JPH04360246A - Device for compressing file - Google Patents

Abstract

PURPOSE:To attain efficient file compression by independently storing a common part and a non-common part excluding the common part when common contents are included in plural files. CONSTITUTION:When a user inputs a file compression processing command and n existing file names required to be compressed, an operating system(OS) mutually compares the contents data of the n files by a comparing processing part 11 and judges whether a completely coincident part of contents exists or not. At the time of judging the existence of the coincident contents part, the part is regarded as a common part, a common part control word indicating the common part is added to the common part by a common part processing part 12 and the common part and its control word are stored in a secondary storage device 30. Then processing for specifying that the compared files have the specific common part on their specific positions is applied to the compared files and the specified files are stored in the device 30. The file processing is repeated for all the n files. Thus efficient file compression can be attained.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a file management method performed by an operating system in a computer system, and more particularly to a file compression device for efficiently compressing a plurality of files having common parts.

0002

[Background Art] In computer systems, data compression technology has been proposed and put into practical use in order to efficiently use secondary storage devices, which are storage areas, to reduce data storage costs and to economize transfer costs. has been made into Data compression refers to suppressing redundancy in data by performing data conversion, and expressing data concisely with a shorter data length.

[0003] As a data compression method that is currently widely used, there is Huffman's optimal encoding method. this is,
The frequency of appearance of patterns in data is statistically investigated, and shorter codes are assigned to patterns that appear more frequently.By increasing the number of patterns, it is possible to shorten the average code length. Furthermore, the Ziv-Lempel data compression method does not require assuming statistical properties or stationarity of data, and can be directly applied to any symbol string. According to this data compression method, long symbol strings can be compressed efficiently, so it is suitable for compressing various files created by computer systems.

However, this type of data compression technology has the following problems. In other words, if multiple files have the same data, the storage area (
Multiple copies of the same data will be stored in the secondary storage device. This wasted area in the secondary storage device increases as the size of the part with common content increases and the number of files that have it increases, which hinders the efficient use of the secondary storage device. .

On the other hand, as described above, data compression processing is performed on data files individually, and each data file is converted into a smaller data file. Therefore, conventional data compression processing cannot eliminate the uneconomical effects of using a secondary storage device due to the common content inherent in a plurality of files.

[0006] Also, in order to solve this problem, integrated processing that takes into account the contents of multiple files is required, but such a processing method has not been implemented so far and has not yet been put into practical use. It has not been.

[0007]

[Problems to be Solved by the Invention] As described above, although conventional data compression processing has a compression effect on individual data files, it is difficult to compress secondary storage devices due to the common content inherent in multiple files. Uneconomical use could not be eliminated.

The present invention has been made to solve the above-mentioned problems, and its purpose is to avoid inconveniences such as redundant storage of content common to multiple files in a secondary storage device. An object of the present invention is to provide a file compression device that enables more efficient use of a secondary storage device.

[0009]

[Means for Solving the Problems] The gist of the present invention is to independently store common parts and non-common parts other than the common parts when a plurality of files have common contents.

That is, the present invention provides a file compression device for efficiently compressing a plurality of files, which includes means for comparing the contents of a plurality of files stored in a first storage unit, and means for extracting a portion in which the common portions match; means for storing the extracted common portion as a common file in a first storage unit or a second storage unit; and means for storing a file after extracting the common portion as a unique file in the first storage unit. or a means for storing the information in the second storage section.

[0011] In the present invention, as a result of comparing the contents of a plurality of files, there may be a plurality of common parts determined within one file. In addition, the comparison is not limited to only between all files specified by the user, but it is also possible to perform comparison processing on arbitrary multiple files among the specified files and determine the common parts between the files. good.

0012

[Operation] According to the present invention, very efficient file compression can be performed in the following situations.

For example, assume that a computer simulation is being performed to investigate the time evolution of a system. In such simulations, time is discretized and
The state of the system is calculated for each discretized time step and the time evolution is followed. Graphical representation of simulation results is an extremely effective means of assisting in understanding the changes in phenomena at each calculated time step. Create a file. and,
Based on that data file, a video is created by connecting still images from each time step.

