JPH06225092A - Storage controller and method therefor - Google Patents

Abstract

PURPOSE:To reduce the overhead of a transfer protocol, and to attain efficient division-storage by obtaining an optimal transfer unit for the various kinds of storage devices which store information on a rotation type storage medium, and operating the division-storage by the optimal transfer unit. CONSTITUTION:A compressor 3 increase the number of originals to be stored in a hard disk 9 by compressing picture data, and shortens data transferring time to the hard disk 9. A buffer memory 4 is used when the data output of the compressor 3 is not a constant speed, long time is required for the start of the data transfer to the hard disk 9, and the transferring speed is not constant. A timer 19 changes the transfer unit for the various kinds of hard disks, and measures processing time for the constant amounts of data, that is, time from the start of data storage in the buffer memory 4 to the end of data storage in the hard disk 9. Then, the transfer unit whose processing time is the minimum is used as the optimal transfer unit for the hard disk 9, and a non-volatile memory 20 operates the division-storage by the optimal transfer unit.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a storage control device and method for temporarily storing data in a buffer memory and storing the data in a storage device using a rotary recording medium.

[0002]

2. Description of the Related Art As a storage controller for temporarily storing data in a buffer memory and storing the data in a storage device using a rotary recording medium, a copying machine or a facsimile machine described in Japanese Patent Laid-Open No. 2-81563. And the like are used in image recording devices. This is because at the same time the image data read by the document reading means is stored in the storage device,
The image recording means records the first copy, and the second and subsequent copies are recorded using the image data read from the storage device. In such an image recording device, it has been proposed to use a large-capacity hard disk or optical disk as a storage device. However, since the data transfer speed of the hard disk or optical disk is lower than the normal signal transfer speed, a buffer memory that absorbs the speed difference. However, the page image data read by the image reading means is temporarily stored in the buffer memory and then stored in the storage device after the storage is completed, which has a drawback that it takes time to store the image.

FIG. 11 shows a timing chart of the operation of a conventional digital copying machine. As can be seen from FIG. 11, the page image data read by the image reading means is temporarily stored in the buffer memory and is stored in the storage device after the storage is completed, so it takes time to store the image.

Further, Japanese Laid-Open Patent Publication No. 3-145266 discloses a facsimile machine which stores image data for transmission reservation in a storage device. In order to solve the problem that this device takes a long time to store an image, the image data is sequentially distributed and stored in a memory composed of a memory block group of a predetermined size, and if a stored memory block is detected during storage, that memory block is stored. The data is stored in a storage device. With such a configuration, writing to the storage device can be started at the time when a certain unit of image data is stored in the memory, and the storage end in the storage device can be completed earlier than when the page image data is stored collectively. Become.

[0005]

Generally, in a storage device using a rotary storage medium, the smaller the data transfer unit is, the more the waiting for rotation of the recording medium and the overhead of the transfer protocol increase. Therefore, the transfer speed to the storage device is high. Get lower.

The divisional storage of page image data as disclosed in Japanese Patent Laid-Open No. 3-145266 seems to be efficient, but the division unit is reduced in order to speed up the start of writing to the storage device. This means that the unit of transfer to the storage device is reduced, and therefore the speed performance of the storage device is reduced. For this reason, if the division unit is made extremely small, the transfer speed to the storage device may be extremely reduced, and the end of storage in the storage device may be slower than in the case where division storage is not performed. In order to effectively perform divided storage, there are two conflicting requirements: one is to reduce the unit of division to speed up the start of writing to the storage device, and the other is to increase the unit of division to improve the storage speed in the storage device. You must choose a division unit that meets the requirements of

The division unit, in other words, the transfer unit to the storage device must be determined based on the data transfer speed to the buffer memory and the speed characteristic of the storage device. However, since various storage devices having different performances may be used depending on the purpose, information regarding the speed characteristics of various storage devices is required. The speed characteristic of the storage device depends on the physical configuration of the storage device, but even if information about the physical configuration of the storage device is obtained, it is difficult to obtain the speed characteristic by calculation, and it is optimal by an experimental method. It is required to obtain a proper data transfer unit.

In view of the above problems, the present invention automatically finds an optimum transfer unit when performing divided storage for various storage devices that store information in a rotary storage medium, and divides the storage by the transfer unit. It is an object of the present invention to provide a storage control device and method for performing the above.

