Backup management program, backup management method, and information processing apparatus

Abstract

To optimize resources used for backup.SOLUTION: An information processing device 101, during transferring data to be backed up to a backup storage 103 while being divided into plural times, acquires a first data amount transmitted to the backup storage 103, among data previously assigned to each data transfer unit 102, and a second data amount not transmitted to the backup storage 103 due to a de-duplication function from each data transfer unit 102. The information processing device 101 calculates the last transmit data amount transmitted to the backup storage 103, on the basis of the first data amount. The information processing device 101 calculates a next time transmit data amount to be transmitted to the backup storage, on the basis of the first data amount, the second data amount, and the size of the data to be backed up. The information processing device 101 increases / decreases the number of the plurality of data transfer devices 102, on the basis of the result obtained by comparing the last transmit data amount with the next time transmit data amount.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a backup management program, a backup management method, and an information processing apparatus.

Background Art

[0002] In recent years, the use of the cloud has advanced, and object storage on the cloud is often used as a secondary storage destination for backed-up data. Daily backups are performed, for example, by calculating the required time based on the backup data size and network bandwidth and securing the backup execution time for that amount of time. On the other hand, since the line speed of the network may vary depending on the time of day, there are cases where the backup may not be completed within the time limit due to a decrease in the line speed. To prevent this situation, a backup method that uses multiple lines for transferring backup data has been devised.

[0003] As prior art, based on information regarding the bandwidth of each network path and the maximum multiplicity and minimum bandwidth of the network paths to be used, those with a bandwidth greater than the minimum bandwidth within the range of the maximum multiplicity or less among each network path are determined as the network paths to be used, and backup data is assigned to the network paths, and the backup data is transmitted and received between the client and the server. There is also a technology that acquires bandwidth information indicating the bandwidth status of the network, determines a communication path for transmitting and receiving data to be backed up based on the bandwidth information, executes a backup of the data received from the client terminal via the determined communication path, and based on the bandwidth information, executes interruption and resumption of the data backup.

[0004] There is also a technique for determining whether or not to perform deduplication processing on block data obtained by dividing the data received from an external device based on network load information indicating the load status of the network with a storage device having a deduplication function. There is also a technique for determining the number of divisions of the received data, determining a data center that transfers each divided data with reference to transfer order information in which the priority order of transferring file replicas is determined, dividing a file for which writing is requested into a determined number of data, and transmitting the divided data to the determined data centers in parallel. There is also a technique in which when a backup GW device receives a packet addressed to a communication device, it attaches a reception time and transmits it to a backup server, and when the backup server receives a packet with a reception time, it advances the clock of emulation by a virtual machine that is a replica of the communication device until it matches the reception time of the packet received one before, and stops the clock when they match.

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Patent Document 2

Patent Document 3

Patent Document 4

Patent Document 5

Summary of the Invention

Problems to be Solved by the Invention

[0006] However, in the prior art, when performing data backup using a plurality of communication lines, it is difficult to control the resources used for backup so that the backup is completed within the assumed time.

[0007] On one side, the present invention aims to optimize the resources used for backup.

Means for Solving the Problem

[0008] In one embodiment, during the transfer of backup target data to a backup storage by dispersing it multiple times via network lines corresponding to each data transfer device of a plurality of data transfer devices, among the data previously assigned to each data transfer device, the first data amount transmitted to the backup storage and the second data amount not transmitted to the backup storage due to the deduplication function are acquired from each data transfer device. Based on the acquired first data amount, the previous transmission data amount transmitted to the backup storage by the plurality of data transfer devices is calculated. Based on the acquired first data amount, the second data amount, and the size of the backup target data, the next transmission data amount to be transmitted to the backup storage is calculated. Based on the result of comparing the calculated previous transmission data amount and the next transmission data amount, a backup management program is provided that increases or decreases the number of the plurality of data transfer devices.

Effect of the Invention

[0009] According to one aspect of the present invention, there is an effect that the resources used for backup can be optimized.

Brief Description of the Drawings

[0010]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Figure 17

[0011] Hereinafter, embodiments of a backup management program, a backup management method, and an information processing apparatus according to the present invention will be described in detail with reference to the drawings.

[0012] (Embodiment) FIG. 1 is an explanatory diagram showing an example of a backup management method according to an embodiment. In FIG. 1, an information processing apparatus 101 is a computer that controls an increase or decrease in resources used for backup when performing a backup of backup target data. The backup target data is, for example, client business data.

[0013] Here, a backup storage on the cloud may be used as a secondary storage destination for the backed-up data. The backup operation is scheduled, and the start time and end time are often determined. The backup operation is required to be completed by the end time, but it is not required to end earlier than the end time.

[0014] On the other hand, if the bandwidth of the network line used for backup fluctuates so as to narrow from the start time to the end time, the backup may not be completed within the assumed time (the scheduled backup window). For this reason, there is a backup method in which the number of network lines for transferring backup data is increased and the backup data is equally divided and transferred.

[0015] However, in this method, among the multiple network lines, the line with a wide bandwidth completes the backup in a short time, but the line with a narrow bandwidth takes longer. For this reason, the overall backup time is dragged by the time of the line with a narrow bandwidth and extended, and the backup may not end within the assumed time.

[0016] Also, in the method of simply dividing the backup data equally, waste may occur in the resources used (network lines and servers for transferring backup data) due to fluctuations in bandwidth, duplicate elimination processing with the data already stored in the backup storage at the transfer destination, and fluctuations in data size due to compression processing.

[0017] Therefore, in this embodiment, a backup management method for optimizing the resources used for backup will be described so that the backup can be completed within the assumed time.

[0018] In FIG. 1, the information processing apparatus 101 transfers backup target data to the backup storage 103 in multiple dispersions via network lines corresponding to each of the plurality of data transfer apparatuses 102. The data transfer apparatus 102 transfers the data assigned from the information processing apparatus 101 among the backup target data to the backup storage 103. In FIG. 1, the plurality of data transfer apparatuses 102 are denoted as "data transfer apparatuses 102-1 to 102-3".

[0019] The backup storage 103 is a computer that serves as the storage destination for the backup target data. The network line corresponding to the data transfer apparatus 102 is a network line that can be used by the data transfer apparatus 102. This network line is not occupied by the data transfer apparatus 102 and may be used for other purposes. For this reason, it is not always possible to secure the bandwidth as planned.

[0020] The backup target data is transferred to the backup storage 103 in multiple dispersions. In one transfer, the data assigned to each data transfer apparatus 102 is transferred to the backup storage 103 respectively. For example, the assumed time until the backup is completed and the number of transmissions indicating how many times the backup target data is divided for transfer are specified in advance.

[0021] The time allowed for one transfer is set based on, for example, the assumed time until the backup is completed and the number of transmissions. For this reason, depending on the status of the network line, in one transfer, the data transfer apparatus 102 may not be able to send all the data assigned to itself. Hereinafter, a processing example of the information processing apparatus 101 executed during the transfer of the backup target data will be described.

[0022] (1) While transferring backup target data to the backup storage 103 in multiple batches, the information processing apparatus 101 acquires from each data transfer apparatus 102 the amount of first data transmitted to the backup storage 103 and the amount of second data not transmitted to the backup storage 103 due to the deduplication function, among the data assigned to each data transfer apparatus 102 in the previous round.

[0023] Here, the deduplication function is a function that analyzes data, detects and eliminates duplicate data. Duplicate data is, for example, data already stored in the backup storage 103 among the backup target data. The data transfer apparatus 102 does not transmit to the backup storage 103 the duplicate data that has already been stored in the backup storage 103 among the data assigned to its own apparatus.

[0024] In the example of FIG. 1, it is assumed that the backup target data 110 is divided into P (number of transmissions) and transferred to the backup storage 103. Among the backup target data 110, the data transferred in the first round is referred to as "data 111". The information processing apparatus 101 divides the data 111 and assigns it to each of the data transfer apparatuses 102-1 to 102-3.

[0025] For example, data 111-1 divided from the data 111 is assigned to the data transfer apparatus 102-1. Data 111-2 divided from the data 111 is assigned to the data transfer apparatus 102-2. Data 111-3 divided from the data 111 is assigned to the data transfer apparatus 102-3.

[0026] Each of the data transfer apparatuses 102-1 to 102-3 performs a duplicate check and does not transmit to the backup storage 103 the duplicate data that has already been stored in the backup storage 103 among the data 111-1 to 111-3 assigned to its own apparatus. For example, the data transfer apparatus 102-1 does not transmit to the backup storage 103 the duplicate data 111-1-1 that has already been stored in the backup storage 103 among the data 111-1.

[0027] Therefore, the first data volume of the data transfer device 102-1 is the cumulative data volume of the data transmitted within the allowable time among the data 111-1'. The data 111-1' is the remaining data of the data 111-1 excluding the duplicate data 111-1-1. Also, the second data volume of the data transfer device 102-1 is the data volume of the data 111-1-1.

[0028] (2) The information processing device 101 calculates the previous transmission data volume transmitted to the backup storage 103 by the plurality of data transfer devices 102 based on the acquired first data volume. Specifically, for example, the information processing device 101 calculates the previous transmission data volume transmitted to the backup storage 103 by the plurality of data transfer devices 102 by adding up the first data volumes acquired from each data transfer device 102.

[0029] Here, since the first data volume of each data transfer device 102 changes depending on the situation of the network line corresponding to each data transfer device 102, for example, it can be said that it is one of the indicators for representing the transfer speed according to the situation of the network line. Also, the previous transmission data volume can be said to be one of the indicators for representing the transfer speed of the entire plurality of data transfer devices 102.

