JP2007122167A - Data transfer program, data transfer method, and data transfer unit - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To perform high speed file transfer by using the disk of a disk device between servers connected through a disk device in a SAN(Storage Area Network) environment. <P>SOLUTION: A disk use circumstance acquiring part 120 acquires the number of partitions available for transfer by confirming the use circumstances of each intermediate disk to be used for the transfer of a file, and a file dividing part 130 equally divides transfer files by a number calculated by totaling the number of available partitions of each intermediate disk, and applies file descriptors to the transfer files, and a file transfer part 140 generates transfer processes whose number is equal to the number of the available partitions of the intermediate disks, and successively reads and transfers in parallel data by using each of the file descriptors of the transfer files as a reference point. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a data transfer program, a data transfer method, and a data transfer device for transferring a file from a server to a server via a disk of a disk array device connected in a SAN environment. The present invention relates to a data transfer program, a data transfer method, and a data transfer apparatus capable of transferring data at high speed by transferring in parallel via a plurality of transfer paths of a plurality of disks provided in the apparatus.

  Conventionally, in the backup of data stored in a business server, a file transfer by FTP connected to a backup server via a LAN has been mainstream. However, file transfer by FTP over LAN causes problems such as a heavy load on the CPU of the server and security risks such as data falsification and data leakage due to unauthorized access by a third party. .

  Therefore, the business server and backup server are connected via a disk array device in a SAN environment, and file transfer is performed using the disk installed in the disk array device, reducing the load on the server CPU. In addition, an application that can perform file transfer that ensures security in terms of security has been commercialized (for example, see Non-Patent Document 1).

"Softek XL data mover features", [October 6, 2005 search], Internet <URL: http://storage-system.fujitsu.com/jp/softek/products/xldatamover/merit/>

  However, in the conventional file transfer using the disk of the disk array device in the SAN environment, the file transfer is performed through one transfer path of one disk. Therefore, when transferring a large amount of data collectively, There is a problem that a long transfer time is required depending on the amount.

  In recent years, as the capacity of disks has increased, the business data stored on the server has been increasing, and the time for backing up these business data has been increasing day by day. There is a need for a system that can transfer.

  The present invention has been made to solve the above-described problems caused by the prior art. The data to be transferred is divided and transferred in parallel via a plurality of transfer paths of a plurality of disks provided in the disk array device. It is an object of the present invention to provide a data transfer program, a data transfer method, and a data transfer apparatus that can transfer data at high speed.

  In order to solve the above-described problems and achieve the object, a data transfer program according to the invention of claim 1 transfers data from a first server to a second server via a disk of a disk device connected in a SAN environment. A data transfer program for transferring data from the first server to the disk at the time of transfer, the transfer data dividing procedure for dividing the data to be transferred into a plurality, and the transfer data dividing by generating a plurality of transfer processes A parallel transfer procedure for transferring transfer data divided according to the procedure to the disk in parallel is caused to be executed by a first server.

  According to the first aspect of the present invention, the data to be transferred is divided into a plurality of parts, a plurality of transfer processes are generated, and the divided transfer data is transferred in parallel to the disk of the disk device. Time can be shortened.

  According to a second aspect of the present invention, there is provided a data transfer program according to the first aspect of the invention, wherein the transfer multiplicity determination procedure is to determine the transfer multiplicity of data by obtaining the number of usable partitions from a disk used for data transfer. Is further executed by the first server, and the transfer data dividing procedure divides the data to be transferred into a plurality of data based on the transfer multiplicity determined by the multiplicity determining procedure.

  According to the invention of claim 2, the number of usable partitions is obtained from the disk used for data transfer to determine the data transfer multiplicity, and the data to be transferred is divided into a plurality of data based on the determined transfer multiplicity. Thus, the transfer data can be divided according to the unused section of the disk used for data transfer, and the transfer can be performed efficiently.

  According to a third aspect of the present invention, in the data transfer program according to the second aspect of the present invention, the transfer multiplicity determining procedure obtains the number of usable partitions from a plurality of disks and transfers the data multiplicity for each disk. If there is a difference in the transfer performance of each disk, a part of the transfer multiplicity determined to be a disk with poor performance is transferred to a disk with good performance.