[0009]

In order to solve the above-mentioned problems, the present invention provides a storage control device for temporarily storing data in a buffer memory and storing the data in a storage device using a rotary recording medium. Corresponding to the data amount set by the identification information reading means for reading the identification information indicating the type of the storage device from the storage device, the transfer unit setting means for setting the unit to be transferred to the storage device, and the transfer unit setting means. Data storage detecting means for detecting that the data to be stored in the buffer memory and time counting means for measuring the time from the start of storage of a certain amount of data in the buffer memory to the end of storage in the storage device And when the data storage detection means detects storage of data in the buffer memory corresponding to the amount of data set by the transfer unit setting means. , Divided storage control means for controlling the divided data to be stored by storing data corresponding to the value in the storage device, and various transfer units are set by the transfer unit setting means, With respect to the value, the processing time required from the start of storage of a certain amount of data in the buffer memory to the end of storage in the storage device is measured by the time measuring means, and a transfer unit that minimizes the processing time is obtained. An optimum transfer unit measuring unit that is an optimum transfer unit to the storage device when storing the transfer unit by dividing the data; identification information indicating the type of the storage device obtained by the identification information reading unit; An optimum transfer unit storing means for storing the optimum transfer unit obtained by the optimum transfer unit measuring means is provided.

Further, according to the present invention, in a storage control method for temporarily storing data in a buffer memory and storing the data in a storage device using a rotary recording medium, identification information indicating a type of the storage device is provided. Whether the unit to be read from the storage device by the identification information reading device and transferred to the storage device is set by the transfer unit setting device, and the data corresponding to the data amount set by the transfer unit setting device is stored in the buffer memory. Whether or not the data storage detection means detects the storage, and when the data storage detection means detects that the data corresponding to the data amount set by the transfer unit setting means is stored in the buffer memory, corresponds to the value. By storing the data in the storage device, the storage in the storage device is divided so that the data is divided and stored. Control time, various transfer units are set by the transfer unit setting means, and the processing time required from the start of storage of a certain amount of data in the buffer memory to the end of storage in the storage device for each value Is measured by the time measuring means, the transfer unit with the minimum processing time is obtained by the optimum transfer unit measuring means, and this transfer unit is set as the optimum transfer unit to the storage device when the data is divided and stored. It is characterized in that the identification information indicating the type of the storage device obtained by the information reading means and the optimum transfer unit obtained by the optimum transfer unit measuring means are stored in the optimum transfer unit storage means.

[0011]

In the present invention, the divided storage control means corresponds to the value when the data storage detection means detects the storage in the buffer memory of the data corresponding to the data amount set by the transfer unit setting means. By storing the data in the storage device, the data is controlled to be divided and stored. When obtaining the optimum transfer unit, the optimum transfer unit measuring means sets various transfer units by the transfer unit setting means, and for each value, stores a certain amount of data in the storage device from the start of storage in the buffer memory. The processing time required until the end is measured by the time measuring means, the transfer unit that minimizes the processing time is obtained, and this is set as the optimum transfer unit to the storage device when the data is divided and stored. During normal data storage, the optimum transfer unit is designated by the transfer unit setting means. The storage device can be replaced with various storage devices having different storage capacities, access times, etc. When the storage device is replaced, the identification information of the storage device is read by the identification information reading means, and the optimum transfer unit storage means is referred from this identification information. If the optimum transfer unit is not stored, the optimum transfer unit measuring means measures the optimum transfer unit.

[0012]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT An embodiment in which the present invention is applied to a digital copying machine will be described below.

FIG. 1 is a basic block diagram of the present invention. The image input means comprises an automatic document feeder 1 and an image reading device 2, and the image reading device 2 is a CCD (Cha).
A document is converted into electrical image data and read by an image pickup device such as a large coupled device. The automatic document feeder 1 sequentially sends out documents to a predetermined reading position of the image reading device 2, and discharges them after the reading is completed. The compressor 3 is used to increase the number of originals that can be stored in the hard disk 9 by compressing the image data and to shorten the data transfer time to the hard disk 9. The buffer memory 4 is used because the data output of the compressor 3 is not constant, and it takes time to start the data transfer of the hard disk 9 and the transfer speed is not constant. Also, it operates as a ring memory in which the start address and the end address are connected. The hard disk 9 is used as a data storage means. SCSI (Small Co
A computer system interface controller 8 controls a hard disk 9 having a SCSI interface. The hard disk 9 is removable, and another SCSI hard disk can be connected. The decompressor 10 decompresses the compressed image data. The image output means comprises an image recording device 12 and an automatic paper binding device 11. The image recording device 12 records the image data expanded by the expander 10 on a paper by an electrophotographic recording method using a laser or the like. The automatic sheet binding device 11 binds sheets for each number of copies sequentially output by the image recording device 12. Reference numeral 7 denotes a buffer memory counter, which counts the image data amount stored in the buffer memory 4, and when the count value matches the value set in the base count register 7a, a CPU (Central Process)
g Unit) 15 is notified, the count value is reset to 0, and then counting is started again. Buffer memory counter 7
Is connected to a current count register 7b indicating the current count value, and the CPU 15 can know the count value at any time via this register.