[0030] In the example of FIG. 1, the information processing device 101 calculates the previous transmission data volume D1 transmitted to the backup storage 103 by the data transfer devices 102-1 to 102-3 by adding up the first data volumes acquired from the data transfer devices 102-1 to 102-3.

[0031] (3) The information processing apparatus 101 calculates the next transmission data amount to be transmitted to the backup storage 103 based on the acquired first data amount, second data amount, and the size of the data to be backed up. Specifically, for example, the information processing apparatus 101 calculates the remaining data amount of the data to be backed up that has not been transferred by subtracting the first data amount and the second data amount acquired from each data transfer apparatus 102 from the size of the data to be backed up.

[0032] Thereby, the information processing apparatus 101 calculates the remaining data amount in consideration of not only the data actually transmitted to the backup storage 103 but also the duplicate data that has not been transmitted to the backup storage 103 due to the deduplication function. Then, the information processing apparatus 101 calculates the next transmission data amount based on the calculated remaining data amount and the remaining number of transmissions.

[0033] In the example of FIG. 1, the information processing apparatus 101 calculates the remaining data amount of the data to be backed up 110 that has not been transferred by subtracting the first data amount and the second data amount acquired from each of the data transfer apparatuses 102-1 to 102-3 from the size of the data to be backed up 110. Then, the information processing apparatus 101 calculates the next transmission data amount D2 based on the calculated remaining data amount and the remaining number of transmissions (P - 1).

[0034] (4) The information processing apparatus 101 increases or decreases the number of the plurality of data transfer apparatuses 102 based on the result of comparing the calculated previous transmission data amount and the next transmission data amount. For example, if the next transmission data amount is the same as the previous data amount, it can be said that it can be transmitted within the allowable time unless the network line status changes rapidly.

[0035] Therefore, for example, when the amount of data to be transmitted next time is larger than the amount of data transmitted last time, the information processing apparatus 101 may increase the number of a plurality of data transfer apparatuses 102. On the other hand, when the amount of data to be transmitted next time is smaller than the amount of data transmitted last time, the information processing apparatus 101 may decrease the number of a plurality of data transfer apparatuses 102.

[0036] In the example of FIG. 1, for example, when the amount of data D2 to be transmitted next time is larger than the amount of data D1 transmitted last time, the information processing apparatus 101 starts a new data transfer apparatus to increase the resources used for backup. On the other hand, when the amount of data D2 to be transmitted next time is smaller than the amount of data D1 transmitted last time, the information processing apparatus 101 ends the operation of at least any one of the data transfer apparatuses 102-1 to 102-3 to reduce the resources used for backup.

[0037] As described above, according to the information processing apparatus 101, while increasing or decreasing the number of a plurality of data transfer apparatuses 102 so that the backup is completed within the assumed time, the data to be backed up can be distributed and transferred to the backup storage 103 in multiple times.

[0038] In the example of FIG. 1, when there is a possibility that the transmission of backup data is not completed within the allowable time, the used resources (data transfer apparatus 102, network line) can be increased to prevent the backup time from being prolonged. Also, when there is a possibility that waste occurs in the used resources, the used resources (data transfer apparatus 102, network line) can be reduced to suppress wasteful used resources.

[0039] (System configuration example of backup system 200) Next, a system configuration example of the backup system 200 including the information processing apparatus 101 shown in FIG. 1 will be described. Here, a case where the information processing apparatus 101 shown in FIG. 1 is applied to the backup management apparatus 201 of the backup system 200 will be described as an example.

[0040] FIG. 2 is an explanatory diagram showing an example of the system configuration of the backup system 200. In FIG. 2, the backup system 200 includes a backup management device 201, a plurality of client devices 202, an administrator terminal 203, a host server 204, proxy servers P1 to Pn (n: a natural number of 2 or more), and a backup storage ST. In the backup system 200, the backup management device 201, the client devices 202, the administrator terminal 203, the host server 204, the proxy servers P1 to Pn, and the backup storage ST are connected via a wired or wireless network 210. The network 210 is, for example, the Internet, a LAN (Local Area Network), a WAN (Wide Area Network), or the like.

[0041] In the following description, any one of the proxy servers P1 to Pn may be referred to as "proxy server Pi" (i = 1, 2,..., n). The data transfer device 102 shown in FIG. 1 corresponds to, for example, the proxy server Pi.

[0042] Here, the backup management device 201 has a primary storage drive 220 and a data management table 230, and performs backup of backup target data. The backup management device 201 is, for example, a server. Also, the backup management device 201 may be realized by a virtual machine on a server (physical server). The primary storage drive 220 corresponds to a temporary storage location for the backup target data. However, the temporary storage location for the backup target data may be, for example, a DB (Database). The storage content of the data management table 230 will be described later with reference to FIG. 4.

[0043] The client device 202 is a computer that has data to be backed up. The data to be backed up is, for example, data of an application to be backed up (such as a business application). The client device 202 may be, for example, a server or a PC (Personal Computer). Also, the client device 202 may be realized by a virtual machine or a container on a server (physical server).

[0044] The administrator terminal 203 is a computer used by the administrator of the backup system 200. On the administrator terminal 203, the administrator can perform various operations related to backup, for example. The administrator terminal 203 is, for example, a PC, a tablet PC, or the like.

[0045] The host server 204 is a computer capable of running a virtual machine. A virtual machine is a virtual computer that operates in an execution environment constructed by dividing the hardware resources of a physical computer. A virtual machine is realized, for example, by virtualizing the hardware resources by a hypervisor.

[0046] Also, the host server 204 may be capable of running a container. A container corresponds to an isolated user space created by dividing the kernel of an OS (Operating System) internally and operates as one of the processes of the OS. The user space is an execution environment that aggregates the resources for the user to execute an application.

[0047] The proxy servers P1 to Pn are realized by virtual machines (or containers) on the host server 204. The proxy server Pi is a server for transferring backup data and has a deduplication function. The proxy server Pi transfers the data assigned from the backup management device 201 to the backup storage ST using the network line corresponding to its own server. However, each proxy server Pi may also be realized by a physical server respectively.

[0048] The backup storage ST is the storage destination of the backup target data. The backup storage ST is, for example, a server. The backup storage 103 shown in FIG. 1 corresponds to the backup storage ST, for example.

[0049] Note that the backup management device 201 may also be realized by a plurality of computers. Also, the client device 202 and the proxy server Pi may also be realized by the same computer (for example, the host server 204).

[0050] (Example of the hardware configuration of the backup management device 201) FIG. 3 is a block diagram showing an example of the hardware configuration of the backup management device 201. In FIG. 3, the backup management device 201 includes a CPU (Central Processing Unit) 301, a memory 302, a disk drive 303, a disk 304, a communication I / F (Interface) 305, a portable recording medium I / F 306, and a portable recording medium 307. Also, each component is connected by a bus 300 respectively.

[0051] Here, the CPU 301 controls the overall operation of the backup management device 201. The CPU 301 may have multiple cores. The memory 302 includes, for example, a ROM (Read Only Memory), a RAM (Random Access Memory), and a flash ROM. Specifically, for example, the flash ROM stores the OS program, the ROM stores the application program, and the RAM is used as the work area of the CPU 301. The program stored in the memory 302 is loaded into the CPU 301 to cause the CPU 301 to execute the coded processing.

[0052] The disk drive 303 controls the read / write of data to / from the disk 304 according to the control of the CPU 301. The disk 304 stores the data written under the control of the disk drive 303. Examples of the disk 304 include a magnetic disk and an optical disk.

[0053] The communication I / F 305 is connected to the network 210 through a communication line and is connected to an external computer (for example, the client device 202, the administrator terminal 203, the proxy server Pi, etc. shown in FIG. 2) via the network 210. Then, the communication I / F 305 serves as the interface between the network 210 and the inside of the device and controls the input / output of data from / to the external computer. For the communication I / F 305, for example, a modem or a LAN adapter can be adopted.

[0054] The portable recording medium I / F 306 controls the read / write of data to / from the portable recording medium 307 according to the control of the CPU 301. The portable recording medium 307 stores the data written under the control of the portable recording medium I / F 306. Examples of the portable recording medium 307 include a CD (Compact Disc)-ROM, a DVD (Digital Versatile Disk), a USB (Universal Serial Bus) memory, etc.

[0055] Note that, in addition to the components described above, the backup management device 201 may also have, for example, an input device, a display, etc. Also, the client device 202, the administrator terminal 203, the proxy server Pi, and the backup storage ST shown in FIG. 2 can also be realized with the same hardware configuration as the backup management device 201. However, the administrator terminal 203 has, in addition to the components described above, for example, an input device, a display, etc.

[0056] (Stored content of the data management table 230) Next, with reference to FIG. 4, the stored content of the data management table 230 included in the backup management device 201 will be described. The data management table 230 is realized by a storage device such as the memory 302 and the disk 304 shown in FIG. 3, for example.

[0057] FIG. 4 is an explanatory diagram showing an example of the stored content of the data management table 230. In FIG. 4, the data management table 230 has fields for the number of transmissions, Address, w, S, L, D rem , line enable, and speed rank. By setting information in each field, data management information (for example, data management information 400-1, 400-2) is stored as a record.

[0058] Here, the number of transmissions indicates the transmission number p representing which transfer it is among the specified number of transmissions P (p = 0, 1,..., p-1). Note that the first row record of each data management information indicates the overall information for the p-th time. Address is an identifier for identifying the proxy server Pi. However, the first row of each data management information is "0000".