  According to the invention of claim 3, the number of usable partitions is obtained from a plurality of disks, the data transfer multiplicity is determined for each disk, and if there is a difference in the transfer performance of each disk, the disk with poor performance Since a part of the determined transfer multiplicity is transferred to a disk with good performance, the transfer multiplicity is distributed in a balanced manner to each disk used for data transfer, and data can be transferred efficiently.

  According to a fourth aspect of the present invention, there is provided a data transfer method for transferring data from a first server to a second server via a disk of a disk device connected in a SAN environment. A data transmission step of dividing the data into a plurality of pieces and transmitting the data from the first server to the disk of the disk device in parallel; and the data transmitted to the disk of the disk device by the data transmission step in parallel to the second server And a data receiving step of combining the transferred data.

  According to the invention of claim 4, the data to be transferred is divided into a plurality of pieces and transmitted in parallel from the first server to the disk of the disk device, and the transmitted data is transferred in parallel to the second server and transferred. Since the data is combined, the time required for data transfer can be shortened.

  According to a fifth aspect of the present invention, there is provided a data transfer device that transfers data from a first server to a second server via a disk of a disk device connected in a SAN environment. A data transfer apparatus for transferring data to a disk, the transfer data dividing means for dividing the data to be transferred into a plurality of data, and the transfer data generated by the transfer data dividing means by generating a plurality of transfer processes And parallel transfer means for transferring data in parallel.

  According to the fifth aspect of the present invention, the data to be transferred is divided into a plurality of parts, a plurality of transfer processes are generated, and the divided transfer data is transferred in parallel to the disk of the disk device. Time can be shortened.

  According to the first, fourth, and fifth aspects of the present invention, the time required for data transfer can be shortened, so that the effect that data can be transferred at high speed is achieved.

  According to the second aspect of the present invention, transfer data can be divided and transferred efficiently according to unused sections of a disk used for data transfer, so that data can be transferred at high speed. There is an effect.

  Further, according to the invention of claim 3, the transfer multiplicity is distributed in a balanced manner to each disk used for data transfer, and the data can be transferred efficiently, so that the data can be transferred at high speed. Play.

  Exemplary embodiments of a data transfer program, a data transfer method, and a data transfer apparatus according to the present invention will be explained below in detail with reference to the accompanying drawings.

  First, the concept of the data transfer system according to the first embodiment will be described. FIG. 1 is an explanatory diagram for explaining the concept of the data transfer system according to the first embodiment. As shown in the figure, in this data transfer system, a transfer file stored on a disk of a sending server is divided into a plurality of files, and a plurality of intermediate disks of a disk array device connected by a SAN (Storage Area Network). Are transferred in parallel to the receiving server.

  Here, the division number of the file to be transferred is determined based on the number of partitions usable in each intermediate disk at the time of file transfer. The number of partitions on each intermediate disk is defined at the time of formatting. In the example shown in FIG. 1, the intermediate disks A and B of the disk array apparatus are set as disks used for file transfer, and five partitions are defined for each. Since the sections A1 and A5 are in use, eight sections of the sections A2 to A4 of the intermediate disk A and the sections B1 to B5 of the intermediate disk B can be used. Therefore, the transfer file is divided into eight files.

  Further, the divided transfer files are transferred to different sections of the intermediate disk by a transfer process in which the same number of files are generated by the data transfer program executed on the transmission side server. In the example of FIG. 1, the transfer file is divided into eight files, and transferred to the sections A2 to A4 of the intermediate disk A and the sections B1 to B5 of the intermediate disk B, respectively, by eight transfer processes.

  Transfer files transferred for each partition of each intermediate disk are received by the receiving server in parallel by a receiving process that is generated by the data receiving program executed by the receiving server in the same number as the number of file divisions. , And stored in the received file on the receiving disk so that it is in the same order as the divided order.

  Here, for convenience of explanation, two disks (intermediate disks A and B) are set as intermediate disks used for file transfer. However, any of the plurality of disks provided in the disk array device is set as an intermediate disk. Whether to use as a disk can be set in advance according to each environment.

  Also, here, for convenience of explanation, the case where there is one transmission server is shown, but a case where a plurality of transmission servers are connected via a SAN and each transfers a file to one reception server. There is also.

  As described above, in the data transfer system according to the first embodiment, the file to be transferred is divided in accordance with the number of usable partitions of the intermediate disk used for file transfer, and each of them uses a plurality of partitions of a plurality of intermediate disks. By transferring in parallel, the time required for file transfer can be shortened.