DMA (Direct Memory A)
access controller 6 has 3 channels, channel 1 is DMA transfer from compressor 3 to buffer memory 4, channel 2 is DMA transfer from buffer memory 4 to decompressor 10, channel 3 is buffer memory 4 and SC.
DMA transfer between SI controllers 8 is performed. The bus arbiter 5 includes a compressor 3, a SCSI controller 8 and an expander 10.
The DMA transfer requests from the three devices are arbitrated in a predetermined priority and a predetermined data unit. Also, the overflow of the buffer memory 4 used as the ring memory is monitored.

The above is a description of the image data system.
The image data bus 21 for transferring the image data is a CPU bus 2 of a control system described later in order to operate at a higher speed.
Separated from 2.

The control means is composed of each device connected to the CPU bus 22 of the control system. The CPU 15 is a main control device and controls each device connected to the CPU bus 22. PR
OM (Programmable Read Only)
The memory 16 stores software for operating the CPU 15. RAM (Random Ac
The cess Memory 17 is a working memory of the CPU 15. The interrupt controller 18 includes the compressor 3 and the S
Interrupts of the CSI controller 8, the decompressor 10, the buffer memory counter 7, and the communication controller 14 are accepted in a predetermined priority order, and the CPU 15 is prompted to perform processing. The communication controller 14 is a device for transmitting / receiving a command / status between the CPU 15, the image input means, the image output means, and the user interface 13. The user interface 13 is a device for detecting an instruction from an operator and displaying the state of the machine. The timer 19 measures the time from the start of data storage in the buffer memory 4 to the end of data storage in the hard disk 9. Non-volatile memory 2
0 stores the measured optimum transfer unit to the hard disk 9.

Next, the structure of the hard disk will be outlined. A hard disk has one or more discs, and a read / write head is provided on each side of the disc. The disk rotates at a constant speed, and the head moves in the radial direction. The address of the data on the hard disk starts from one round (= 1 track) of the outermost disk of the head 0, and then the head 1
This is followed by one track on the outermost periphery of the disk and one track of the head n, and these n tracks are one cylinder. When one cylinder is over, it becomes an adjacent inner cylinder, and the head returns to 0. The same applies to the innermost cylinder. It takes time because the head electrically switches each time one track is completed. When the cylinders are switched, the head moves electrically (= seek) by one cylinder at the same time as the head is electrically switched. The time required for this is the cylinder switching time, which is longer than the normal head switching time.

Next, the measurement of the optimum transfer unit will be described. The measurement process of the optimum transfer unit may be performed only when the hard disk is newly installed, such as when the hard disk is replaced, and therefore, the copying machine is operated in the diagnostic (diagnostic) mode. The diagnostic mode is one of the operation modes of the copying machine and exists for maintenance of the machine. From the user interface 13,
When the measurement of the optimum transfer unit is instructed, the processes shown in the flowcharts of FIGS. 2 and 3 are started.

In FIG. 2 and FIG. 3, first, information regarding the identification of the hard disk is read from the hard disk 9 (S1). Specifically, issue the SCSI Inquiry command to read the product name of the hard disk,
Stored in a predetermined area of AM17. Then, the optimum transfer unit management table in the non-volatile memory 20 is referred to. The optimum transfer unit management table is shown in FIG. The optimum transfer unit management table is a correspondence table between the name of the hard disk and the optimum transfer unit, and records the optimum transfer unit that has already been measured. By referring to this table, it is judged whether or not the optimum transfer unit of the currently connected hard disk has already been obtained (S2). If the optimum transfer unit of the currently connected hard disk has already been obtained, the transfer unit recorded in the optimum transfer unit management table is set in the base count register 7a (S18), and the optimum transfer unit measurement process is performed. To finish.

