JP5966184B2 - Management system - Google Patents

Description

  The present invention relates to a technique for managing a file by dividing it into a plurality of data.

In recent years, a deduplication technique has been proposed as one of data backup methods (see Patent Document 1). The deduplication technology is a technology that can efficiently use storage capacity by detecting duplication of block data in the storage and eliminating one of the duplicated block data.
By using such a technique, the data center can eliminate the accumulation of block data storing the same information in the storage. Furthermore, the data center can reduce the amount of data stored in the storage.

Data centers perform deduplication of block data between data centers by cooperating with other data centers via a network (hereinafter referred to as “global deduplication processing”). A plurality of data centers perform global deduplication processing, thereby eliminating duplication of block data between data centers and making it possible to efficiently use storage throughout the data centers.
When such a data center is realized, the user acquires desired data by performing restoration. Specifically, the user obtains block data necessary for restoring a desired file from each data center, and restores the desired file by performing a restoration process on the obtained block data.

JP 2009-205201 A

  However, when global deduplication processing is performed, the deduplicated block data becomes the only block data in the entire data center. When performing the restore, the user must go to the data center that holds the block data necessary for restoring the file to obtain the necessary block data. For this reason, there is a problem that the restoration time is significantly increased due to the influence of the delay in the block data transfer at the time of restoration.

  In view of the above circumstances, an object of the present invention is to provide a technique for shortening the restoration time.

  According to one embodiment of the present invention, one piece of data composed of a plurality of pieces of block data can be restored at a transmission destination by transmitting the plurality of pieces of block data in response to a request. Can be generated by combining a plurality of partial block data, which is a part of other block data, and is requested by a user from a plurality of data centers respectively installed at a plurality of points and storing one or more block data. A management system in a communication system for transmitting the block data or the partial block data necessary for generating data, and a combination of the plurality of partial block data capable of generating the block data for each block data A block list storage unit for storing the block data and the block data In addition, a processing time calculation unit that calculates a processing time required for generating the block data by combining the plurality of partial block data for each combination stored in the block list storage unit, and for each data The acquisition method for each block data is determined based on the processing time required to receive the block data constituting the data from the data center and the processing time calculated by the processing time calculation unit. And a determination unit.

One aspect of the present invention is the above-described management system, the processing time calculation unit, a communication time required to receive a plurality of the partial block data required to generate the block data, the The processing time is calculated based on the sum of the generation processing time required to generate the block data by combining a plurality of partial block data.

  One aspect of the present invention is the management system described above, wherein the determination unit calculates, for each block data, a processing time required to receive the block data from the data center and the processing time calculation unit. A candidate selection unit that compares the obtained processing times with each other and selects an acquisition method in the shortest processing time as a candidate, and the longest processing time among the processing times of the candidates of the plurality of block data constituting the data A longest time determination unit for determining the longest time candidate as the longest candidate, and for each candidate other than the longest candidate, the generation processing time within a range not exceeding the longest time A final candidate determination unit that determines the acquisition method that is the shortest as a final candidate, and a combination of the longest candidate and the final candidate when acquiring the data And an output unit for outputting as the processing method, the.

  According to the present invention, the restoration time can be shortened.

It is a figure showing the relationship of the data which comprise a file. It is a figure which shows the system configuration | structure of the management system in this embodiment. 2 is a schematic block diagram showing functional configurations of a data center 10 and a central management device 20. FIG. It is a figure which shows the specific example of a partial block list and a block arrangement list. It is a block diagram of the block data list regarding the block data “A”. FIG. 4 is a diagram illustrating a block data list for each block data constituting the file 1. FIG. 4 is a diagram illustrating a block data list for each block data constituting the file 1. It is a flowchart which shows the flow of a process of index list creation of this embodiment. It is a flowchart which shows the flow of a process of index list creation of this embodiment. It is a flowchart which shows the flow of a restore process of this embodiment.

Hereinafter, an embodiment of the present invention will be described with reference to the drawings.
FIG. 1 is a diagram showing the relationship of data constituting a file. FIG. 1A is a diagram illustrating a configuration example of a file 1 backed up by a user. File 1 is composed of a plurality of block data. Block data is data required to restore a file. In FIG. 1A, for example, the block data constituting the file 1 are “A”, “B”, and “C”.
FIG. 1B is a diagram illustrating a configuration example of block data. Block data can be generated using a plurality of partial block data. Partial block data is data generated by performing a division process on block data. In other words, the partial block data is data configured by cutting out a part of the block data.