[0059] w indicates the planned transmission data volume for the p-th time (unit: B (Byte)). The planned transmission data volume is the transmission data volume scheduled to be transmitted to the backup storage. S indicates the data volume for the p-th time transmitted from the proxy server Pi to the backup storage ST (unit: B). For example, S 10is the data volume of the p-th (p = 0) transmission from the proxy server P1 to the backup storage ST. However, the first line of each data management information is the total sum R of the data volumes of the p-th transmission from each proxy server Pi to the backup storage ST p indicates.

[0060] L indicates the data volume of the p-th transmission that was not sent to the backup storage ST due to the deduplication function of the proxy server Pi (unit: B). For example, L 10 is the data volume of the p-th (p = 0) transmission that was not sent to the backup storage ST due to the deduplication function of the proxy server P1. However, the first line of each data management information is the total sum L of the data volumes of the p-th transmission that were not sent to the backup storage ST due to the deduplication function of each proxy server Pi p indicates.

[0061] D rem indicates the remaining data volume among the backup target data at the end of the p-th transfer (unit: B). However, from the second line onwards of each data management information, it is "nul". The line enable "Y" indicates that the proxy server Pi is in use. The line enable "N" indicates that the proxy server Pi is not in use. However, the first line of each data management information is "nul".

[0062] The speed rank is the rank obtained by ranking the S values of each proxy server Pi in ascending order for the p-th time. The higher the rank of the speed rank, the slower the transfer speed. For example, the proxy server Pi with the speed rank of "1st place" indicates that the transfer speed is the slowest. However, the first line of each data management information is "nul".

[0063] (Functional configuration example of the backup management device 201) FIG. 5 is a block diagram showing a functional configuration example of the backup management apparatus 201. In FIG. 5, the backup management apparatus 201 includes a reception unit 501 and a backup control unit 502. The reception unit 501 and the backup control unit 502 are functions that serve as control units. Specifically, for example, the functions are realized by causing the CPU 301 to execute a program stored in a storage device such as the memory 302, the disk 304, and the portable recording medium 307 shown in FIG. 3, or by the communication I / F 305. The processing results of each functional unit are stored in a storage device such as the memory 302 and the disk 304, for example.

[0064] The reception unit 501 receives a backup instruction. The backup instruction is an instruction to back up data. The backup instruction includes, for example, information that can identify the backup target data. Further, the backup instruction includes, for example, an assumed time E and a transmission count P.

[0065] The assumed time E is the assumed time until the backup is completed (unit: h). The transmission count P indicates how many times the backup target data is transferred. Specifically, for example, the reception unit 501 receives a backup instruction from the administrator terminal 203 (or the client device 202) shown in FIG. 2. Further, the reception unit 501 may receive a backup instruction by a user's operation input using an input device (not shown).

[0066] The backup control unit 502 performs a backup of the backup target data. Specifically, for example, the backup control unit 502 distributes the backup target data to the backup storage ST in multiple times via the network lines corresponding to the respective proxy servers Pi of the proxy servers P1 to Pn in accordance with the backup instruction. The network line corresponding to each proxy server Pi is the communication resource allocated to each proxy server Pi. The start time of the backup may be set in advance.

[0067] More specifically, for example, the backup control unit 502 includes a collection unit 503, a division unit 504, an allocation unit 505, an acquisition unit 506, a first calculation unit 507, a second calculation unit 508, and a resource control unit 509.

[0068] The collection unit 503 acquires backup target data. Specifically, for example, the collection unit 503 acquires backup target data from the client device 202 shown in FIG. 2. Which data of which client device 202 to acquire is specified, for example, from a backup instruction.

[0069] Next, the collection unit 503 writes the acquired backup target data to the primary storage drive 220. At this time, the collection unit 503 may, for example, perform a primary deduplication process on the backup target data. The primary deduplication process is a process of comparing the previous backup target data with the current backup target data and removing duplicate data from the current backup target data.

[0070] The division unit 504 divides the backup target data. The size for dividing the backup target data is calculated, for example, by the resource control unit 509. Specifically, for example, the division unit 504 measures the data size W of the backup target data (for example, the backup target data after primary deduplication). Next, the resource control unit 509 calculates the transfer data size T (unit: B) based on the measured data size W (unit: B), the number of transmission times P, and the number n of proxy servers P1 to Pn.

[0071] The transfer data size T can be calculated, for example, using the following formula (1). The number of transmission times P is specified, for example, from a backup instruction.

[0072] T = W / (n * P) ···(1)

[0073] Then, the splitting unit 504 splits the data to be backed up according to the calculated transfer data size T. For example, the splitting unit 504 splits the data to be backed up so that the split data becomes data of the transfer data size T. As a result, the data to be backed up can be split into, for example, (n*P) pieces of data of approximately the same size (backup data).

[0074] The allocation unit 505 allocates the split data to each of the proxy servers P1 to Pn. The data to be allocated is different among the proxy servers. Specifically, for example, the allocation unit 505 transmits the split data as backup data to each of the proxy servers P1 to Pn.

[0075] At this time, the allocation unit 505 may, for example, specify the allowable time e for each proxy server Pi. The allowable time e is the time allowed for one transmission (unit: h). The allowable time e can be calculated using, for example, the following formula (2). The assumed time E and the number of transmissions P are specified from, for example, the backup instruction.

[0076] e = E / P ···(2)

[0077] Each proxy server Pi transfers the received backup data to the backup storage ST using the network line corresponding to its own server. At this time, each proxy server Pi performs, for example, deduplication processing on the backup data. Thereby, for the data already stored in the backup storage ST, it is prevented from being newly transmitted.

[0078] Specifically, for example, the proxy server Pi splits the backup data into blocks. The block size can be arbitrarily set and is set to about 1 [M], for example. Next, the proxy server Pi calculates the hash value of the split block. Then, the proxy server Pi transmits a duplicate check request including the calculated hash value to the backup storage ST.

[0079] When the backup storage ST receives a duplicate check request, it compares the hash value included in the duplicate check request with the hash value of the already stored data. Here, if there is data with a matching hash value, the backup storage ST sends a duplicate check response indicating duplication to the proxy server Pi. On the other hand, if there is no data with a matching hash value, the backup storage ST sends a duplicate check response indicating no duplication to the proxy server Pi.

[0080] When the proxy server Pi receives a duplicate check response, it determines whether to send the corresponding block (the block from which the hash value was calculated). For example, if the duplicate check response indicates no duplication, the proxy server Pi sends the corresponding block to the backup storage ST. On the other hand, if the duplicate check response indicates duplication, the proxy server Pi does not send the corresponding block to the backup storage ST.

[0081] Also, the proxy server Pi measures the first data amount among the backup data that was sent to the backup storage ST. Specifically, for example, the proxy server Pi measures the first data amount by accumulating the sizes of the blocks sent to the backup storage ST.

[0082] Also, the proxy server Pi measures the second data amount among the backup data that was not sent to the backup storage ST due to the deduplication function. Specifically, for example, the proxy server Pi measures the second data amount by accumulating the sizes of the blocks not sent to the backup storage ST.

[0083] In the following description, the first data amount among the backup data that was sent to the backup storage ST may be referred to as the "sent data amount". Also, the second data amount among the backup data that was not sent to the backup storage ST may be referred to as the "assumed sent data amount".

[0084] When the transmission of backup data is completed, the proxy server Pi transmits the measured transmission data volume and the assumed transmission data volume to the backup management device 201. Also, due to network congestion or the like, the transmission of backup data may not be completed within the allowable time e.

[0085] In this case, the proxy server Pi transmits the measured transmission data volume and the assumed transmission data volume to the backup management device 201 together with the remaining backup data. The remaining backup data is data (for example, unprocessed blocks) in the backup data that could not be transmitted to the backup storage ST.

[0086] The acquisition unit 506 acquires the assumed transmission data volume from each proxy server Pi. Here, the transmission data volume is the volume of data (backup data) previously assigned to the proxy server Pi that has been transmitted to the backup storage ST. The assumed transmission data volume is the volume of data (backup data) previously assigned to the proxy server Pi that has not been transmitted to the backup storage ST due to the deduplication function.

[0087] The first calculation unit 507 calculates the previous transmission data volume R transmitted to the backup storage by the proxy servers P1 to Pn based on the acquired transmission data volume. p Specifically, for example, the first calculation unit 507 calculates the previous (p-th) transmission data volume R by obtaining the sum of the transmission data volumes acquired from each proxy server Pi using the following formula (3). p However, S ip is the transmission data volume of the proxy server Pi at the p-th time.

[0088] R p =ΣS ip ···(3)

[0089] The calculated transmission data volume R pis stored, for example, in the data management table 230 shown in FIG. 4.

[0090] The second calculation unit 508 calculates the next transmission data amount w to be transmitted to the backup storage ST based on the acquired transmission data amount regarded as such, the transmission data amount regarded as such, and the size of the backup target data (data size W). Specifically, for example, the second calculation unit 508 subtracts the transmission data amount regarded as such acquired from each proxy server Pi from the data size W of the backup target data to calculate the remaining data amount (remaining data amount) of the backup target data that has not been transferred. Then, the second calculation unit 508 calculates the next transmission data amount w based on the calculated remaining data amount and the remaining number of transmissions.

[0091] Here, the case of calculating the second transmission data amount w as the next transmission data amount w will be described as an example. In this case, the second calculation unit 508 refers to the data management table 230 and calculates the remaining data amount D of the first time using the following formula (4). However, L rem0 is the assumed transmission data amount of the p-th proxy server Pi. ΣL ip is the assumed transmission data amount (total) of the p-th time. ip is the assumed transmission data amount (total) of the p-th time.

