 9 figure/ρ (2) ka12 figure 13 mouth 14 ku6/l (2) chrysanthemum 17 prisoner

Claims (1)

[Claims] 1. For information media such as digital video tapes and optical discs, a shelf for storing a plurality of media, a plurality of recording/playback devices, and a storage space between the shelf and the recording/playback device. In a system that is equipped with multiple handling devices for transporting media, and records and plays back data in files created on media, it is possible to predict the future request time, frequency, etc. for each media, and to predict future usage times and frequencies based on the results. An information medium automatic storage system characterized by controlling a medium that is likely to be stored permanently on a drive device for a certain period of time. 2. For information media such as digital video tapes and optical discs, shelves for storing multiple media, multiple recording/playback devices, and multiple storage devices for transporting media between the shelves and recording/playback devices. In a system that records and plays back data in files created on media and is equipped with several handling devices, the future of each media is determined based on data regarding the past usage history of the media or data regarding the execution scheduling of computer jobs. 1. An information medium automatic storage system characterized by predicting a requested time and controlling the medium whose request time is early to reside on a recording/reproducing device for a certain period of time. 3. For information media such as digital video tapes and optical discs, shelves for storing multiple media, multiple recording/playback devices, and multiple storage devices for transporting media between the shelves and the recording/playback devices. In a system that records and plays back data in files created on media and is equipped with several handling devices, the future of each media is determined based on data regarding the past usage history of the media or data regarding the execution scheduling of computer jobs. An information medium automatic storage system characterized by predicting the request frequency of , and controlling such that a medium with a high request frequency resides on a recording/reproducing device for a certain period of time. 4. For information media such as digital video tapes and optical discs, shelves for storing multiple media, multiple recording/playback devices, and multiple storage devices for transporting media between the shelves and recording/playback devices. In a system equipped with a handling device for recording and reproducing data in files created on a medium, in addition to the medium on which recording or playback processing is to be started or the medium in which processing is being performed, the medium that has completed processing is also recorded as is. /mounted on a playback device, and if some of the media on the device need to be demounted in order to mount new media, future request times, frequency, etc. for each media on the device. An automatic storage system for information media is characterized in that it predicts information media and, based on the results, selects and demounts them in order of decreasing possibility of future use. 5. For information media such as digital video tapes and optical discs, shelves for storing multiple media, multiple recording/playback devices, and multiple storage devices for transporting media between the shelves and the recording/playback devices. In a system equipped with a handling device for recording and reproducing data in files created on a medium, in addition to the medium on which recording or playback processing is to be started or the medium in which processing is being performed, the medium that has completed processing is also recorded as is. / mounted on a playback device, and you need to demount some of the media on the device in order to mount new media. An information medium automatic storage system is characterized in that the future request time of each medium is predicted based on data regarding execution scheduling, and the information medium is selected and demounted in descending order of the request time. 6. For information media such as digital video tapes and optical discs, shelves for storing multiple media, multiple recording/playback devices, and multiple storage devices for transporting media between the shelves and the recording/playback devices. In a system equipped with a handling device for recording and reproducing data in files created on a medium, in addition to the medium on which recording or playback processing is to be started or the medium in which processing is being performed, the medium that has completed processing is also recorded as is. / mounted on a playback device, and you need to demount some of the media on the device in order to mount new media. An information medium automatic storage system is characterized in that it predicts the future request frequency of each medium based on data regarding execution scheduling, and selects and demounts the media in descending order of frequency. 7. In the information media automatic storage system set forth in Paragraphs 1 to 6, if it is necessary to leave a predetermined number of recording/playback devices in an empty state with no media mounted, the empty state is For those devices, the medium is demounted at the end of recording and playback processing, and the controls described in items 1 to 6 are performed using other devices. 8. In the information media automatic storage system set forth in paragraphs 4 to 6, if it is necessary to leave a predetermined number of recording/playback devices in an empty state with no media mounted, the media on those devices Some of them are selected and demounted in the same manner as in Items 4 to 6. 9. In the information medium automatic storage system set forth in paragraphs 2 to 6, as data regarding the past usage history of each medium,