If the optimum transfer unit unit of the currently connected hard disk has not been measured, the optimum transfer unit unit is measured in the subsequent processing. First, information about the standard document is input from the user interface 13 (S3). The information about the standard document is the resolution, the amount of information per pixel, the compression rate, and the standard document size. Then, the image information amount of the standard document is calculated based on these information (S
4). Table 1 shows an example of information about standard manuscripts. In addition,
dpi means dots / inch.

[0021]

[Table 1] From Table 1, the image information amount of the standard document can be calculated as follows.

Image information amount of standard document = (number of pixels of standard document size) × (information amount per pixel) × (compression rate) =
(4672 pixels / line x 3296 lines / page) x
(4 bits / pixel) × 1 / 5≈1.54 Mbytes Next, the transfer unit is variously changed, and the time from the storage of the image data in the buffer memory 4 to the end of the storage in the hard disk 9 is measured to obtain the optimum value. Calculate the transfer unit. The initial value of the transfer unit is 16 Kbytes (S5). Initial value is 16
The reason for setting K bytes is that this value corresponds to approximately one track, and in a transfer unit smaller than this, it is considered that transfer is inefficient because the transfer waits at least once for each transfer. . The transfer unit is set in the base count register 7a (S6), and the timer is started at the same time when the image reading apparatus 2 starts reading the image (S7). The read image data is compressed by the compressor 3, and the compressed image data is sequentially stored in the buffer memory 4. The buffer memory counter 7 counts the amount of data stored in the buffer memory 4. The CPU 15 receives the count end signal (CN
T) or waiting for a signal (EOC) indicating the end of compression of the image data from the compressor 3 (S8), and when receiving the CNT signal (S9), the hard disk 9 stores the image data of the transfer unit. Control (S10). The buffer memory counter 7 resets the count value at the same time when the CNT signal is output, and counts from 0 again. CPU15
Waits for the reception of the CNT signal or EOC signal again (S
8). When a further CNT signal is received (S9), image data for the transfer unit is stored (S10), and the next detection is waited (S8). When the CPU 15 receives the EOC signal (S9), it stores the unstored image data in the buffer memory 4 corresponding to the value indicated by the current count register 7b.
It is controlled to be stored in the hard disk (S11). When the storage on the hard disk is completed, the timer is stopped (S12), the time required from the start of reading the image to the end of the storage on the hard disk is read out, and the RAM 17 is read.
It is stored in the measurement result storage table (S13). The measurement result storage table is a table for temporarily storing the measurement result.

Next, a new transfer unit is obtained by doubling the transfer unit (S14). Here, the transfer unit is changed by a factor of two because the transfer speed of the hard disk is generally logarithmically proportional to the transfer unit as shown in FIG. In this embodiment, the transfer unit was changed by 2 times, but
When dealing with a small amount of image information such as a value image,
It is desirable to change the transfer unit smaller.

If the new transfer unit is less than the image information amount of the standard original, the measurement is further continued (S15). If not, the minimum processing time is selected from the processing times for each transfer unit recorded in the measurement result storage table, and this is set as the optimum transfer unit (S16). The obtained optimum transfer unit is stored in the optimum transfer unit management table together with the product name of the hard disk read in S1 (S17). Finally, the optimum transfer unit is set in the base count register 7a (S1
8), the image data for the optimum transfer unit is stored in the buffer memory 4
When it is stored in, the CPU 15 is notified and the optimal transfer unit measurement process is terminated.

In this embodiment, the measurement of the optimum transfer unit is performed in the diagnostic mode. However, by storing the information about the standard document in the non-volatile memory in the copying machine in advance, S3 in FIG. S4 can be omitted, and the measurement process of the optimum transfer unit can be performed at system startup after the power of the copying machine is turned on.

As a concrete example, the scanning speed is 160 mm.
FIG. 6 shows a measurement result storage table when a standard original of one page is read and stored in the hard disk by using the image reading device of 1 / sec and the hard disk having the specifications shown in Table 2.

[0027]

[Table 2] From FIG. 6, the processing time is the shortest when the transfer unit is 256 Kbytes, so the optimum transfer unit is 256 Kbytes, and this is stored in the optimum transfer unit management table together with the hard disk name as shown in FIG.