[0092] D remp =W - ΣS ip - ΣL ip ···(4)

[0093] According to the above formula (4), the remaining data amount D rem0 of the first time (p = 0) is "D rem0 =W - R0 - L0".

[0094] Note that the remaining data amount D remp may be calculated using, for example, the following formula (5). L p is the assumed transmission data amount (total) of the p-th time and corresponds to "ΣL ip ".

[0095] Dremp = D remp-1 - R p - L p ···(5)

[0096] Then, the second calculation unit 508 calculates the remaining data amount D calculated, for example, using the following formula (6). rem0 Based on the remaining data amount D p+1 and the remaining number of transmission times (P - 1), the next transmission data amount w

[0097] w p+1 = D remp / (P - p - 1) ···(6)

[0098] According to the above formula (6), the next (second) transmission data amount w1 is "D rem0 / (P - 1)".

[0099] The calculated remaining data amount D rem0 and the transmission data amount w1 are stored, for example, in the data management table 230.

[0100] The resource control unit 509 increases or decreases the number n of proxy servers P1 to Pn based on the result of comparing the calculated previous transmission data amount R p and the next transmission data amount w p+1 Specifically, for example, when the next transmission data amount w p+1 is larger than the previous transmission data amount R p the resource control unit 509 increases the number n of proxy servers P1 to Pn.

[0101] On the other hand, when the next transmission data amount w p+1 is smaller than the previous transmission data amount R p the resource control unit 509 decreases the number n of proxy servers P1 to Pn. Note that when the previous transmission data amount R p and the next transmission data amount w p+1 are the same, the resource control unit 509 may choose not to increase or decrease the number n of proxy servers P1 to Pn.

[0102] Here, assume a case where the number n of proxy servers P1 to Pn is increased. The number x (the first number) of proxy servers to be increased may be determined in advance, for example. Also, the resource control unit 509 calculates the amount of transmission data w for the next time p+1 from the amount of transmission data R for the previous time p and the assumed amount of transmission data L for the previous time p and may calculate the number x of proxy servers to be increased based on the result of the subtraction.

[0103] Specifically, for example, the resource control unit 509 refers to the data management table 230 and determines the minimum transmission data amount S 1p ~S np among the transmission data amounts S ip obtained from each proxy server Pi. Then, the resource control unit 509 uses the following formula (7) to divide the result of subtracting the amount of transmission data R p+1 for the previous time from the amount of transmission data w p for the next time and the assumed amount of transmission data L p for the previous time by the specified minimum transmission data amount S ip _1 to calculate the number x of proxy servers to be increased (for example, rounding up).

[0104] x = (w p+1 - R p - L p ) / S ip _1 ··· (7)

[0105] In this case, the resource control unit 509 increases the number n of proxy servers P1 to Pn by the calculated number x. Specifically, for example, the resource control unit 509 cooperates with the host server 204 shown in FIG. 2 and newly activates x proxy servers as backup data transfer servers (proxy servers P1 to Pn). Thereby, the number n of proxy servers P1 to Pn can be increased by the number x.

[0106] Next, assume a case where the number n of proxy servers P1 to Pn is decreased. The number y (the second number) of proxy servers to be decreased may be determined in advance, for example. Also, the resource control unit 509 uses the previous transmission data amount R p and the next transmission data amount w p+1 to calculate the number y of proxy servers to be decreased based on the difference therebetween.

[0107] Specifically, for example, the resource control unit 509 refers to the data management table 230 and selects, in ascending order, the smallest transmission data amounts S 1p ~S np among the previous (p-th) transmission data amounts, and subtracts the selected transmission data amount S ip from the previous transmission data amount R ip . Then, the resource control unit 509 determines whether the value after subtraction is equal to or greater than the next transmission data amount w p . p+1

[0108] Here, when the value after subtraction is equal to or greater than the next transmission data amount w p+1 , the proxy server Pi corresponding to the selected transmission data amount S ip is determined as the proxy server to be decreased. When the proxy server Pi is determined as the proxy server to be decreased, for example, "N" is set in the line enable of the proxy server Pi in the data management table 230.

[0109] The resource control unit 509 repeatedly determines the proxy servers to be decreased until the value after subtraction becomes less than the next transmission data amount w p+1 . The number of determined proxy servers becomes the number y of proxy servers to be decreased. Thereby, the number y of proxy servers to be decreased can be calculated, and the proxy server Pi to be decreased can be determined.

[0110] ​Also, it is possible to determine the proxy servers to be decreased in order from the proxy servers with a small amount of transmission data in the previous (p-th) time. As a result, it is possible to preferentially stop using the proxy servers whose transfer speed is slow due to congestion of the network line or the like, and efficiently suppress the used resources.

[0111] Further, the dividing unit 504 divides the remaining backup target data according to the size set by the resource control unit 509. Here, the remaining backup target data is the remaining data among the backup target data that has not been transferred to the backup storage ST. The remaining backup target data includes, for example, data (remaining backup data) returned from the proxy server Pi because the transmission of the backup data was not completed within the allowable time e.

[0112] Specifically, for example, when the resource control unit 509 increases the number n of the proxy servers P1 to Pn, it refers to the data management table 230 and, among the increased proxy servers P1 to Pn, the size of the data (the first transfer data size T p+1 ) allocated to the proxy servers before the increase is set to the same size (transfer data size T p ) as the data allocated last time.

[0113] Also, the resource control unit 509 refers to the data management table 230 and sets the size of the data (the second transfer data size T p+1 ) allocated to the increased proxy servers to the transmission data amount S ip _1. The transmission data amount S ip _1 is the minimum transmission data amount (speed rank "1") among the transmission data amounts S 1p to S np in the previous (p-th) time.

[0114] Then, the dividing unit 504 divides according to the set sizes (the first and second transfer data sizes T p+1According to (), the remaining backup target data is divided. More specifically, for example, the dividing unit 504 divides, from the remaining backup target data, data of the size of the first transfer data size T by the number of proxy servers before the increase. p+1 The dividing unit 504 also divides, from the remaining backup target data, data of the size of the second transfer data size T by the number of increased proxy servers. p+1

[0115] In this case, the allocating unit 505 allocates the divided data of the size of the first transfer data size T to the proxy servers before the increase. The allocating unit 505 also allocates the divided data of the size of the second transfer data size T to the increased proxy servers. p+1 p+1

[0116] Also, when the resource control unit 509 decreases the number n of proxy servers P1 to Pn, it refers to the data management table 230 and sets the size of the data (transfer data size T) to be allocated to the decreased proxy server Pi to the same size (transfer data size T) as the previously allocated data. Then, the dividing unit 504 divides the remaining backup target data according to the set size (transfer data size T). p+1 p p+1

[0117] More specifically, for example, the dividing unit 504 divides, from the remaining backup target data, data of the size of the transfer data size T by the number of proxy servers after the decrease. In this case, the allocating unit 505 allocates the divided data of the size of the transfer data size T to the proxy servers after the decrease. p+1 p+1

[0118] Note that in the above description, as the specific processing content for calculating the previous (p-th) transmission data amount R, the case of obtaining the sum of the transmission data amounts S obtained from each proxy server Pi using the above formula (3) has been described, but it is not limited to this. p ip ​​​​​​​​​​

[0119] For example, the first calculation unit 507 calculates the transfer speed v ip calculated from the transmission data amount S ip to calculate the previous (p-th) transmission data amount R p if desired. The transfer speed v ip indicates the transfer speed of each proxy server Pi in the previous (p-th) time (unit: [B / h]).

[0120] More specifically, for example, the proxy server Pi measures the time (transmission time t) taken to transmit the data for the transmission data amount S ip (0 < t < e). Then, the proxy server Pi calculates the transfer speed v ip from the measured transmission time t and the transmission data amount S ip and notifies the calculated transfer speed v ip to the backup management device 201.

[0121] The backup management device 201 calculates the transmission data amount R ip from the notified transfer speed v p and the allowable time e. For example, the transmission data amount R p is "R p = v ip * e". Note that the calculation of the transfer speed v ip may also be calculated by the backup management device 201 from the transmission time t and the transmission data amount S ip notified from the proxy server Pi.

[0122] Also, in the above description, the number x of proxy servers to be increased is calculated by dividing the result obtained by subtracting the previous transmission data amount R p+1 and the previous estimated transmission data amount L p from the next transmission data amount w p by the minimum transmission data amount S ip _1, but it is not limited to this. For example, the resource control unit 509 divides the subtracted result by the maximum transmission data amount S 1p ~ S np among Sip _ n By dividing, the number x of proxy servers to be increased may be calculated. In this case, the resource control unit 509, for example, the size of the data to be allocated to the increased proxy server (the second transfer data size T p+1 ) is set to the transmission data amount S ip _ n . Thereby, the resources to be used at the next transmission can be suppressed. Further, the resource control unit 509 divides the subtracted result by the average transmission data amount of the transmission data amounts S 1p ~S np to calculate the number x of proxy servers to be increased. In this case, the resource control unit 509, for example, sets the size of the data to be allocated to the increased proxy server (the second transfer data size T p+1 ) to the average transmission data amount.

[0123] (One embodiment of the backup management device 201) Next, with reference to FIG. 6, one embodiment of the backup management device 201 will be described.

[0124] FIG. 6 is an explanatory diagram showing one embodiment of the backup management device 201. In FIG. 6, the backup management device 201 includes a backup data collection module 601, a backup data division module 602, a calculation module 603, and a scheduler 604.