Generally, the amount of data output for graphics is large even for one time step, and the size of the entire data file is enormous. Incidentally, one of the characteristics of simulation is its arbitrariness in that it is possible to trace the state of the system under the initial values and conditions desired by the executor. Therefore, an example of executing a simulation is to perform calculations many times with the same initial value while changing conditions, and to examine differences in time evolution due to different conditions.

In this way, when simulations are repeated with the same initial values but with different conditions, and a data file is created to graphically represent the results, the initial data (for one time step) is output to the data file. data (which is the initial screen in the video) is common to all data files. However, as mentioned above, the size of graphics data is extremely large even for one time step, so
Having multiple data files sharing the same initial data creates wasted space in the secondary storage device, which hinders efficient use of the storage device.

[0016] In such a case, if the file compression process of the present invention is performed, only one piece of initial data will be saved, so the secondary storage device can be used efficiently, and the user can The data files that have undergone this process can be treated as each holding initial data. Needless to say, if the initial data is common between files, it is not necessary to output the initial data to all data files, but it is possible to visually accurately capture the time evolution of the system and to In order to better understand the physical and chemical phenomena occurring, it is highly desirable that initial data be included in the graphics data file.

[0017]Also, it is possible to divide the initial data and the data of subsequent time steps and output them to separate files, but if the data files are divided in this way, it will be difficult to create graphics afterward. Processing and data file management become quite complicated. Therefore,
It can be said that file compression processing is effective not only for effective use of secondary storage devices, but also for users to use computer systems efficiently.

Another example in which file compression processing is effective is the execution of a program that calculates the value of x at each time t and outputs the values of t and x. If you run this program many times with the same time step but different conditions, the sequence of numbers related to time t output to the data file will be:
It will be the same for all data files created. Therefore, this file compression processing is also effective for these data files that include common content, such as the value of t. Also, if you want to add new or different functionality to an executable program that can be achieved with only a few changes,
Copy the source file, modify it, and use it. Even in such a case, several files containing common contents will be created, so performing file compression processing can contribute to effective use of the secondary storage device.

Furthermore, by performing comparison processing not only on all files but also on any plurality of specified files, it is possible to maximize compression efficiency. For example, suppose that there are three files, and the second file has many common parts with the first file, and the third file has very few common parts. In this case, if all the files are compared, the only common part will be the common part between the first and third files, and the compression efficiency will be extremely low. On the other hand,
By specifying the first and second files and performing comparison processing, the amount of data in the common portion increases, making it possible to improve compression efficiency.

[0020]

Embodiments Hereinafter, embodiments of the present invention will be described with reference to the drawings.

FIG. 1 is a block diagram showing a schematic configuration of a file compression device according to an embodiment of the present invention. 1 in the diagram
0 is a file compression processing unit according to the present invention, which includes a comparison processing unit 11 that compares multiple files, a common part processing unit 12 that extracts common parts of the compared files, and divides each file into common parts and non-common parts. File processing section 1
It is made up of 3rd grade. 20 is a secondary storage device (first storage section) such as a magnetic disk, and 30 is also a secondary storage device (second storage section) such as a magnetic disk. Note that these secondary storage devices 20 and 30 are not necessarily independent, and may be a common device. The operation of file compression by the above device will be explained with reference to the flowcharts and file structures shown in FIGS. 2 to 13.

First, the basic file compression processing procedure will be explained according to the flowchart shown in FIG. When the user inputs a command for file compression processing and the names of n existing files to be compressed (step S1), the operating system compares the content data between the specified n files (step S2). , it is determined whether there is a part whose contents completely match (step S3). If there is no matching part, the file compression process ends.

On the other hand, if it is determined that there is a part whose contents match, that part is regarded as a common part, a common part control word indicating that it is a common part is added thereto, and the part is stored in the secondary storage device. (This processing is called common part processing) (step S4). Furthermore, the file after the comparison is processed to specify that the file owns a specific common part at a specific location, and the file is saved in a secondary storage device (this process (referred to as post-comparison file processing) (step S5). When this post-comparison file processing is performed on all n files, the file compression processing is completed. A file newly created by file compression processing, including common parts, is called a compressed file.