Next, the manner in which divided storage is performed in the normal copy mode of the digital copying machine of this embodiment will be described with reference to the flow charts of FIGS.

The copy mode and the number of copies are set on the user interface 13, and the operator presses the start button to start the copy job (S101). The image input means starts inputting an image (S102), and the input image data is compressed by the compressor 3 (S103). The compressed image data is stored in the buffer memory 4, and the buffer memory counter 7
Starts counting (S104). The CPU 15 waits for reception of the CNT signal from the buffer memory counter 7 or the EOC signal from the compressor 3 (S105). Since the optimum transfer unit is already set in the base count register 7a in the measurement process of the optimum transfer unit in the diagnostic mode, the buffer memory counter 7 stores the image information amount corresponding to the optimum transfer unit in the buffer memory 4. The CNT signal is output at the point of time. When the CNT signal is received (S106), the hard disk 9 is controlled to store the image data corresponding to the optimum transfer unit (S107). The buffer memory counter 7 resets the count value at the same time when the CNT signal is output, and counts from 0 again. CPU15 is CNT again
It waits for the reception of a signal or an EOC signal (S105). When a further CNT signal is received (S106), image data corresponding to the optimum transfer unit is stored (S107), and the next detection is awaited (S105). When the CPU 15 receives the EOC signal (S106), the current count register 7b
The unstored image data in the buffer memory 4 corresponding to the value indicated by is stored in the hard disk 9 (S108). The image data on the buffer memory 4 is stored in the hard disk 9 (S107, S1).
08), and is transferred to the decompressor 5 to undergo decompression processing (S10
9) and recorded by the image recording device 12 (S110). By simultaneously storing the image data (S107, S108) and expanding and recording the image data (S109, S110), it is possible to output the first copy at high speed. After the storage of the image data is completed, if the input of all originals is not completed (S111), the process from image input (S102) is repeated. When the input of all originals is completed, the copy output for the second and subsequent copies is started. First, one page of image data is read from the hard disk 9 (S112), and the buffer memory 4
(S113). When the storage of one page of image data in the buffer memory 4 is completed, the image data in the buffer memory 4 is transferred to the decompressor 5 and decompressed (S11).
4) Then, the image is sent to the image recording device 12 for recording (S115). This is continued until the final output is completed (S116).

A normal digital copying machine develops image data page by page in order to realize high-speed copy output.
It uses an image recording device that feeds, transfers, fixes, and ejects paper. To use such an image recording device, 1
Since it is necessary to transfer the image data for a page to the image recording device at a constant speed, from a storage device such as a hard disk that cannot read data at a constant speed,
Data cannot be transferred directly to the image recording device. For this reason, decompression and recording are performed after securing one page of data in the buffer memory.

FIG. 11 shows a timing chart of the operation of the conventional digital copying machine, and FIG. 10 shows a timing chart of the operation of the digital copying machine according to the present embodiment. In the present embodiment, as a result of accelerating the end of storage in the storage device due to efficient division storage, the end of read from the storage device for the output of the second copy is accelerated, so that the output timing of the second copy can be accelerated. , CPM (Copy Per Minute)
s) is improved.

According to the present embodiment, any hard disk having a SCSI interface automatically obtains the optimum transfer unit and performs divided storage in the optimum transfer unit, so that the speed performance of each hard disk is maximized. It can be used to the limit. This makes it possible to flexibly deal with cases such as connecting a high-performance hard disk to improve the performance of the entire device, or connecting an inexpensive and low-performance hard disk to reduce the cost of the entire device. It will be possible. Although SCSI is used as the hard disk interface in this embodiment,
As the interface, PC / AT (registered trademark of IBM) or SASI (Shugart Associate)
Other interfaces such as s System Interface) may be used.

[0033]

As described above, according to the present invention, various storage devices for storing information on a rotary recording medium are provided.
Since the optimum transfer unit is automatically obtained and the divided storage is performed in the optimum transfer unit, the overhead of the recording medium rotation and the transfer protocol, which are problems in the divided storage, are reduced, and the efficient divided storage can be performed.

[Brief description of drawings]

FIG. 1 is a block diagram showing a basic configuration of an embodiment of the present invention.

FIG. 2 is a first part of a flowchart of a measurement process of an optimum transfer unit.

FIG. 3 is a second part of the flowchart of the measurement process of the optimum transfer unit.