[0125] Here, the backup data collection module 601 acquires backup target data from the client device 202 and writes the acquired backup target data to the primary storage drive 220. The collection unit 503 shown in FIG. 5 corresponds to, for example, the backup data collection module 601.

[0126] The backup data splitting module 602 splits the data to be backed up according to the size notified by the scheduler 604, and allocates the split data (backup data) to each proxy server Pi. Each proxy server Pi transmits the backup data to the backup storage ST. The splitting unit 504 and the allocation unit 505 shown in FIG. 5 correspond to, for example, the backup data splitting module 602.

[0127] The calculation module 603 obtains the amount of transmitted data S ip regarded as such and the amount of transmitted data L ip from each proxy server Pi. Based on the obtained amount of transmitted data S ip , the calculation module 603 calculates the previous amount of transmitted data R p transmitted to the backup storage ST by the proxy servers P1 to Pn.

[0128] Furthermore, the calculation module 603 calculates the next amount of transmitted data w ip regarded as such and the amount of transmitted data L ip to be transmitted to the backup storage ST based on the obtained amount of transmitted data S p+1 and the data size W of the data to be backed up. The acquisition unit 506, the first calculation unit 507, and the second calculation unit 508 shown in FIG. 5 correspond to, for example, the calculation module 603.

[0129] The scheduler 604 increases or decreases the number n of the proxy servers P1 to Pn based on the result of comparing the calculated previous amount of transmitted data R p and the next amount of transmitted data w p+1 . For example, the scheduler 604 increases or decreases the number n of the proxy servers P1 to Pn by notifying the host server 204 of the number of proxy servers to be increased or decreased.

[0130] Furthermore, the scheduler 604 notifies the backup data splitting module 602 of the size (transfer data size) for splitting the data to be backed up. The resource control unit 509 shown in FIG. 5 corresponds to, for example, the scheduler 604.

[0131] Note that each of the above-described various modules 601 to 604 may be implemented by a plurality of computers. For example, the backup data collection module 601 and the backup data division module 602 may be implemented by a repository server, and the calculation module 603 and the scheduler 604 may be implemented by a management server. In this case, the backup management device 201 may be implemented by, for example, two computers, namely, a repository server and a management server.

[0132] (Operation example of backup system 200) Next, with reference to FIGS. 7 to 9, an operation example of the backup system 200 will be described.

[0133] FIGS. 7 to 9 are sequence diagrams showing an operation example of the backup system 200. In FIG. 7, the scheduler 604 controls the host server 204 to deploy the proxy servers P1 to Pn (step S701). The deployment of the proxy server Pi means starting the proxy server Pi on the host server 204 to make it available.

[0134] Next, the backup data collection module 601 acquires backup target data from the client device 202 (step S702). Then, the backup data collection module 601 writes the acquired backup target data to the primary storage drive 220 (step S703).

[0135] Next, the backup data division module 602 divides the backup target data written to the primary storage drive 220 according to the size notified by the scheduler 604 (step S704). Then, the backup data division module 602 assigns the divided data (backup data) to each proxy server Pi (step S705).

[0136] In FIG. 8, the proxy server Pi divides the assigned backup data into blocks (step S801). Next, the proxy server Pi calculates the hash value of the divided blocks (step S802). Then, the proxy server Pi transmits the calculated hash value to the backup storage ST (step S803).

[0137] When the backup storage ST receives the hash value, it compares the received hash value with the hash value of the already stored data (step S804). Then, the backup storage ST transmits the comparison result to the proxy server Pi (step S805).

[0138] Based on the received comparison result, the proxy server Pi performs deduplication processing on the backup data (step S806). Next, the proxy server Pi transmits the backup data after deduplication to the backup storage ST (step S807).

[0139] In FIG. 9, the proxy server Pi transmits the measured transmission data volume S ip and the assumed transmission data volume L ip to the calculation module 603 (step S901). The calculation module 603 writes the received transmission data volume S ip and the assumed transmission data volume L ip to the data management table 230 (step S902).

[0140] Next, the calculation module 603 refers to the data management table 230 and calculates the previous transmission data volume R p , the remaining data volume D remp and the next transmission data volume w p+1 (step S903). Then, the calculation module 603 writes the calculated previous transmission data volume R p , the remaining data volume D remp and the next transmission data volume w p+1 to the data management table 230 (step S904).

[0141] Next, the scheduler 604 calculates the number of increased or decreased proxy servers (x, y) with reference to the data management table 230 (step S905). Then, the scheduler 604 refers to the data management table 230 and determines the next transfer data size T of each proxy server Pi after the increase or decrease p+1 for setting (step S906).

[0142] Next, the scheduler 604 notifies the host server 204 of the calculated number of increased or decreased proxy servers (x, y) (step S907). As a result, in the host server 204, the number n of the proxy servers P1 to Pn is increased or decreased.

[0143] Then, the scheduler 604 notifies the backup data division module 602 of the next transfer data size T of each set proxy server Pi p+1 (step S908). As a result, in the backup data division module 602, according to the notified next transfer data size T p+1 , the backup target data (remaining backup target data) is divided and assigned to each proxy server Pi after the increase or decrease.

[0144] Thereby, while increasing or decreasing the number n of the proxy servers P1 to Pn so that the backup is completed within the assumed time E, the backup target data can be transferred to the backup storage ST in multiple times and dispersed.

[0145] (Backup Example of Backup Target Data) FIG. 10 is an explanatory diagram showing a backup example of backup target data. In FIG. 10, when transferring the backup target data 1000 to the backup storage ST in multiple times and dispersing it via the network lines corresponding to the respective proxy servers Pi of the proxy servers P1 to Pn, the backup times t1, t2,..., tn of each proxy server Pi are shown. The backup data transmitted from each of the proxy servers P1 to Pn is stored in the partitions 1 to n corresponding to each of the proxy servers P1 to Pn.

[0146] In the backup system 200, it is possible to adjust the size of backup data and resources (network line, proxy server Pi) that are transmitted in response to fluctuations in the bandwidth of the network line almost in real time. As a result, it is possible to shorten the backup time so that the backup is completed within the assumed time E, and to use resources efficiently. For example, although the backup time at each proxy server Pi varies only for the first time, it can be expected to be equal from the second time onwards.

[0147] (Backup management processing procedure of backup management device 201) Next, with reference to FIGS. 11 and 12, the backup management processing procedure of the backup management device 201 will be described. The backup management process is executed, for example, in response to a backup instruction from the administrator terminal 203 or the client device 202.

[0148] FIGS. 11 and 12 are flowcharts showing an example of the backup management processing procedure of the backup management device 201. In FIG. 11, the backup management device 201 controls the host server 204 to deploy the proxy servers P1 to Pn (step S1101).

[0149] Next, the backup management device 201 acquires backup target data from the client device 202 (step S1102). Then, the backup management device 201 writes the acquired backup target data to the primary storage drive 220 (step S1103).

[0150] Next, the backup management device 201 measures the data size W of the backup target data with reference to the primary storage drive 220 (step S1104). At this time, the backup management device 201 may, for example, perform a primary deduplication process on the backup target data.

[0151] Then, the backup management device 201 calculates the transfer data size T based on the measured data size W, the number of transmission times P (predetermined number of transmission times), and the number n of proxy servers P1 to Pn (step S1105). Next, the backup management device 201 divides the backup target data according to the calculated transfer data size T (step S1106).

[0152] Then, the backup management device 201 assigns the divided data to each proxy server Pi of the proxy servers P1 to Pn (step S1107). Next, the backup management device 201 sets the number of transmission times p to "p = 0" (step S1108) and proceeds to step S1201 shown in FIG. 12.

[0153] In the flowchart of FIG. 12, first, the backup management device 201 transmits the data assigned to each proxy server Pi as backup data to each proxy server Pi (step S1201). Next, the backup management device 201 determines whether it has received the transmission data amount S ip and the assumed transmission data amount L ip from each proxy server Pi (step S1202).

[0154] Here, the backup management device 201 waits to receive the transmission data amount S ip and the assumed transmission data amount L ip from each proxy server Pi (step S1202: No). If the transmission of the backup data is not completed within the allowable time e in the proxy server Pi, the backup management device 201 receives the remaining backup data from the proxy server Pi.

[0155] Then, when the backup management device 201 receives the transmission data amount S ip and the assumed transmission data amount L ip from each proxy server Pi (step S1202: Yes), the transmission data amount S of each received proxy server Pi ip and the assumed transmission data amount L ipRecord it in the data management table 230 (step S1203). Next, the backup management device 201 executes a resource control process for increasing or decreasing the number n of proxy servers P1 to Pn (step S1204). The specific processing procedure of the resource control process will be described later with reference to FIGS. 13 and 14.

[0156] Then, the backup management device 201 p+1 divides the remaining backup target data according to the next transfer data size T (step S1205). Next, the backup management device 201 allocates the divided data to each proxy server Pi (step S1206).

[0157] Next, the backup management device 201 determines whether the transmission count p is "p = P - 1" (step S1207). Here, when the transmission count p is not "p = P - 1" (step S1207: No), the backup management device 201 increments the transmission count p (step S1208) and returns to step S1201.

[0158] On the other hand, when the transmission count p is "p = P - 1" (step S1207: Yes), the backup management device 201 ends the series of processes according to this flowchart. In step S1205, if there is no remaining backup target data, the backup management device 201 ends the series of processes according to this flowchart.

[0159] Thereby, while increasing or decreasing the number n of proxy servers P1 to Pn so that the backup is completed within the assumed time E, the backup target data can be transferred to the backup storage ST in multiple dispersions.