In the example of FIG. 1B, the block data “E” is represented by a binary number “110111”. The block data “F” is represented by a binary number “001101”. The block data “E” is divided into partial block data “E1” and partial block data “E2” by performing a division process. The partial block data “E1” is represented by a binary number “110”. The partial block data “E2” is represented by a binary number “111”.
The block data “F” is divided into partial block data “F1” and partial block data “F2” by performing a division process. The partial block data “F1” is represented by a binary number “001”. The partial block data “F2” is represented by a binary number “101”.

  The block data “A” is represented by a binary number “110101”. The block data “A” is divided into partial block data “A1” and partial block data “A2” by performing the division process. The partial block data “A1” is represented by a binary number “110”. The partial block data “A2” is represented by a binary number “101”. The partial block data “A1” and the partial block data “E1” are represented by the same binary number. Further, the partial block data “A2” and the partial block data “F2” are represented by the same binary number. That is, as shown in FIG. 1B, the block data “A” can be generated by a combination of the partial block data “E1” and the partial block data “F2”.

FIG. 2 is a diagram showing a system configuration of the management system in the present embodiment. The management system of this embodiment includes a plurality of data centers 10, a central management device 20, a plurality of PCs 30, and a network 40.
The data center 10 stores a file backed up from the PC 30 via the network 40 (hereinafter referred to as “backup file”).
The central management device 20 is configured using an information processing device. The central management device 20 manages block data transferred from each data center 10.

The PC 30 backs up files to the data center 10 via the network 40. Further, the PC 30 receives data from the data center 10 via the network 40 and restores a file desired by the user.
The network 40 may be a network configured in any way. For example, it may be configured using the Internet.

FIG. 3 is a schematic block diagram showing the functional configuration of the data center 10 and the central management device 20. First, the functional configuration of the data center 10 will be described.
The data center 10 includes a CPU (Central Processing Unit), a memory, an auxiliary storage device, and the like connected by a bus, and executes a data center program. By executing the data center program, the data center 10 has a communication unit 101, a configuration list creation unit 102, a restore data storage unit 103, a deduplication unit 104, a block data storage unit 105, a block list storage unit 106, and a processing time calculation unit 107. , Functions as a device including the determination unit 108. The determination unit 108 functions as a candidate selection unit 109, a longest time determination unit 110, a final candidate determination unit 111, an output unit 112, an index list storage unit 113, and a partial block data generation unit 114. All or some of the functions of the data center 10 may be realized using hardware such as an application specific integrated circuit (ASIC), a programmable logic device (PLD), and a field programmable gate array (FPGA). The data center program may be recorded on a computer-readable recording medium. The computer-readable recording medium is, for example, a portable medium such as a flexible disk, a magneto-optical disk, a ROM, a CD-ROM, or a storage device such as a hard disk built in the computer system. The data center program may be transmitted / received via a telecommunication line.

The communication unit 101 transmits / receives data to / from other data centers 10, the central management device 20, and the PC 30.
The configuration list creation unit 102 creates a restore data configuration list. The restore data configuration list is a list in which block data information necessary for restoration is defined for each backup file.
The restore data storage unit 103 is configured using a storage device such as a magnetic hard disk device or a semiconductor storage device. For example, the semiconductor memory device is an SSD (Solid State Drive). The restore data storage unit 103 stores the restore data configuration list created by the configuration list creation unit 102.

The deduplication unit 104 performs global deduplication processing between the plurality of data centers 10. The deduplication unit 104 selects block data that does not exist in other data centers 10 by performing global deduplication processing.
The block data storage unit 105 is configured using a storage device such as a magnetic hard disk device or a semiconductor storage device. The block data storage unit 105 stores block data that does not exist in other data centers 10.

  The block list storage unit 106 is configured using a storage device such as a magnetic hard disk device or a semiconductor storage device. For example, the semiconductor memory device is an SSD. The block list storage unit 106 stores a partial block list and a block arrangement list. The partial block list is a list in which combinations of partial block data capable of generating block data for each block data (hereinafter referred to as “partial block group”) are described. The partial block group may include a partial block group that directly acquires block data from the data center 10. The block arrangement list is a list in which information on each partial block data is described. For example, the block arrangement list may describe the block size of the partial block data and the arrangement information of the partial block data (location of the data center 10 storing the partial block data).