  Next, the configuration of the data transfer system according to the first embodiment will be described. FIG. 2 is a functional block diagram illustrating the configuration of the data transfer system according to the first embodiment. As shown in the figure, this data transfer system includes a transmitting server 50 storing business data, a disk array device 60 equipped with an intermediate disk for data transfer, and receiving backup data for business data. The side server 70 is connected in the SAN environment.

  The data transfer program 100 executed by the transmission-side server 50 includes a disk information acquisition unit 110, a disk usage status acquisition unit 120, a file division unit 130, a file transfer unit 140, and a control unit 150.

  The disk information acquisition unit 110 is a processing unit that acquires disk information about an intermediate disk of the disk array device 60 from a disk information storage unit 210 of the receiving server 70 described later.

  The disk usage status acquisition unit 120 is a processing unit that checks the usage status of the intermediate disk of the disk array device 60 and acquires the number of partitions that can be used for transfer. Specifically, the disk usage status acquisition unit 120 extracts the disk number of the intermediate disk used for file transfer from the disk information acquired by the disk information acquisition unit 110. Then, based on the extracted disk numbers, the intermediate disks of the disk array device 60 are referred to, and the number of partitions that can be used for transfer is acquired for each intermediate disk.

  The file dividing unit 130 is a processing unit that divides a transfer file on the transmission side disk of the transmission side server 50 into units to be transferred. Specifically, the file dividing unit 130 acquires the size of the file to be transferred, and equally divides the transfer file into the total number of usable partitions for each intermediate disk acquired by the disk usage status acquisition unit 120. A file descriptor (file reading position) is assigned. Further, the size of the file to be transferred and the information of the file descriptor are transferred to the receiving server 70 as transfer file information.

  The file transfer unit 140 is a processing unit that transfers the transfer file in which the file descriptor is set by the file dividing unit 130 to each usable partition of the intermediate disk used for transfer in parallel. Specifically, the file transfer unit 140 generates the same number of transfer processes as the total number of usable partitions for each intermediate disk acquired by the disk usage status acquisition unit 120, and transfers the file. Each transfer process sequentially reads data starting from another file descriptor set in the transfer file, and transfers the data in parallel to each usable partition of the intermediate disk.

  The file transfer unit 140 generates a plurality of transfer processes and transfers data in parallel with each file descriptor set by the file dividing unit 130 as a base point, thereby transferring the transfer file at high speed.

  The control unit 150 is a processing unit that controls the disk information acquisition unit 110, the disk usage status acquisition unit 120, the file division unit 130, and the file transfer unit 140. Specifically, the control unit 150 performs control between the functional units. By performing the movement, the data transfer program 100 is caused to function as one program.

  The data receiving program 200 executed by the receiving server 70 includes a disk information storage unit 210 and a file receiving unit 220.

  The disk information storage unit 210 is a storage unit that stores information on an intermediate disk used for file transfer. Specifically, the disk information storage unit 210 stores a disk number of a disk used as an intermediate disk in file transfer among a plurality of disks provided in the disk array device 60.

  The file receiving unit 220 is a processing unit that receives, in parallel, transfer files transferred in a divided manner from the transmission side server 50 for each usable partition of the intermediate disk, and stores them in the reception side disk of the reception side server 70. . Specifically, the file receiving unit 220 periodically monitors each intermediate disk of the disk array device 60 and, when detecting the transfer file information, receives the transfer file information, and from the received transfer file information Extract the size of the transfer file and the file descriptor information. Then, a receiving file having the same size as the size of the extracted transfer file is secured on the receiving disk. Further, a file descriptor similar to the transfer file is assigned to the reception file based on the extracted file descriptor information.

  Further, when the file receiving unit 220 detects data of the transfer file divided and transferred from the transmission side server 50, the file reception unit 220 generates a reception process for receiving the transfer file. The reception process receives the data of the detected transfer file, and sequentially stores the data in the reception file based on the file descriptor corresponding to the transfer file. A plurality of reception processes are generated for each section of the intermediate disk, and each is executed in parallel.

  Next, a processing procedure of the data transfer program 100 according to the first embodiment will be described. FIG. 3 is a flowchart illustrating the processing procedure of the data transfer program 100 according to the first embodiment. As shown in the figure, in this data transfer program 100, the disk information acquisition unit 110 acquires disk information from the disk information storage unit 210 of the receiving server 70 (step S101).