[0160] Next, with reference to FIGS. 13 and 14, the specific processing procedure of the resource control process in step S1204 shown in FIG. 12 will be described.

[0161] Figures 13 and 14 are flowcharts showing an example of the specific processing procedure of the resource control process. In the flowchart of Figure 13, first, the backup management device 201 determines whether the transmission count p is "p = 0" (step S1301). Here, when the transmission count p is "p = 0" (step S1301: Yes), the backup management device 201 calculates the previous (first) transmission data amount R0 and the assumed transmission data amount L0 based on the received transmission data amount S ip and the assumed transmission data amount L ip of each proxy server Pi (step S1302).

[0162] Then, the backup management device 201 refers to the data management table 230 using the above formula (5) to calculate the remaining data amount D rem0 for the first time (step S1303). Next, the backup management device 201 calculates the next transmission data amount w1 based on the calculated remaining data amount D rem0 and the remaining number of transmissions (P - 1) using the above formula (6) (step S1304).

[0163] Then, the backup management device 201 records the calculated remaining data amount D rem0 and the next transmission data amount w1 in the data management table 230 (step S1305). Next, the backup management device 201 ranks the transmission data amounts S 1p ~S np in ascending order and records the smallest transmission data amount S ip in the data management table 230 (step S1306).

[0164] Then, the backup management device 201 determines whether the next transmission data amount w1 is greater than the previous (first) transmission data amount R0 (step S1307). Here, when the next transmission data amount w1 is greater than the previous transmission data amount R0 (step S1307: Yes), the backup management device 201 refers to the data management table 230 and calculates the number x of proxy servers to be increased using the above formula (7) (step S1308).

[0165] Then, the backup management device 201 controls the host server 204 to increase the number n of proxy servers P1 to Pn by x (step S1309). Next, the backup management device 201 refers to the data management table 230 and determines the size of the data (transfer data size T p+1 ) to be assigned to the existing (before the increase) proxy servers among the increased proxy servers P1 to Pn to be the same size as the data assigned last time (transfer data size T p ). (Step S1310).

[0166] Then, the backup management device 201 refers to the data management table 230 and determines the size of the data (transfer data size T p+1 ) to be assigned to the increased proxy servers to be the transmission data volume S ip _1 (step S1311), and returns to the step where the resource control process was called. The transmission data volume S ip _1 is the minimum transmission data volume (speed rank "1") among the transmission data volumes S 1p ~S np in the previous (p-th) time.

[0167] Also, in step S1307, when the next transmission data volume w1 is less than or equal to the previous (first) transmission data volume R0 (step S1307: No), the backup management device 201 executes a decrease number calculation process for calculating the number x of proxy servers to be decreased (step S1312). The specific processing procedure of the decrease number calculation process will be described later with reference to FIG. 15.

[0168] Then, the backup management device 201 controls the host server 204 to decrease the number n of proxy servers P1 to Pn by y (step S1313). Next, the backup management device 201 refers to the data management table 230 and determines the size of the data (transfer data size T p+1 ) to be assigned to the proxy server Pi after the decrease to be the same size as the data assigned last time (transfer data size T p) is set to (step S1314), and the process returns to the step where the resource control process was called.

[0169] Also, in step S1301, when the number of transmissions p is not "p = 0" (step S1301: No), the backup management device 201 proceeds to step S1401 shown in FIG. 14.

[0170] In the flowchart of FIG. 14, first, the backup management device 201 calculates the transmission data volume S ip and the assumed transmission data volume L ip of each received proxy server Pi, and based on these, calculates the previous (first) transmission data volume R p and the assumed transmission data volume L p (step S1401).

[0171] Then, the backup management device 201 calculates the remaining data volume D remp for the p-th time using the above formula (5) (step S1402). Next, the backup management device 201 calculates the next transmission data volume w remp based on the calculated remaining data volume D p+1 and the remaining number of transmissions (P - p - 1) using the above formula (6) (step S1403).

[0172] Then, the backup management device 201 records the calculated remaining data volume D remp and the next transmission data volume w p+1 in the data management table 230 (step S1404). Next, the backup management device 201 ranks the transmission data volumes S 1p ~S np in ascending order and records the smallest transmission data volume S ip in the data management table 230 (step S1405).

[0173] Then, the backup management device 201 determines whether the next transmission data volume w p+1 is greater than the previous transmission data volume R p (step S1406). Here, the next transmission data volume wp+1 is greater than the previous transmission data volume R p If it is larger (step S1406: Yes), the backup management device 201 refers to the data management table 230 and calculates the number x of proxy servers to be increased using the above formula (7) (step S1407).

[0174] Then, the backup management device 201 controls the host server 204 to increase the number n of proxy servers P1 to Pn by x units (step S1408). Next, the backup management device 201 refers to the data management table 230 and, among the proxy servers P1 to Pn after the increase, the size of the data (transfer data size T p+1 ) to be assigned to the existing (before the increase) proxy servers is set to the same size as the previously assigned data (transfer data size T p ) (step S1409).

[0175] Then, the backup management device 201 refers to the data management table 230 and, for the size of the data (transfer data size T p+1 ) to be assigned to the increased proxy servers, sets it to the transmission data volume S ip _1 (step S1410), and returns to the step where the resource control process was called.

[0176] Also, in step S1406, if the next transmission data volume w p+1 is less than or equal to the previous transmission data volume R p (step S1406: No), the backup management device 201 executes a decrease number calculation process for calculating the number y of proxy servers to be decreased (step S1411). The specific processing procedure of the decrease number calculation process will be described later with reference to FIG. 15.

[0177] Then, the backup management device 201 controls the host server 204 to decrease the number n of proxy servers P1 to Pn by y units (step S1412). Next, the backup management device 201 refers to the data management table 230 and, for the size of the data (transfer data size Tp+1 ) is set to the same size as the previously assigned data (transfer data size T p ) (step S1413), and the process returns to the step where the resource control process was called.

[0178] As a result, it is possible to adjust the size of the backup data and the resources (network line, proxy server Pi) to be transmitted in response to fluctuations in the network line bandwidth almost in real time.

[0179] Next, with reference to FIG. 15, the specific processing procedures of the decrease amount calculation process in step S1312 shown in FIG. 13 and step S1411 shown in FIG. 14 will be described.

[0180] FIG. 15 is a flowchart showing an example of the specific processing procedure of the decrease amount calculation process. In FIG. 15, the backup management device 201 sets the parameter u to "u = 1" (step S1501) and sets the parameter V to "V = R p " (step S1502).

[0181] Next, the backup management device 201 subtracts the transmission data amount S ip _ u from the parameter V (step S1503). Note that the transmission data amount S ip _ u is the u-th smallest transmission data amount among the transmission data amounts S 1p ~S np . Then, the backup management device 201 determines whether the parameter V is greater than the next transmission data amount w p+1 (step S1504).

[0182] Here, when the parameter V is greater than the next transmission data amount w p+1 (step S1504: Yes), the backup management device 201 determines the transmission data amount S ip _ uSet "N" in the line enable of the proxy server Pi corresponding thereto (step S1505). Next, the backup management device 201 increments the parameter u (step S1506) and returns to step S1503.

[0183] On the other hand, when the parameter V is less than the next transmission data volume w p+1 In the following case (step S1504: No), the backup management device 201 sets "Y" in the line enable of the proxy server Pi corresponding to the transmission data volume S ip _ u (step S1507). Then, the backup management device 201 calculates the number y (= u - 1) of proxy servers to be decreased (step S1508) and returns to the step where the decrease number calculation process was called.

[0184] Thus, based on the difference between the previous transmission data volume R p and the next transmission data volume w p+1 it is possible to calculate the number y of proxy servers to be decreased so that the next transmission data volume w p+1 can be transmitted within the allowable time e and the used resources are minimized.

[0185] In step S1412 shown in FIG. 14, when decreasing the number n of proxy servers P1 to Pn, the backup management device 201 may not delete the proxy servers (for example, y proxy servers) but may set them to the stopped state. In this case, the backup management device 201 sets, for example, the size of the data (transfer data size T p+1 ) assigned to the stopped proxy server to "0". Thereby, for example, in the host server 204, although the hardware resources are still assigned to the stopped proxy server, the network line is not used.

[0186] Further, when the backup management device 201 increases the number n of proxy servers P1 to Pn by x, it may resume the stopped proxy servers. In this case, the backup management device 201 sets, for example, "Y" in the line enable of the resumed proxy server Pi. As a result, for example, the number n of proxy servers P1 to Pn can be increased more quickly than when starting the proxy servers from scratch.

[0187] (Various processing procedures of proxy server Pi) Next, the data transfer processing procedure of the proxy server Pi will be described. First, with reference to FIG. 16, the data transfer processing procedure of the proxy server Pi will be described. Here, the case where the proxy server Pi can send all the backup data to the backup storage ST within the allowable time e will be described.

[0188] FIG. 16 is a flowchart showing an example of the data transfer processing procedure of the proxy server Pi. In the flowchart of FIG. 16, first, the proxy server Pi determines whether it has received backup data from the backup management device 201 (step S1601). Here, the proxy server Pi waits to receive the backup data (step S1601: No).

[0189] When the proxy server Pi receives the backup data (step S1601: Yes), it divides the received backup data into blocks (step S1602). Next, the proxy server Pi calculates the hash value of the divided blocks (step S1603).

[0190] Then, the proxy server Pi sends a duplicate check request including the calculated hash value to the backup storage ST (step S1604). Next, the proxy server Pi determines whether it has received a duplicate check response indicating duplication from the backup storage ST (step S1605).