  The processing time calculation unit 107 calculates the total restoration time (processing time) based on the partial block list and the block arrangement list stored in the block list storage unit 106. The total restore time is the total processing time required to restore a file. The processing time calculation unit 107 calculates the total restoration time for each partial block group using the partial block list and the block arrangement list.

The determination unit 108 determines one partial block group for each block data constituting the file. Hereinafter, a specific configuration of the determination unit 108 will be described.
The candidate selection unit 109 selects a candidate partial block group for each block data constituting the file. The candidate partial block group is a partial block group having the shortest total restoration time among the partial block groups.
The longest time determination unit 110 determines a partial block group (hereinafter referred to as “longest partial block group”) having the longest total restoration time from the candidate partial block groups selected by the candidate selection unit 109. The longest time determination unit 110 acquires the total restore time of the longest partial block group as the longest restore time.

  The final candidate determination unit 111 first acquires a total restore time (hereinafter referred to as “direct restore time”) required for acquiring block data directly from the data center 10 for each candidate partial block group. Then, the final candidate determination unit 111 compares the longest restore time with the direct restore time for each candidate partial block group other than the longest partial block group.

When the direct restore time is shorter than the longest restore time, the final candidate determination unit 111 sets a partial block group (hereinafter referred to as “direct restore group”) that directly acquires block data from the data center 10 as a final partial block group. To decide.
On the other hand, when the direct restore time is longer than the longest restore time, the final candidate determining unit 111 determines the candidate partial block group as the final partial block group. The final candidate determination unit 111 determines the final partial block group for each block data by performing such processing for each block data.

  The output unit 112 generates an index list for each file based on the longest partial block group and the final partial block group. The index list is a list in which combinations of partial block data that minimize the total restoration time are described. The output unit 112 outputs the generated index list to the index list storage unit 113. The output unit 112 records the head pointer information and the number of bytes of each partial block data in the index list when generating the index list.

The index list storage unit 113 is configured using a storage device such as a magnetic hard disk device or a semiconductor storage device. For example, the semiconductor memory device is an SSD. The index list storage unit 113 stores the index list output by the output unit 112. The index list storage unit 113 adds and stores an index list each time the output unit 112 outputs the index list.
The partial block data generation unit 114 performs block data division processing by referring to the index list stored in the index list storage unit 113, and generates partial block data. For example, the partial block data generation unit 114 acquires data corresponding to the number of bytes recorded in the index list as partial block data from the address indicated by the head pointer information.

  Next, the functional configuration of the central management device 20 will be described. The central management device 20 includes a CPU, a memory, an auxiliary storage device, and the like connected by a bus, and executes a management program. By executing the management program, the central management device 20 functions as a device including a communication unit 201, a block data storage unit 202, a division processing unit 203, a detection unit 204, and an update unit 205. Note that all or some of the functions of the central management device 20 may be realized using hardware such as an ASIC, PLD, or FPGA. The management program may be recorded on a computer-readable recording medium. The computer-readable recording medium is, for example, a portable medium such as a flexible disk, a magneto-optical disk, a ROM, a CD-ROM, or a storage device such as a hard disk built in the computer system. Further, the management program may be transmitted / received via a telecommunication line.

The communication unit 201 transmits / receives data to / from a plurality of data centers 10.
The block data storage unit 202 is configured using a storage device such as a magnetic hard disk device or a semiconductor storage device. For example, the semiconductor memory device is an SSD. The block data storage unit 202 stores block data received from the plurality of data centers 10 by the communication unit 201.
The division processing unit 203 performs a division process on the block data stored in the block data storage unit 202. The division process is a process of subdividing one block data into a plurality of partial block data. The division processing unit 203 generates partial block data by dividing the block data.
The detection unit 204 detects duplication of the generated partial block data.
The update unit 205 updates the partial block list and the block arrangement list according to the detection result by the detection unit 204.

FIG. 4 is a diagram illustrating specific examples of the partial block list and the block arrangement list.
FIG. 4A is a diagram illustrating a specific example of the partial block list. The partial block list has a plurality of records 50. The record 50 has block data and partial block group values.
The value of the block data represents identification information of the block data.