  Then, the disk usage status acquisition unit 120 extracts the disk number of the intermediate disk used for transfer from the disk information (step S102), and the usage status of each intermediate disk of the disk array device 60 based on the extracted disk number. The number of partitions that can be used for transfer is acquired (step S103).

  Then, the transfer file dividing unit 130 acquires the size of the transfer file, equally divides the number of usable partitions of each intermediate disk by the total number, and assigns a file descriptor (step S104).

  Then, the file transfer unit 140 generates the same number of transfer processes as the number of usable partitions of the intermediate disk, and each sequentially reads data from another file descriptor of the transfer file as a base point and transfers the data in parallel (step S105). .

  As described above, the file transfer unit 140 transfers the file in parallel to the unused partition of each intermediate disk used for the file transfer, thereby reducing the time required for the file transfer.

  As described above, in the first embodiment, the disk usage status acquisition unit 120 checks the usage status of each intermediate disk used for file transfer, acquires the number of partitions usable for transfer, and divides the transfer file The unit 130 equally divides the transfer file by the total number of usable partitions of each intermediate disk and assigns file descriptors, and the file transfer unit 140 performs the same number of transfer processes as the number of usable partitions of the intermediate disk. The data transfer program 100 transfers the files at high speed from the transmission side server 50 to the reception side server 70 because the data is sequentially read and transferred in parallel with each file descriptor of the transfer file as a base point. be able to.

  By the way, in the first embodiment, the case has been described where data is transferred by generating the same number of transfer processes as the number of partitions that can be used for transferring the intermediate disk. However, in day-to-day operations, file transfer by a plurality of users may be concentrated depending on the day of the week or the time zone, and there may be contention between partitions on the intermediate disk. In addition, when a disk is added to the disk array device, the transfer performance may differ between the intermediate disks used for transfer. Therefore, in the second embodiment, a case will be described in which the multiplicity of transfer is determined according to the use state of the intermediate disk according to the day of the week and the time zone for transferring the file and the transfer performance between the intermediate disks.

  First, a method for determining the multiplicity of the data transfer system according to the second embodiment will be described. In the data transfer system according to the second embodiment, the overall transfer multiplicity in the current file transfer is determined based on the transfer time average of the file transfer performed on the same day and the same time period in the past predetermined period. Furthermore, based on the transfer rate of each intermediate disk used for transfer, by reducing the multiplicity assigned to the intermediate disk with poor transfer performance, and increasing the multiplicity assigned to the intermediate disk with excellent transfer performance, Optimize transfer multiplicity allocation.

  FIG. 4 is an explanatory diagram for explaining the multiplicity determination method of the data transfer system according to the second embodiment. As shown in the figure, in this data transfer system, intermediate disks C, D, and E each defined with seven partitions are used for file transfer. Here, it is assumed that all the partitions of the intermediate disks C and D are unused, and that one partition E7 is in use for the intermediate disk E. In addition, the transfer rate of each intermediate disk is 11 MB / sec, 6 MB / sec, and 4 MB / sec, respectively.

  The transfer multiplicity of each intermediate disk is determined based on a preset reference value of the transfer multiplicity assigned to each intermediate disk in one file transfer. This reference value is set as appropriate according to the system environment, and in the example shown in FIG.

  If each intermediate disk has a partition in use, a value obtained by subtracting the number of partitions in use from the reference value of the transfer multiplicity is set as the transfer multiplicity to be assigned to each intermediate disk. In the example shown in FIG. 4, the transfer multiplicity of the intermediate disks C, D, and E is 5 multiplex, 5 multiplex, and 4 multiplex, respectively, based on the usage status of each intermediate disk.

  Further, based on the average transfer time of the file transfer performed on the same day and the same time period in the past predetermined period, the overall transfer multiplicity used for the current transfer is determined. First, the average transfer time is compared with a preset reference transfer time. If the transfer average time exceeds the reference transfer time, the average transfer time is used as a reference until the average transfer time becomes smaller than the reference transfer time. The transfer time is repeatedly deducted, and each time the multiplicity of each intermediate disk is reduced by one.

  In the example shown in FIG. 4, it is assumed that the average transfer time per time zone of the same time zone on the same day in the past four weeks is 13 minutes, and the reference time for calculating the transfer multiplicity is 10 minutes. As a result, the transfer multiplicity of the disks C, D, and E in the example shown in FIG. 4 is 4 multiplex, 4 multiplex, and 3 multiplex, respectively.