[0024] Here, a method for comparing content data between files and determining common parts will be briefly described. As an example, consider a case where a common portion between file 1 and file 2 is determined. First, assuming that file 1 has been divided into appropriate data lengths (for example, one line in this case), file 1
Compare all lines of file 2 with all lines of file 2 line by line starting from the first line. As a result of comparing a certain pair, if the two lines are not the same, the next pair is compared, but if they are the same, the following processing is performed. For example, if the i1th line of file 1 and the i2th line of file 2 are the same, then the (i1 +1)th line of file 1 and the (i
Compare the 2+1)th line. If these two lines are also the same, the (i1+2)th line and (i2+2)th line are compared.

In this way, as long as the two lines are the same, the next lines of file 1 and file 2 are compared. If it turns out that the (i1 + k) line of file 1 and the (i2 + k) line of file 2 are not the same, then the range from the i1 line to the (i1 + k-1) line of file 1 and the file From line i2 of 2 (i2 +k
-1) It is determined that the range up to the line is a common part. If a common portion of three files is to be determined, the common portion of files 1 and 2 and file 3 are compared in the same manner as described above.

Next, the common portion processing and the post-comparison file processing will be explained in more detail. First, in the common part processing in step S4 of FIG. 2, a common part control word is added to the common part as described above, and the file is saved in the secondary storage device as a file as shown in FIG. 3 (this is called a common file). . Here, the common part control word is the number of compressed files that share the common part (this is called the number of shared files).
Record. This common part control word is provided for updating, deleting, etc. the compressed file, as will be explained later.

As the post-comparison file processing in step S5 in FIG. 2, an index sequential file method can be used. First, a conventionally used index sequential file will be explained. An index sequential file is one index and several records (
The index has a structure that consists of a split file), and the index stores the values for specifying the record (the start and end positions in the secondary storage device where the record is saved) in order. . The operating system can process indexed sequential files by reading information about the records from the index and accessing the records in order. According to this, it is also possible to update, insert, and delete records.

In the post-comparison file processing of the present invention, this index sequential file method is applied, and the operating system performs processing according to the flowchart shown in FIG. First, the file after comparison is divided into a common part and other parts (this is called a unique part), and each part is regarded as a record (step S6). That is, a record is either a common part or a unique part, but if it is a common part, a common file containing a common part control word is regarded as a record. Next, values indicating the start and end positions of each record and a value distinguishing whether the record is a common part or a unique part are recorded in the newly provided index (step S7). Finally, the index and the unique portion of the record are saved in the secondary storage device (step S8), and the post-comparison file processing is completed.

Therefore, when file compression processing is performed using this index sequential file method, the compressed file has a structure as shown in FIG. It is possible to perform the same processing (update, deletion, etc.) on the compressed file created in this manner as on a normal index sequential type file. However, if the record is a common file, a common part control word is added to the beginning of the record, so the common part control word is skipped and only the common part is processed.

It is also possible to execute the post-comparison file processing using a processing method called a marking method. In this method, the operating system performs processing according to the flowchart shown in FIG. First, a common part is removed from the files that have been subjected to the comparison process, and marking is performed at that position to indicate that the common part exists (step S9). Next, a header is added to the file, and information regarding the common parts removed from the file is recorded therein (step S10). Step S9 and Step S1
The file processed by 0 is called a unique file,
Its structure is shown in FIG. Finally, the unique file is saved in the secondary storage device (step S11), and the post-comparison file processing ends.

[0031] As a specific marking method, a special character (this is called a common part designation character) is recorded at the position where the common part is extracted, and the extracted common part (common file) is recorded after that. Record the value you specify. This value is
For example, it is a value indicating the start position and end position in the secondary storage device where the common file is stored. Here, the common part specification character and the value following it are called a common part specification word. In addition, values indicating the start and end positions where the common file is saved are also recorded in the header. Therefore, a compressed file created by file compression processing using the marking method is composed of one unique file and a common file specified by the unique file. For example, when an operating system updates a compressed file created using a marking method, the compressed file can be handled in the same way as a normal file by performing the following process.

That is, in order to update the compressed file,
The unique file is transferred from the secondary storage device to the main storage or secondary virtual address space, which is a work area. When transferring,
It is determined for each bit whether or not it is a common part designation character, and if it is determined that it is a common part designation character, the common part designation word is replaced with the common part specified by it. In other words, the common part specification word is removed and the common part is read at that position. Through this process, the unique file transferred to the work area is converted into the same file as before the file compression process. However, the header of the specific file is not transferred.