The value of the partial block group represents identification information of partial block data that can generate block data. In FIG. 4, the number of partial block data included in the partial block group is two, but may be one or three or more.
In FIG. 4A, for example, block data “A” can be generated by a combination of partial block data “E1” and partial block data “F2”. The block data “A” can be generated by a combination of the partial block data “D1” and the partial block data “J2”.

FIG. 4B is a diagram illustrating a specific example of the block arrangement list. The block arrangement list has a plurality of records 51. The record 51 has partial block data, block size, and placement destination values.
The value of the partial block data represents identification information of the partial block data. The block size represents the size of the partial block data capacity. The block size may be expressed in units of B (bytes), for example. The placement destination represents placement information of partial block data (location of the data center 10 storing the partial block data).
In FIG. 4B, for example, the partial block data “E1” has a size of 100 MB (megabytes) and is stored in the data center 10 of “Ehime”.

  A specific method for creating an index list will be described with reference to FIGS. In the following description, a case where an index list of file 1 is created will be described for the sake of simplicity. The file 1 is composed of, for example, block data “A”, block data “B”, and block data “C”.

FIG. 5 is a configuration diagram of a block data list regarding the block data “A”.
The block data list has a plurality of records 52 representing restore information. The record 52 has values of a block data partial block group, a CPU processing time (generation processing time), a transfer time (communication time), and a total restore time.
The value of the block data represents identification information of the block data.
The value of the partial block group represents identification information of partial block data included in each partial block group.

  The value of the CPU processing time represents the generation processing time required for generating one block data. Note that the CPU processing time of the record 52 representing the direct restore group (the uppermost record 52 in FIG. 5) is “0” because it is not necessary to combine partial block data. The CPU processing time may be held as a table for each number of partial block data. In this case, the processing time calculation unit 107 calculates the CPU processing time according to the number of partial block data included in the partial block group by referring to the table.

The transfer time value represents the time required to transfer the block data and partial block data necessary for restoration from each data center 10 to the restoration request source (PC 30). For example, the transfer time is obtained by dividing the block size of block data and partial block data required for restoration based on the block arrangement list by the delay time required to receive the block data and partial block data from the data center 10. It is calculated by.
The value of the total restore time represents the total restore time of each partial block group. Specifically, this is the total time calculated by the sum of the CPU processing time and the transfer time for each record 52.

FIG. 6 is a diagram showing a block data list for each block data constituting the file 1.
FIG. 6A is a diagram illustrating a specific example of a block data list regarding the block data “A”. FIG. 6B is a diagram illustrating a specific example of a block data list regarding the block data “B”. FIG. 6C is a diagram illustrating a specific example of a block data list regarding the block data “C”.

  The candidate selection unit 109 selects a candidate partial block group for each block data constituting the file. In FIG. 6A, the total restoration time of the partial block group composed of the partial block data “E1” and “F2” is “10”, and is the most total restoration among the partial block groups related to the block data “A”. The time is short. Therefore, the candidate selection unit 109 selects a partial block group composed of the partial block data “E1” and “F2” as a candidate partial block group of the block data “A” (thick frame in FIG. 6A).

  In FIG. 6B, the total restore time (direct restore time) of the direct restore group composed only of the block data “B” is “14”, which is the most total among the partial block groups related to the block data “B”. Therefore, the restore time is short, so that the candidate selection unit 109 selects a direct restore group including only block data “B” as a candidate partial block group of block data “B” (thick frame in FIG. 6B). .

  In FIG. 6C, the total restoration time of the partial block group composed of the partial block data “P2” and “O2” is “11”, which is the most total restoration among the partial block groups related to the block data “C”. The time is short. Therefore, the candidate selection unit 109 selects a partial block group composed of the partial block data “P2” and “O2” as a candidate partial block group of the block data “C” (thick frame in FIG. 6C).

  Next, the longest time determination unit 110 determines a partial block group having the longest total restore time from candidate partial block groups for each block data list. In FIG. 6, the total restoration time of the candidate partial block group of the block data “B” is the longest. Therefore, the longest time determination unit 110 determines the total restore time of the block data “B” as the longest restore time of the file 1. Further, the longest time determination unit 110 determines the partial block group having the longest restore time as the longest partial block group.