FIG. 4 is a diagram showing an optimum transfer unit management table.

FIG. 5 is a diagram showing a relationship between a transfer unit of a hard disk and speed.

FIG. 6 is a diagram showing a measurement result storage table.

FIG. 7 is a diagram showing an optimum transfer unit management table in which a newly measured optimum transfer unit is entered.

FIG. 8 is a first flowchart of the copying operation according to the embodiment.
Part of.

FIG. 9 is a second flowchart of the copying operation according to the embodiment.
Part of.

FIG. 10 is an operation timing chart of the digital copying machine of the embodiment.

FIG. 11 is an operation timing chart of a conventional digital copying machine.

[Explanation of symbols]

1 ... Automatic document feeder, 2 ... Image reading device, 3 ... Compressor, 4 ... Buffer memory, 5 ... Bus arbiter, 6 ... D
MA controller, 7 ... Buffer memory counter, 7a
... base count register, 7b ... current count register, 8 ... SCSI controller, 9 ... hard disk, 10 ... expander, 11 ... automatic paper binding device, 12 ... image recording device, 13 ... user interface, 14 ... communication controller, 15 ... CPU, 16 ... PROM, 17 ...
RAM, 18 ... Interrupt controller, 19 ... Timer, 20 ... Non-volatile memory, 21 ... Image data bus, 2
2 ... CPU bus

Claims (4)

[Claims] 1. A storage control device for temporarily storing data in a buffer memory and storing the data in a storage device using a rotary recording medium, comprising: identifying information indicating a type of the storage device from the storage device. Identification information reading means for reading, transfer unit setting means for setting a unit to be transferred to the storage device, and detecting that data corresponding to the data amount set by the transfer unit setting means is stored in the buffer memory Data storage detection means, time measurement means for measuring the time from the start of storage of a certain amount of data in the buffer memory to the end of storage in the storage device, the data storage detection means, the transfer unit setting means When the storage of the data corresponding to the set data amount in the buffer memory is detected, the data corresponding to the value is stored in the storage device. The Rukoto sets the divided memory control means for controlling so as to divide and store data, and various transfer unit by the transfer unit setting means,
For each value, the processing time required from the start of storage of a fixed amount of data in the buffer memory to the end of storage in the storage device is measured by the time measuring means, and the transfer unit that minimizes the processing time is obtained. An optimum transfer unit measuring unit which is an optimum transfer unit to the storage device when storing the transfer unit by dividing the data, and identification information indicating the type of the storage device obtained by the identification information reading unit, A storage control device comprising: an optimum transfer unit storage unit for storing the optimum transfer unit obtained by the optimum transfer unit measuring unit. 2. A storage control method for temporarily storing data in a buffer memory and storing the data in a storage device using a rotary recording medium, wherein identification information indicating the type of the storage device is read out as identification information. The transfer unit setting means sets a unit to be read from the storage device and transferred to the storage device, and a data indicating whether or not data corresponding to the data amount set by the transfer unit setting means is stored in the buffer memory. When the storage detection means detects that the data storage detection means stores the data corresponding to the data amount set by the transfer unit setting means in the buffer memory, the data corresponding to the value is stored. By storing in the device, the storage in the storage device is controlled by the divided storage control means so that the data is divided and stored. Set the various transfer unit by the transfer unit setting means,
For each value, the processing time required from the start of storage of a certain amount of data in the buffer memory to the end of storage in the storage device is measured by the time measuring means, and the processing time is minimized by the optimum transfer unit measuring means. Find the transfer unit
This transfer unit is an optimum transfer unit to the storage device when dividing and storing the data, and the identification information indicating the type of the storage device obtained by the identification information reading means and the optimum transfer unit measuring means are obtained. A storage control method characterized in that the optimum transfer unit thus obtained is stored in an optimum transfer unit storage means. 3. The storage according to claim 2, wherein during normal data storage, an optimum transfer unit is designated in the transfer unit setting means, and the divided storage control means stores the data in a divided manner. Control method. 4. The storage device can be replaced with various storage devices having different storage capacities, access times, etc., and when the storage device is replaced, identification information indicating the type of the storage device by the identification information reading means. Reading the optimum transfer unit from the identification information, and when the optimum transfer unit is not stored, the optimum transfer unit measuring unit measures the optimum transfer unit. Item 4. The storage control method according to Item 3.