[0191] Here, when receiving a duplicate check response indicating no duplicates (step S1605: No), the proxy server Pi transmits the divided blocks to the backup storage ST (step S1606). Then, the proxy server Pi updates the transmission data volume S ip by accumulating the sizes of the transmitted blocks (step S1607), and proceeds to step S1609.

[0192] Also, in step S1605, when receiving a duplicate check response indicating duplicates (step S1605: Yes), the proxy server Pi updates the assumed transmission data volume L ip by accumulating the sizes of the divided blocks (step S1608).

[0193] Next, the proxy server Pi determines whether there are any unprocessed blocks in the backup data (step S1609). Here, if there are unprocessed blocks (step S1609: Yes), the proxy server Pi returns to step S1603 to calculate the hash values of the unprocessed blocks.

[0194] On the other hand, if there are no unprocessed blocks (step S1609: No), the proxy server Pi transmits the transmission data volume S ip and the assumed transmission data volume L ip to the backup management device 201 (step S1610), and ends the series of processes according to this flowchart.

[0195] Thereby, the proxy server Pi can transfer the backup data assigned from the backup management device 201 to the backup storage ST. Also, the proxy server Pi can measure the transmission data volume S ip and the assumed transmission data volume L ip and notify the backup management device 201.

[0196] Next, with reference to FIG. 17, the data return processing procedure of the proxy server Pi will be described. The data return processing is executed when the proxy server Pi cannot finish sending the backup data to the backup storage ST within the allowable time e.

[0197] FIG. 17 is a flowchart showing an example of the data return processing procedure of the proxy server Pi. In the flowchart of FIG. 17, first, the proxy server Pi determines whether it has received a data return request from the backup management device 201 (step S1701). The data return request requests to return the remaining data among the backup data that has not been transferred to the backup storage ST.

[0198] For example, when the allowable time e has elapsed after the backup management device 201 has sent the backup data to the proxy server Pi, the backup management device 201 sends a data return request to the proxy server Pi. However, if the backup management device 201 has received the transmitted data volume S ip and the assumed transmitted data volume L ip from the proxy server Pi before the allowable time e has elapsed, the backup management device 201 does not send a data return request to the proxy server Pi.

[0199] The proxy server Pi waits to receive a data return request (step S1701: No). When the proxy server Pi receives a data return request (step S1701: Yes), it ends the execution of the data transfer process shown in FIG. 17 (step S1702). Then, the proxy server Pi sends the remaining backup data together with the transmitted data volume S ip and the assumed transmitted data volume L ip to the backup management device 201 (step S1703), and ends a series of processes according to this flowchart.

[0200] Thereby, when the proxy server Pi cannot finish sending the backup data to the backup storage ST within the allowable time e, it can return the remaining backup data to the backup management device 201.

[0201] As described above, according to the backup management apparatus 201 according to the embodiment, while transferring backup target data to the backup storage ST in multiple dispersions via the network lines corresponding to the respective proxy servers Pi of the proxy servers P1 to Pn, the transmission data volume (S ip ) can be regarded as and the transmission data volume (L ip ) can be acquired. The transmission data volume is the data volume of the data (backup data) previously assigned to the proxy server Pi that has been transmitted to the backup storage ST. The assumed transmission data volume is the data volume of the data (backup data) previously assigned to the proxy server Pi that has not been transmitted to the backup storage ST due to the deduplication function. Further, according to the backup management apparatus 201, based on the acquired transmission data volume, the previous transmission data volume R p transmitted to the backup storage ST by the proxy servers P1 to Pn is calculated, and based on the acquired transmission data volume, the assumed transmission data volume, and the size of the backup target data (data size W), the next transmission data volume w p+1 to be transmitted to the backup storage ST can be calculated. Then, according to the backup management apparatus 201, based on the result of comparing the calculated previous transmission data volume R p and the next transmission data volume w p+1 , the number n of the proxy servers P1 to Pn can be increased or decreased.

[0202] Thereby, while increasing or decreasing the number n of the proxy servers P1 to Pn so that the backup is completed within the assumed time E, the backup target data can be transferred to the backup storage ST in multiple dispersions.

[0203] Further, according to the backup management apparatus 201, when the next transmission data volume w p+1 is larger than the previous transmission data volume R p , the number n of the proxy servers P1 to Pn can be increased. Also, according to the backup management apparatus 201, when the next transmission data volume wp+1 is smaller than the previous transmission data volume R p the number n of proxy servers P1 to Pn can be decreased.

[0204] Thus, when there is a possibility that the transmission of backup data will not be completed within the allowable time e at the next (p + 1)-th transmission, the resources used can be increased to prevent the backup time from becoming long. Also, when there is a possibility that waste will occur in the resources used at the next (p + 1)-th transmission, the resources used can be decreased to improve the efficiency of the resources used.

[0205] Also, according to the backup management device 201, when the next transmission data volume w p+1 is larger than the previous transmission data volume R p the number x of proxy servers to be increased is calculated based on the result of subtracting the previous transmission data volume R p+1 from the next transmission data volume w p and the previous assumed transmission data volume L p and the number n of proxy servers P1 to Pn can be increased by the calculated number x.

[0206] Thus, not only the data volume (R p ) actually transmitted to the backup storage ST but also the data volume (L p ) not transmitted due to the deduplication function are considered, and the number x of the insufficient proxy servers required at the next transmission can be calculated.

[0207] Also, according to the backup management device 201, the minimum transmission data volume S 1p ~S np among the previous (p-th) transmission data volumes S ip _1 acquired from each proxy server Pi is specified, and the result of subtracting the previous transmission data volume R p+1 from the next transmission data volume w p and the previous assumed transmission data volume L p is divided by the specified minimum transmission data volume S ipBy dividing by _1, the number x of proxy servers to be increased can be calculated.

[0208] Thus, based on the previous minimum transmission data volume S ip _1, by estimating the next transfer speed, it is possible to prevent a shortage of resources from occurring during the next transmission.

[0209] Also, according to the backup management device 201, when the number n of proxy servers P1 to Pn is increased, among the increased proxy servers P1 to Pn, the size of the data assigned to the proxy servers before the increase (the first transfer data size T p+1 ) can be set to the same size as the data assigned previously (transfer data size T p ). Also, according to the backup management device 201, the size of the data assigned to the increased proxy servers (the second transfer data size T p+1 ) can be set to the transmission data volume S ip _1.

[0210] Thus, the next transfer data size T p+1 of each proxy server Pi after the increase can be adjusted so that the data for the next transmission data volume w p+1 can be transmitted within the allowable time e.

[0211] Also, according to the backup management device 201, when the next transmission data volume w p+1 is smaller than the previous transmission data volume R p , based on the difference between the previous transmission data volume R p and the next transmission data volume w p+1 , the number y of proxy servers to be decreased can be calculated, and the number n of proxy servers P1 to Pn can be decreased by the calculated number y.

[0212] Thus, based on the difference between the previous transmission data volume R p and the next transmission data volume w p+1 , the next transmission data volume w p+1 can be within the allowable time e.It is possible to calculate the number y of proxy servers to be reduced so that the data can be transmitted and the resources used are minimized as much as possible.

[0213] Also, according to the backup management device 201, when the number n of proxy servers P1 to Pn is reduced, the size of the data (transfer data size T p+1 ) assigned to the reduced proxy server Pi is set to the same size as the data assigned last time (transfer data size T p ).

[0214] As a result, the next transfer data size T p+1 of each reduced proxy server Pi can be adjusted so that the data for the next transmission amount w p+1 can be transmitted within the allowable time e.

[0215] Also, according to the backup management device 201, by subtracting the transmission data amount (S ip ) regarded as the data obtained from each proxy server Pi and the transmission data amount (L ip ) from the data size W of the backup target data, the remaining data amount D remp of the backup target data that has not been transferred can be calculated. And according to the backup management device 201, based on the calculated remaining data amount D remp and the remaining number of transmissions, the next transmission data amount w p+1 can be calculated.

[0216] As a result, the next transmission data amount w remp can be calculated so that the remaining backup target data (remaining data amount D p+1 ) is evenly distributed and transferred each time.

[0217] From these, according to the backup management device 201 according to the embodiment, it is possible to optimize the resources used (the number n of proxy servers P1 to Pn, the network line) so that the backup is completed within the assumed time, and perform the backup of the backup target data.

[0218] Note that the backup management method described in this embodiment can be realized by executing a pre-prepared program on a computer such as a personal computer or a workstation. This backup management program is recorded on a computer-readable recording medium such as a hard disk, a floppy disk, a CD-ROM, a DVD, or a USB memory, and is executed by being read from the recording medium by a computer. Further, this backup management program may be distributed via a network such as the Internet.

[0219] Also, the backup management device 201 (information processing device 101) described in this embodiment can also be realized by a specific-purpose IC such as a standard cell or a structured ASIC (Application Specific Integrated Circuit), or a PLD (Programmable Logic Device) such as an FPGA.

[0220] Regarding the above-described embodiment, the following additional remarks are disclosed.

[0221] (Supplementary Note 1) While transferring backup target data to a backup storage in multiple dispersions via network lines corresponding to respective data transfer devices of a plurality of data transfer devices, among the data previously assigned to each data transfer device, a first data amount transmitted to the backup storage and a second data amount not transmitted to the backup storage due to a deduplication function are acquired from each data transfer device. Based on the acquired first data amount, a previous transmission data amount transmitted to the backup storage by the plurality of data transfer devices is calculated. Based on the acquired first data amount, the second data amount, and the size of the backup target data, a next transmission data amount to be transmitted to the backup storage is calculated. Based on the result of comparing the calculated previous transmission data volume and the next transmission data volume, increase or decrease the number of the plurality of data transfer devices. A backup management program characterized by causing a computer to execute processing.