According to the method described above, it is possible to realize the file compression processing of the present invention. next,
A process for deleting or updating a compressed file created by the method described above, and a specific procedure for further performing file compression processing on a compressed file will be described.

When deleting a compressed file created by the index sequential file method, the operating system performs processing according to the flowchart shown in FIG. When the user enters a deletion command and the name of the compressed file to be deleted (step S12), the operating system accesses the index of the specified compressed file (step S13) and determines whether the records are common or unique. Determine if there is (
Step S14). If the record is a unique part, the operating system deletes the record (unique part) (step S15).

On the other hand, if the record is a common part, the operating system accesses the common part control word of the common file and rewrites the number of shared files (step S16). In this case, reduce the number of shared files by one. Next, it is determined whether the number of shared files is 0 or not, and if the number of shared files is 0, the common part is deleted (this process is called zero determination deletion) (step S
17). When the processes from step S13 to step S17 are completed for all records owned by the compressed file specified by the user, the index is finally deleted (step S18), and the deletion of the compressed file is completed.

When deleting a compressed file using the marking method, processing is performed according to the flowchart of FIG. First, when the user inputs a deletion command and the name of the compressed file to be deleted (step S19)
, the operating system accesses the unique file header of the specified compressed file and reads information regarding the common file (step S20). next,
The number of shared files in the common file is decreased by one (step S21), and zero determination deletion is performed (step S22). Finally, the unique file is deleted (step S23), and the deletion of the compressed file is completed.

The procedure for updating a compressed file using the index sequential file method will be explained with reference to the flowchart shown in FIG. The user enters the command to edit a compressed file and the compressed file name (
Step S24), the compressed file is edited. (Step S25). When saving an edited compressed file, the operating system checks each record to see if changes have been made within it (step S
26). If no changes were made within a record, no action is taken on that record. However, if a change is made within the record, it is determined whether the record is a common part or a unique part (step S27), and if the record is a unique part, it is updated (step S27).
28).

On the other hand, if the changed record is a common part, the record is updated as a unique part (step S29). In other words, it is newly saved with the common part control word removed. Then, the information of the corresponding record in the index is rewritten to a value specifying the newly saved record (step S30). Furthermore, for the common file before being changed, the number of shared files in the common part control word is decreased by one (step S31), and zero determination deletion is performed (step S32). The processes from step S26 to step S32 are performed on all records owned by the compressed file, and the update of the compressed file is completed.

When updating a compressed file using the marking method, the changed file is saved as a normal file. Then, the number of shared files of the common file owned by the compressed file before the change is reduced by one, and zero determination deletion is performed. Therefore, once an update is made to a compressed file using the marking method, it is no longer a compressed file, but reverts to a regular file.

File compression processing for a compressed file is performed again after the compressed file is restored to a normal file. The process of converting a compressed file back to a normal file is performed using the following steps. In the case of compressed files processed using the index-sequential file method, records are recombined into one file according to the index. At this time, if the record is a common file, only the common parts excluding the common part control word are connected. In the case of a compressed file using the marking method, the common part specification word is replaced with the common part specified by it. Then, reduce the number of shared files owned by the compressed file by one,
Delete zero judgment. Finally, if the index sequential file method is used, the index is deleted, and if the marking method is used, the header portion is deleted. The above procedure returns the compressed file to a normal file, so the file compression process is performed on the normal file again.

Up to now, file compression processing has been explained on the assumption that there is only one common part to be processed, but even if there are multiple common parts, there is no problem in realizing file compression processing. It is not something that causes Therefore, a processing method when there are a plurality of common parts to be determined will be described.

[0042] File compression processing when, for example, k common parts are included in n files can basically be performed using the same procedure as shown in the flowchart in Fig. 2, but there are some differences. . If k common parts are determined as a result of comparing n files in step S2 of FIG. 2, step S4 is repeated k times, and common part processing is performed for each of the k determined common parts. . Next, post-comparison file processing (step S5) is performed on the n files, but in the case of the index sequential file method, k of the records are common parts, and in the case of the marking method, the unique k common part specification words are recorded in the file. Further, information on all k common parts is recorded in the index and header. This post-comparison file processing is repeated n times, and when all n files have been processed, the file compression processing ends.