  Furthermore, the assignment of the transfer multiplicity to each intermediate disk is optimized based on the transfer performance of each intermediate disk. First, the transfer rate of the disc with the highest transfer rate is compared with the transfer rate of the disc with the lowest transfer rate. If the difference between the transfer rates exceeds the preset reference transfer rate, the transfer rate is One multiplicity is transferred from the disk having the lowest transfer rate to the disk having the highest transfer rate.

  In the example shown in FIG. 4, the reference transfer rate is 4 MB / sec. Also, the intermediate disk C with the highest transfer rate and. When the transfer rates of the intermediate disk E having the lowest transfer rate are compared, the difference is 7 MB / sec, which exceeds the reference transfer rate of 4 MB / sec. One multiplicity is transferred to the disk C (in the example shown in the figure, the transfer assigned to the section E4 of the intermediate disk E is transferred to the section C3 of the intermediate disk C). As a result, the transfer multiplicity of the disks C, D, and E is 5 multiplexing, 4 multiplexing, and 2 multiplexing, respectively.

  Based on the transfer multiplicity for each intermediate disk determined in this way, the transfer file on the transmission-side disk of the transmission-side server is divided into the same number, and the data is stored in parallel for each usable partition of the intermediate disk of the disk array device Forward.

  As described above, in the data transfer system according to the second embodiment, the overall transfer multiplicity in the current file transfer is calculated based on the transfer time average of the file transfer performed on the same day and the same time period in the past predetermined period. By determining, it is possible to suppress a decrease in the overall transfer efficiency including other file transfers, and to perform file transfer efficiently.

  In the data transfer system according to the second embodiment, the transfer multiplicity of each disk is optimized by optimizing the assignment of the transfer multiplicity to each intermediate disk based on the transfer rate of each intermediate disk used for transfer. Data is distributed in a well-balanced manner and data can be transferred efficiently.

  Next, the configuration of the data transfer system according to the second embodiment will be described. FIG. 5 is a functional block diagram of the configuration of the data transfer system according to the second embodiment. Here, for convenience of explanation, functional units that play the same functions as the respective units shown in FIG.

  As shown in FIG. 5, this data transfer system includes a transmitting server 50 storing business data, a disk array device 60 equipped with an intermediate disk for data transfer, and receiving backup data for business data. The side server 70 is connected via a SAN.

  The data transfer program 300 executed by the transmission server 50 includes a disk information acquisition unit 310, a disk usage status acquisition unit 120, a file division unit 130, a file transfer unit 340, a control unit 350, and a transfer multiplicity setting. Part 360.

  The disk information acquisition unit 310 is a processing unit that acquires disk information about an intermediate disk used for file transfer stored in a disk information storage unit 410 of the receiving server 70 described later. FIG. 6 is a diagram illustrating an example of disc information according to the second embodiment. As shown in the figure, this disc information includes the difference between the disc number of the disc used as an intermediate disc in file transfer, the transfer rate of each disc, and the transfer rate used as a reference when transferring the transfer multiplicity. The transfer multiplicity for one disk, the reference time for reducing the transfer multiplicity, and the transfer time for each predetermined time zone for each day of the week for file transfer performed in the past predetermined period (in FIG. (Transfer time is set at 30-minute intervals for each day of the week for file transfer performed in 4 weeks), file transfer start date and time (transfer start date and time), and last file transfer start time ( Last transfer date).

  Returning to the explanation of FIG. 5, the file transfer unit 340 is a processing unit that transfers the transfer file divided by the file dividing unit 130 in parallel to each usable partition of the intermediate disk used for transfer. Specifically, the file transfer unit 340 generates the same number of transfer processes as the total transfer multiplicity for each intermediate disk determined by the transfer multiplicity setting unit 360 described later. Then, these transfer processes sequentially read data from the transfer file based on different file descriptors, and transfer them in parallel to each usable partition of the intermediate disk.

  The control unit 350 is a processing unit that controls the disk information acquisition unit 310, the disk usage status acquisition unit 120, the file division unit 130, the file transfer unit 340, and the transfer multiplicity setting unit 360. Causes the data transfer program 300 to function as one program by transferring control between the functional units.