FIG. 7 is a diagram illustrating a block data list for each block data constituting the file 1.
In FIG. 7, the final candidate determination unit 111 determines a final partial block group for each candidate partial block group other than the longest partial block group. The final candidate determination unit 111 compares the longest restore time with the direct restore time for each candidate partial block group. When the direct restore time is shorter than the longest restore time, the final candidate determining unit 111 determines the direct restore group as the final partial block group. On the other hand, when the direct restore time is longer than the longest restore time, the final candidate determination unit 111 determines the candidate partial block group as the final partial block group.

  In FIG. 7A, the final candidate determination unit 111 compares the direct restore time of the block data “A” with the longest restore time. The direct restore time of the block data “A” is “12”, and the longest restore time is “14”. That is, the direct restore time of the block data “A” is shorter than the longest restore time. Therefore, the final candidate determination unit 111 determines the direct restore group as the final partial block group of the block data “A”.

In FIG. 7C, the final candidate determination unit 111 compares the direct restore time of the block data “C” with the longest restore time. The direct restore time of the block data “C” is “15”, and the longest restore time is “14”. That is, the direct restore time of the block data “C” is longer than the longest restore time. Therefore, final candidate determination section 111 determines the candidate partial block group as the final partial block group of block data “C”.
The output unit 112 generates an index list of the file 1 based on the longest partial block group and the final partial block group.

8 and 9 are flowcharts showing the flow of processing for creating an index list according to this embodiment.
The communication unit 101 receives backup files from the plurality of PCs 30 (step S101). The configuration list creation unit 102 creates a restore data configuration list. The configuration list creation unit 102 describes backup files from a plurality of PCs 30 in the restore data configuration list in units of files (step S102). The deduplication unit 104 eliminates duplication of block data between the data centers 10 by performing deduplication processing (step S103). Each data center 10 transfers the block data stored in its own device (data center 10) to the central management device 20 (step S104).

  The communication unit 201 of the central management device 20 receives the block data transferred from each data center 10. The division processing unit 203 performs a division process on the received block data. The division processing unit 203 generates partial block data that can generate block data. The detection unit 204 detects the overlap of the partial block data generated by the division processing unit 203 (step S105).

  The update unit 205 updates the partial block list according to the detection result by the detection unit 204 (step S106). The update unit 205 updates the block arrangement list according to the detection result by the detection unit 204 (step S107). The communication unit 201 transfers the partial block list and the block arrangement list to each data center 10 (step S108).

  The communication unit 101 of the data center 10 receives the partial block list and the block arrangement list transferred from the central management device 20. The communication unit 101 records the received partial block list and block arrangement list in the block list storage unit 106. The processing time calculation unit 107 calculates the total restoration time of the partial block group for each partial block data constituting the file based on the partial block list and the block arrangement list stored in the block list storage unit 106 (step S109). ).

The candidate selection unit 109 selects a partial block group (candidate partial block group) having the shortest total restoration time from the partial block groups for each block data constituting the file (step S110).
The longest time determination unit 110 determines the longest partial block group having the longest total restoration time from the candidate partial block groups. The longest time determination unit 110 acquires the total restore time of the longest partial block group as the longest restore time (step S111).

  The final candidate determination unit 111 compares the direct restore time with the direct restore time for each candidate partial block group other than the longest partial block group (step S112). If the direct restore time is longer than the longest restore time (step S112—YES), the final candidate determining unit 111 determines the candidate partial block group as the final partial block group (step S113).

On the other hand, when the direct restore time is shorter than the longest restore time in the process of step S112 (step S112—NO), the final candidate determination unit 111 determines the direct restore group as the final partial block group (step S114).
The output unit 112 generates an index list based on the final partial block group and the longest partial block group (step S115). The output unit 112 outputs the generated index list to the index list storage unit 113. The index list storage unit 113 adds and stores an index list every time the output unit 112 outputs the index list (step S116).

FIG. 10 is a flowchart showing the flow of restoration processing according to this embodiment.
The communication unit 101 of the data center 10 receives the user restore request transmitted from the PC 30 via the network 40 (step S201). The communication unit 101 refers to the index list according to the received restore request, thereby determining the block data and partial block data constituting the file requested to be restored and the arrangement destination thereof (step S202). The communication unit 101 reads block data from the block data storage unit 105 for data stored in its own device (data center 10) among the block data and partial block data constituting the file requested to be restored. The communication unit 101 refers to the index list to transmit block data constituting the file requested to be restored to the request source (PC 30).