The comparison process is not limited to all n files specified by the user, but may be performed between any files among them. Therefore, when performing the file compression process, the number i of files to be compared is changed and the comparison process and the common part process are performed in stages. That is, in stage i, a comparison is made between i files (i≦n), a common portion i for the i files is determined, and common portion processing is performed. The value of i is from n to m(n
≧m≧2).

The specific processing procedure will be explained with reference to the flowchart shown in FIG. When the user inputs a command for file compression processing and n file names (step S33), the operating system first performs a comparison process on the n files (step S34,
= n), the common part (common part n) of all files is determined (step S35), and if there is a common part n, common part processing is performed (step S36). Next, contents are compared between arbitrary n-1 files, excluding the common part n (step S34, i=n-1), and the common part n
-1, execute common part processing (step S3
6). However, the processing from step S34 to step S36 is repeated nCi times (the number of i combinations to be selected from n). In other words, when i is n-1, the number of repetitions is nCn-1 = n, and all combinations of n-
Comparison processing and common portion processing are performed between one file.

In this way, m pieces (n≧m
≧2) Processing is performed step by step, including comparison processing between files and common portion processing. Finally, for the n input files, post-comparison file processing is performed on all common parts determined in the files (step S37);
The file compression process ends. Below, auxiliary matters related to file compression processing will be described.

(1) If the size of the common portion determined as a result of comparison between files is very small, the effect of file compression processing will not be sufficiently obtained, so it is necessary to specify the lower limit value of the common portion size. be. Therefore, in this case, for example, the lower limit value is determined as the size of the common part control word.

(2) In common part processing, the information recorded in the common part control word of a common file is the number of shared files, but in addition, the names and storage locations of compressed files that share the common file are also recorded. may be recorded.

(3) In post-comparison file processing using the index sequential file method, if the record size is very large, it may be further divided into several records.

(4) In the post-comparison file processing using the marking method, the value recorded following the common part designation character in the common part designation word of the unique file may be any value that can designate the common file. Values indicating the start and end positions where the common file is saved may be recorded, or the name of the common file may be recorded. However, if the name of the common file is recorded in the common part specification word, the name of the common file must also be recorded in the header in addition to the values indicating the start and end positions of the common file.

(5) The operating system may notify the user of the processing status before completing the file compression processing. For example, when a user is performing an operation online, the processing status is displayed as a message on the terminal screen. That is, if a common part is determined, the name or number of the determined common part and a list of compressed file names that share the common part, and if a common part is not determined, the content that there is no common part. message may be displayed.

(6) In the file compression process (FIG. 11) in which common part determination and common part processing are performed in stages while reducing the number of files to be compared, the number m of files to be compared at the final stage is It may be implicitly specified, or the user may be able to specify it again.

(7) As a method for determining the common portion n between n files, there is, for example, the following procedure. n
The first and second files are numbered and the common parts are extracted by comparing the first and second files. Next, the extracted common portion is compared with the third file, and the common portion is similarly extracted. In this way, the files are compared one by one in order, and the common part among the last n-1 files is compared with the n-th file to determine the common part n. Therefore, when a common part n is determined, a comparison has already been performed and a common part has been extracted between the first to i-th (2≦i≦n-1) files. Here, these common parts extracted in the process of determining the common part n are called temporary common parts. By the way, file compression processing (Figure 11) when determining common parts and processing common parts is performed in stages.
), the number of files to be compared is changed in a decreasing direction, so at the stage of determining the common portion i, the common portions n to i+1 have already been determined. Therefore, when determining the common part n using the above method, some temporary common parts have already been extracted, so in order to reduce the number of steps in the file compression process, the extracted temporary common parts are It may be reused for determining the common part i.

(8) If there are several common parts within one file, file compression processing may be performed on this file alone. To achieve this, the user may input one file compression processing command and file name, and compare the contents within one file to determine the common part 1, or as shown in Figure 11 In the file compression process, the number m of files to be finally compared may be set to 1.

(9) The file compression method in the present invention is based on the conventional data compression technology (Huffman
(optimal encoding method, Ziv-Lempel data compression method, etc.), so there is no point in performing further conventional data compression on the compressed file after file compression processing. No problem. Furthermore, some modified examples of file compression processing will be explained.