  The transfer multiplicity setting unit 360 is a processing unit that determines the transfer multiplicity for each intermediate disk of the disk array device 60. Specifically, the transfer multiplicity setting unit 360 extracts the transfer multiplicity for one disk from the disk information acquired by the disk information acquisition unit 310. The extracted transfer multiplicity is compared with the number of usable partitions for each intermediate disk acquired by the disk usage status acquisition unit 310, and the smaller value is set as the transfer multiplicity for each intermediate disk.

  Further, the transfer multiplicity setting unit 360 extracts, from the disk information, the average transfer time α (min) for the same day and the same time period in a predetermined period and the time β (min) that serves as a reference for reducing the transfer multiplicity. . Here, if α−β ≧ 0, the transfer multiplicity for each intermediate disk is reduced by one, and the value obtained by subtracting β from α is set as α, and the transfer number is increased until α−β <0. Decrease the severity one by one.

  Thereafter, the transfer multiplicity setting unit 360 determines, based on the disk information, the transfer rate x (MB / sec) of the intermediate disk having the highest transfer rate, the transfer rate y (MB / sec) of the intermediate disk having the lowest transfer rate, and A transfer rate difference d (MB / sec) which is a reference for transferring the transfer multiplicity is extracted. Here, when xy> d, a value obtained by transferring one transfer multiplicity from each intermediate disk having the lowest transfer rate to the intermediate disk having the highest transfer rate and subtracting d from x Is set as x, a value obtained by adding d to y is set as y, and the transfer multiplicity is changed one by one until x−y ≦ d.

  As described above, the transfer multiplicity setting unit 360 determines the transfer time average of the file transfer performed on the same day of the past predetermined period and the same time zone based on the transfer multiplicity with respect to one predetermined disk. The transfer multiplicity is adjusted based on the transfer rate of each intermediate disk used for transfer, and finally the transfer multiplicity for each intermediate disk is determined.

  The transfer multiplicity setting unit 360 assigns transfer multiplicity to each intermediate disk in a well-balanced manner, so that each intermediate disk can be efficiently used to perform file transfer.

  The data receiving program 400 executed by the receiving server 70 includes a disk information storage unit 410 and a file receiving unit 420. The disk information storage unit 410 is a storage unit that stores the above-described disk information.

  The file receiving unit 420 is a processing unit that receives, in parallel, transfer files transferred in a divided manner from the transmission side server 50 to each usable partition of the intermediate disk and stores them in the reception side disk of the reception side server 70. . Specifically, the file receiving unit 420 periodically monitors each intermediate disk of the disk array device 60 and, when detecting the transfer file information, receives the transfer file information, and receives the transfer file information from the received transfer file information. Extract the size of the transfer file and the file descriptor information. Then, a receiving file having the same size as the size of the extracted transfer file is secured on the receiving disk. Further, a file descriptor similar to the transfer file is assigned to the reception file based on the extracted file descriptor information.

  In addition, when the file receiving unit 420 detects data of a transfer file divided and transferred from the transmission-side server 50, the file reception unit 420 generates a reception process for receiving the transfer file. This reception process receives the data of the detected transfer file, and sequentially stores the data in the received file based on the file descriptor corresponding to the received transfer file. A plurality of reception processes are generated for each section of the intermediate disk, and each is executed in parallel.

  Further, the file receiving unit 420 updates the transfer start date and time of the disc information stored in the disc information storage unit 410 at the timing when reception of one transfer file starts, using the date and time at that time. In addition, the file receiving unit 420 updates the last transfer date and time of the disk information using the transfer start date and time at the timing when reception of one file is completed. At the same timing, the transfer rate is calculated based on the transfer record for each intermediate disk used for transfer, and the transfer rate of each intermediate disk in the disk information is updated. Further, the time required for data transfer is added to the transfer time of the week, day of the week, and time zone corresponding to the last transfer date and time of the disk information.

  Next, a processing procedure of the data transfer program 300 according to the second embodiment will be described. FIGS. 7A and 7B are flowcharts (1) and (2) illustrating the processing procedure of the data transfer program 300 according to the second embodiment. As shown in the figure, in the data transfer program 300, the disk information acquisition unit 310 acquires the disk information from the disk information storage unit 410 of the receiving server 70 (step S201).

  Then, the disk usage status acquisition unit 120 extracts the disk number of the intermediate disk used for transfer from the disk information file (step S202), and uses each intermediate disk of the disk array device 60 based on the extracted disk number. The status is confirmed, and the number of partitions that can be used for transfer is acquired (step S203).