The time when the media was last used is used as the reference. 10. In the information medium automatic storage system of Sections 2 to 6, the data regarding the past usage history of each medium is based on the number of times it has been used within a predetermined period. 11. In the information medium automatic storage system set forth in paragraphs 1 to 6, if a certain recording/playback device is designated as the recording/playback device that should process the specified medium,
Management data such as the start and end of use of the device is reported to the specified device as if it were to be used virtually, and in reality, the specified medium is already mounted. If you have a playback device, use that device. 12. In the information medium automatic storage system set forth in paragraphs 1 to 6, if a specific recording/playback device is designated as the recording/playback device that should process the specified medium,
Management data such as the start and end of use of the device is reported to the specified device as if it were to be used virtually, and in reality, the specified medium is already mounted. If there is no playback device, determine the recording/playback device to be demounted based on the criteria described in Sections 4 to 6 and use that device. 13. In the information medium automatic storage system set forth in paragraphs 1 to 6, the man-hours required for mounting/demounting the medium, which occurs when the medium containing the requested file is not on the recording/playback device, is reduced as much as possible. Schedule the execution order of jobs. 14. In the information medium automatic storage system of Section 13,
For each job that is requested to be executed, the number of man-hours required for mounting/demounting the media is predicted, and the job is executed starting from the one with the smallest number of man-hours. 15. In the information media automatic storage system of Section 13,
The total man-hours for mounting/demounting operations are predicted for all combinations of execution orders for all or some of the jobs requested to be executed, and the jobs are executed in accordance with the execution order that minimizes the total man-hours. 16. In the information media automatic storage system described in Paragraphs 1 to 6, in order to process requests from multiple jobs in parallel as much as possible, data regarding past media usage history or computer job execution schedule Determine file placement on media based on data about the ring. 17. In the information medium automatic storage system set forth in Section 16,
A plurality of files used in the same job are arranged in a concentrated manner on as few media as possible based on data regarding past media usage history or data regarding execution scheduling of computer jobs. 18. In the information medium automatic storage system of Section 16,
Recording and playback requests are made from multiple jobs to a group of files stored on the same medium during the same time period based on data related to past media usage history or data related to execution scheduling of computer jobs. Determine file placement on the media so that conflicts do not occur. 19. In the information media automatic storage system set forth in paragraphs 1 to 6, data from multiple jobs in the same time period is based on data regarding past media usage history or data regarding execution scheduling of computer jobs. To change job execution scheduling so that competing recording and reproduction requests are not generated for a group of files stored in the same medium. 20. In the information medium automatic storage system set forth in paragraphs 1 to 6, when writing data, the requested medium is not recorded/recorded.
If it is not mounted on the playback device, a storage means prepared in advance or another medium already mounted is used as a temporary storage area, and after that, writing is performed normally on the requested medium. 21. In the information media automatic storage system of Section 20,
When reading data, if the requested medium is not mounted on the recording/reproducing device, if data to be read exists in the temporary storage area, that data is read. In the information medium automatic storage system of Items 22, 20, and 21, a medium for temporary storage is resident in a recording/reproducing device in an automatic storage, and this medium is used. 23. In the information medium automatic storage system of Sections 20 and 21, the free space of another medium that has already been mounted is used as a temporary storage area. 24. In the information medium automatic storage system of Section 20,
The frequency of occurrence of medium requests is monitored at regular intervals, and when the frequency is lower than a predetermined value, data in the temporary storage area is normally written to the requested medium. 25. In the information media automatic storage system of Section 20,
Based on data regarding job execution scheduling, data in a temporary storage area is normally written to a requested medium when the frequency of occurrence of a medium request is lower than a predetermined value. 26. In the information medium automatic storage system set forth in Section 20,
At a predetermined time or when specified by the user,
The data in the temporary storage area is properly written to the requested medium.