  Further, the partial block data generation unit 114 refers to the head pointer information and the number of bytes stored in the index list. The partial block data generation unit 114 performs block data division processing based on the head pointer information and the number of bytes, and generates partial block data. The communication unit 101 transmits the generated partial block data to the request source (PC 30). Of the block data and partial block data constituting the file requested to be restored, the communication unit 101 makes a request to the data center 10 of the arrangement destination for data not stored in its own device (data center 10). A request is made to transfer block data or partial block data to the source (PC 30) (step S203).

  According to the management system configured as described above, each data center 10 uses the data transfer time between its own device and each data center 10 to minimize the time required for restoration for each file. Create an index list. For this reason, it is possible to reduce the time required for restoration of a file requested to be restored by the user.

  For block data that can be restored in a time shorter than the longest restoration time, an index list is generated so that the block data is directly transferred without combining partial block data. In general, processing for combining partial block data to generate block data is performed by a user device (PC 30). Therefore, the processing load of the CPU in the user device (PC 30) can be reduced by reducing the number of times of combining.

<Modification>
The block arrangement list record 51 may have information other than the block size and the arrangement destination.
The processes in steps S109 to S115 described above may be performed in a place other than the data center 10. For example, the central management device 20 may create an index list and transfer the index list to each data center 10. The communication unit 101 of each data center 10 records the transferred index list in the index list storage unit 113.
One data center 10 among the plurality of data centers 10 may be configured to function as the central management device 20.
The device operated by the user need not be limited to the PC 30 described above. For example, instead of a PC (personal computer), any device may be used as long as it is a communication terminal device such as a smartphone, a mobile phone, a game device, a television receiver, and a video recording device.

  The embodiment of the present invention has been described in detail with reference to the drawings. However, the specific configuration is not limited to this embodiment, and includes designs and the like that do not depart from the gist of the present invention.

DESCRIPTION OF SYMBOLS 10 ... Data center, 20 ... Central management apparatus, 30 ... PC, 40 ... Network, 101 ... Communication part, 102 ... Configuration list creation part, 103 ... Restore data storage part, 104 ... Deduplication part, 105 ... Block data storage part 106: Block list storage unit, 107: Processing time calculation unit, 108 ... Determination unit, 109 ... Candidate selection unit, 110 ... Longest time determination unit, 111 ... Final candidate determination unit, 112 ... Output unit, 113 ... Index list storage , 114 ... Partial block data generation unit, 201 ... Communication unit, 202 ... Block data storage unit, 203 ... Division processing unit, 204 ... Detection unit, 205 ... Update unit

Claims (3)

One data composed of a plurality of block data can be restored at a transmission destination by transmitting the plurality of block data in response to a request, and the block data includes other block data It can be generated by combining a plurality of partial block data, which is a part, and is necessary for generating data requested by a user from a plurality of data centers installed at a plurality of points and storing one or more of the block data. A management system in a communication system for transmitting the block data or the partial block data,
For each block data, a block list storage unit that stores a combination of the plurality of partial block data capable of generating the block data;
For each block data, a processing time calculation unit that calculates a processing time required for generating the block data by combining the plurality of partial block data for each combination stored in the block list storage unit;
Acquisition for each block data based on the processing time required to receive the block data constituting the data from the data center and the processing time calculated by the processing time calculation unit for each data A determination unit for determining a method;
A management system comprising: The processing time calculation unit, a communication time required to receive a plurality of the partial block data required to generate the block data, to generate the block data by combining said plurality of partial block data The management system according to claim 1, wherein the processing time is calculated based on a sum of a generation processing time required for the processing. The determination unit
For each block data, the processing time required to receive the block data from the data center is compared with each processing time calculated by the processing time calculation unit, and an acquisition method in the shortest processing time is selected. A candidate selector to select as,
A longest time determination unit that determines the longest processing time among the processing times of each candidate of a plurality of block data constituting the data as the longest time, and determines the longest time candidate as the longest candidate;
For each candidate other than the longest candidate, a final candidate determining unit that determines, as a final candidate, the acquisition method in which the generation processing time is the shortest in a range in which the processing time does not exceed the longest time;
An output unit that outputs a combination of the longest candidate and the final candidate as a processing method when acquiring the data;
The management system according to claim 2, comprising:

Images