[0055] As a special case of update processing for compressed files, updating of common parts can be considered. This is a process that rewrites the common part itself, and by performing this process, the contents of the common part of compressed files that include the common part are changed all at once. In order to update the common part, the common part control word may not be changed and the update may be performed as is. In this way, if you want to change the common parts of multiple files that contain the same content all at once, you can perform a process that combines file compression processing and updating the common parts (this is called batch file modification). It is very convenient to do.

Batch modification of files is performed according to the flowchart shown in FIG. First, the user inputs a command for batch modification of files and the names of n files desired to be modified at once (step S38). The operating system performs file compression processing on these files (step S39), and displays the name or number of the common part and a list of compressed file names that share it (step S40). Next, the user edits the common part to be updated (step S41), and the operating system updates the modified common part (step S42). Modification and updating of this common part can be repeated. When there are no common parts to be updated, the process ends.

It is also possible to use file compression processing as copy processing. The procedure of this process will be explained with reference to the flowchart shown in FIG. First, the user inputs a copy command (by file compression processing) and the copy source and copy destination file names (step S
43). The operating system regards the entire file to be copied as a common part (step S44), and records the number of shared files as 2 in common part processing (step S45). Furthermore, the post-comparison file processing (step S46) is performed twice to create copy source and copy destination compressed files.

In the case of post-comparison file processing using the index sequential file method, the process of step S6 in FIG. 4 is not necessary, and the information recorded in the index is only that of one record (common file) (step S
7). Then, in step S8, only the index is saved. Furthermore, in the case of post-comparison file processing using the marking method, processing is performed according to the flowchart shown in FIG. However, the unique file created by this post-comparison file processing consists only of a header and a common part specification word. By performing the above processing, only one common part is saved in the secondary storage device, and this is shared between the copy source and copy destination compressed files.

Note that the present invention is not limited to the embodiments described above. In the embodiment, a magnetic disk is used as a storage unit for storing files, but a secondary storage device (external storage device) such as a magnetic tape or an optical disk can be used instead. Further, the configuration of the file compression processing section shown in FIG. 1 may be realized by hardware or software. In addition, various modifications can be made without departing from the gist of the present invention.

[0060]

[Effects of the Invention] As detailed above, according to the present invention, when there are multiple files including portions with common contents, each file is divided into the common portion and the non-common portion excluding the common portion. , these are stored independently in a storage unit such as a secondary storage device. Therefore, it is possible to realize a file compression device that can avoid inconveniences such as redundant storage of contents common to a plurality of files in a secondary storage device, and can make more efficient use of the secondary storage device. becomes possible.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing a schematic configuration of a file compression device according to an embodiment of the present invention;

[Figure 2] Flowchart showing the basic steps of file compression processing,

[Figure 3] Schematic diagram showing the structure of a common file,

FIG. 4 is a flowchart showing post-comparison file processing using the index sequential file method;

[Fig. 5] A schematic diagram showing the structure of a compressed file created by file compression processing using the index sequential file method.

FIG. 6 is a flowchart showing post-comparison file processing using the marking method;

[Fig. 7] A schematic diagram showing the structure of a unique file of a compressed file created by file compression processing using a marking method.

FIG. 8 is a flowchart showing a procedure for deleting a compressed file processed by the index sequential file method;

[
FIG. 9 is a flowchart showing the procedure for deleting compressed files processed by the marking method;

FIG. 10 is a flowchart showing a procedure for updating a compressed file processed by the index sequential file method;

FIG. 11 is a flowchart showing the procedure of file compression processing when the number of files to be compared is changed and comparison processing and common part processing are performed in stages;

[Fig. 12] Flowchart showing the procedure for batch modification of files,

FIG. 13 is a flowchart showing a procedure when file compression processing is performed as copy processing.

[Explanation of symbols]

10...File compression processing section, 11...Comparison processing section, 12...Common part processing section, 13...File processing section, 20...Secondary storage device (first storage section), 30...Secondary storage device (second storage section) memory).

Claims (1)

[Claims] Claim 1: means for comparing the contents of a plurality of files stored in a first storage unit; means for extracting portions whose contents match as a result of the comparison by the means; means for storing a common portion as a common file in a first storage unit or a second storage unit; and a means for storing a file after extracting the common portion as a unique file in a first storage unit.
1. A file compression device comprising: a storage unit or a second storage unit.