[0222] (Appendix 2) The process of increasing or decreasing the number of the plurality of data transfer devices is When the next transmission data volume is larger than the previous transmission data volume, increase the number of the plurality of data transfer devices. When the next transmission data volume is smaller than the previous transmission data volume, decrease the number of the plurality of data transfer devices. The backup management program according to Appendix 1, characterized by the above.

[0223] (Appendix 3) Based on the obtained second data volume, cause the computer to execute a process of calculating the previous assumed transmission data volume by the deduplication function. The process of increasing or decreasing the number of the plurality of data transfer devices is When the next transmission data volume is larger than the previous transmission data volume, calculate the first number of data transfer devices to be increased based on the result of subtracting the previous transmission data volume and the previous assumed transmission data volume from the next transmission data volume. Increase the number of the plurality of data transfer devices by the calculated first number. The backup management program according to Appendix 2, characterized by the above.

[0224] (Appendix 4) The process of calculating the first number is Calculate the first number by dividing the result of subtracting the previous transmission data volume and the previous assumed transmission data volume from the next transmission data volume by the minimum data volume among the obtained first data volumes. The backup management program according to Appendix 3, characterized by the above.

[0225] (Supplementary Note 5) When the number of the plurality of data transfer devices is increased, among the plurality of data transfer devices after the increase, the size of the data assigned to the data transfer device before the increase is set to the same size as the data assigned last time, and the size of the data assigned to the increased data transfer device is set to the minimum data amount. The backup management program according to Supplementary Note 4, wherein the computer is caused to execute the process.

[0226] (Supplementary Note 6) The process of increasing or decreasing the number of the plurality of data transfer devices is When the amount of data to be transmitted next is smaller than the amount of data transmitted last time, based on the difference between the amount of data transmitted last time and the amount of data to be transmitted next, a second number of data transfer devices to be decreased is calculated. The number of the plurality of data transfer devices is decreased by the calculated second number. The backup management program according to any one of Supplementary Notes 3 to 5, characterized in that.

[0227] (Supplementary Note 7) When the number of the plurality of data transfer devices is decreased, the size of the data assigned to the data transfer device after the decrease is set to the same size as the data assigned last time. The backup management program according to Supplementary Note 6, wherein the computer is caused to execute the process.

[0228] (Supplementary Note 8) The process of calculating the amount of data to be transmitted next is By subtracting the acquired first data amount and the second data amount from the size of the data to be backed up, the remaining data amount of the data to be backed up that has not been transferred is calculated. Based on the calculated remaining data amount and the remaining number of transmissions, the amount of data to be transmitted next is calculated. The backup management program according to Supplementary Note 1, characterized in that.

[0229] (Appendix 9) While transferring backup target data to a backup storage in multiple dispersions via network lines corresponding to respective data transfer devices of a plurality of data transfer devices, among the data previously assigned to each of the data transfer devices, a first data amount transmitted to the backup storage and a second data amount not transmitted to the backup storage due to a deduplication function are acquired from each of the data transfer devices. Based on the acquired first data amount, a previous transmission data amount transferred to the backup storage by the plurality of data transfer devices is calculated. Based on the acquired first data amount, the second data amount, and the size of the backup target data, a next transmission data amount to be transferred to the backup storage is calculated. Based on the result of comparing the calculated previous transmission data amount and the next transmission data amount, the number of the plurality of data transfer devices is increased or decreased. A backup management method characterized in that a computer executes the process.

[0230] (Appendix 10) While transferring backup target data to a backup storage in multiple dispersions via network lines corresponding to respective data transfer devices of a plurality of data transfer devices, among the data previously assigned to each of the data transfer devices, a first data amount transmitted to the backup storage and a second data amount not transmitted to the backup storage due to a deduplication function are acquired from each of the data transfer devices. Based on the acquired first data amount, a previous transmission data amount transferred to the backup storage by the plurality of data transfer devices is calculated. Based on the acquired first data amount, the second data amount, and the size of the backup target data, a next transmission data amount to be transferred to the backup storage is calculated. Based on the result of comparing the calculated previous transmission data amount and the next transmission data amount, the number of the plurality of data transfer devices is increased or decreased. An information processing apparatus characterized by having a control unit.

Description of Symbols

[0231] 101 Information processing device 102 Data transfer device 103, ST backup storage 110, 1000 Backup target data 111 Data 200 Backup system 201 Backup management device 202 Client device 203 Administrator terminal 204 Host server 210 Network 220 Primary storage drive 230 Data management table 300 Bus 301 CPU 302 Memory 303 Disk drive 304 Disk 305 Communication I / F 306 Removable recording medium I / F 307 Removable recording medium 501 Reception unit 502 Backup control unit 503 Collection unit 504 Division unit 505 Allocation unit 506 Acquisition unit 507, 508 Calculation unit 509 Resource control unit 601 Backup data collection module 602 Backup data division module 603 Calculation module 604 Scheduler P1~Pn, Pi Proxy server

Claims

1. While transferring backup target data to be backed up using the plurality of data transfer devices to a backup storage in multiple portions via network lines corresponding to each of the plurality of data transfer devices, among the data previously assigned to each data transfer device, the first data amount transmitted to the backup storage and the second data amount not transmitted to the backup storage due to the deduplication function are acquired from each data transfer device, Based on the first data amount acquired from each data transfer device, the previous transmission data amount transmitted to the backup storage by the plurality of data transfer devices is calculated, Based on the first data amount, the second data amount, and the size of the backup target data acquired from each data transfer device, the next transmission data amount scheduled to be transmitted to the backup storage is calculated, Based on the result of comparing the calculated previous transmission data amount and the next transmission data amount, the number of the plurality of data transfer devices is increased or decreased, A backup management program characterized by causing a computer to execute the process.

2. The process of increasing or decreasing the number of the plurality of data transfer devices is When the next transmission data amount is larger than the previous transmission data amount, increasing the number of the plurality of data transfer devices, When the next transmission data amount is smaller than the previous transmission data amount, decreasing the number of the plurality of data transfer devices, The backup management program according to claim 1, characterized by the above.

3. Causing the computer to execute a process of calculating the previous assumed transmission data amount by the deduplication function based on the acquired second data amount, The process of increasing or decreasing the number of the plurality of data transfer devices is When the next transmission data volume is larger than the previous transmission data volume, based on the result of subtracting the previous transmission data volume and the previous assumed transmission data volume from the next transmission data volume, calculate the first number of data transfer devices to be increased, Increase the number of the plurality of data transfer devices by the calculated first number, The backup management program according to claim 2, characterized in that.

4. The process of calculating the first number is, Calculate the first number by dividing the result of subtracting the previous transmission data volume and the previous assumed transmission data volume from the next transmission data volume by the minimum data volume among the first data volumes obtained from the respective data transfer devices, The backup management program according to claim 3, characterized in that.

5. When increasing the number of the plurality of data transfer devices, among the plurality of data transfer devices after the increase, set the size of the data assigned to the data transfer device before the increase to the same size as the data assigned previously, and set the size of the data assigned to the increased data transfer device to the minimum data volume, The backup management program according to claim 4, characterized in that the process is executed by the computer.

6. The process of increasing or decreasing the number of the plurality of data transfer devices is, When the next transmission data volume is smaller than the previous transmission data volume, based on the difference between the previous transmission data volume and the next transmission data volume, calculate the second number of data transfer devices to be decreased, Decrease the number of the plurality of data transfer devices by the calculated second number, The backup management program according to any one of claims 3 to 5, characterized in that.

7. When reducing the number of the plurality of data transfer devices, set the size of the data assigned to the data transfer device after the reduction to the same size as the previously assigned data. The backup management program according to claim 6, wherein the computer is caused to execute the processing.

8. During the transfer of backup target data to be backed up using the plurality of data transfer devices to a backup storage in multiple portions via network lines corresponding to the respective data transfer devices of the plurality of data transfer devices, among the data previously assigned to each of the data transfer devices, obtain from each of the data transfer devices a first data amount transmitted to the backup storage and a second data amount not transmitted to the backup storage due to a deduplication function. Calculate the previous transmission data amount transmitted to the backup storage by the plurality of data transfer devices based on the first data amount obtained from each of the data transfer devices. Calculate the next transmission data amount to be transmitted to the backup storage based on the first data amount, the second data amount, and the size of the backup target data obtained from each of the data transfer devices. Increase or decrease the number of the plurality of data transfer devices based on the result of comparing the calculated previous transmission data amount and the next transmission data amount. A backup management method, characterized in that a computer executes the processing.

9. During the transfer of backup target data to be backed up using the plurality of data transfer devices to a backup storage in multiple portions via network lines corresponding to the respective data transfer devices of the plurality of data transfer devices, among the data previously assigned to each of the data transfer devices, obtain from each of the data transfer devices a first data amount transmitted to the backup storage and a second data amount not transmitted to the backup storage due to a deduplication function. Based on the first data amount acquired from each of the data transfer devices, calculate the previous transmission data amount transmitted to the backup storage by the plurality of data transfer devices. Based on the first data amount, the second data amount, and the size of the backup target data acquired from each of the data transfer devices, calculate the next transmission data amount scheduled to be transmitted to the backup storage. Based on the result of comparing the calculated previous transmission data amount and the next transmission data amount, increase or decrease the number of the plurality of data transfer devices. An information processing apparatus characterized by having a control unit.

Images