  Then, the transfer multiplicity setting unit 360 compares the number of usable partitions of each intermediate disk with the transfer multiplicity for one disk in the disk information, and assigns the smaller value to each intermediate disk. (Step S204).

  Then, the transfer multiplicity setting unit 360 extracts, from the disk information, the average transfer time α (min) for the same day of the week and the same time period, and the time β (min) serving as a reference for reducing the transfer multiplicity. (Step S205, Step S206). If α−β ≧ 0 (step S207, Yes), 1 is subtracted from the transfer multiplicity of each intermediate disk (step S208), and a value obtained by subtracting β from α is set to α (step S209). ). Then, the process returns to step S207, and the processes of steps S208 and S209 are repeated until α−β <0.

  Thereafter, if α−β <0 (No in step S207), the file dividing unit 130 acquires the size of the transfer file and equally divides it by the same number as the total transfer multiplicity for each intermediate disk. A file descriptor is assigned (step S210).

  Also, the transfer multiplicity setting unit 360 determines from the disk information that the transfer rate x (MB / sec) of the intermediate disk having the highest transfer rate and the transfer rate y (MB / sec) of the intermediate disk having the lowest transfer rate. Are extracted (steps S211, S212). Further, the transfer multiplicity setting unit 360 obtains a transfer rate difference d (MB / sec) serving as a reference for transferring the transfer multiplicity from the disk information (steps S213 and S214).

  If xy> d (step S215, Yes), one transfer multiplicity is transferred from each intermediate disk having the lowest transfer rate to the intermediate disk having the highest transfer rate (step S216). . Further, a value obtained by subtracting d from x is set to x, and a value obtained by subtracting d from y is set to y (step S217). And it returns to step S215 and repeats the process of step S216 and S217 until it becomes x-y <= d.

  Thereafter, when x−y ≦ d is satisfied (step S215, No), the file transfer unit 340 generates the same number of transfer processes as the transfer multiplicity assigned to each intermediate disk, and each file of the transfer file Data is sequentially read starting from the descriptor and transferred in parallel (step S218).

  In this way, by dividing the transfer file based on the transfer multiplicity assigned to each intermediate disk in a balanced manner by the transfer multiplicity setting unit 360, and transferring the files in parallel to each intermediate disk, It is possible to transfer files using each intermediate disk efficiently.

  As described above, in the second embodiment, the transfer multiplicity setting unit 360 is performed on the same day and at the same time in the past predetermined period stored in the disk information storage unit 410 of the receiving server 70. Determine the overall transfer multiplicity for the current file transfer based on the average transfer time of the file transfer, and optimize the assignment of the transfer multiplicity to each intermediate disk based on the transfer rate of each intermediate disk used for transfer Therefore, file transfer can be performed efficiently using the intermediate disk, and file transfer can be performed at higher speed.

  In the first embodiment, the case has been described in which the use status of the intermediate disk is confirmed at the time when the file transfer is started, and the file is transferred using an unused partition at that time. However, while the file is being transferred, the partition that was in use at the start of the transfer may be released due to the completion of another file transfer, and the number of unused partitions may increase. Therefore, in the third embodiment, a tuning point is provided in the middle of the file transfer, the use status of the intermediate disk is confirmed for each tuning point, and if there are unused partitions, the files are also used using those partitions. The case of transferring will be described.

  First, the concept of the data transfer system according to the third embodiment will be described. FIG. 8 is an explanatory diagram for explaining the concept of the data transfer system according to the third embodiment. As shown in the figure, in this data transfer system, tuning points are set for each transmission file with a predetermined size, the range delimited by each tuning point is handled as one file, and sequential transfer is performed.

  Then, at the time of starting the transfer of each range, check the usage status of the intermediate disk, divide each range by the same number as the number of unused partitions at that time, and generate the same number of transfer processes in parallel Transfer data. Here, when the use status of the intermediate disk is confirmed, a partition that has already been used for transfer is secured as it is, and is subsequently used for transfer.

In the example shown in FIG. 8, when the range from the beginning of the transfer file to the tuning point 1 is transferred, the three sections F1, F2, and F3 of the intermediate disk F are unused at the time of starting the transfer. Thus, the range of the transfer file is divided into _three , and each of them is transferred in parallel by the _three transfer processes 31 ₁ , 31 _2, and 31 ₃ generated by the transmitting server.

  Then, at the timing when one of the transfer processes already executed, the transfer status of the intermediate disk is confirmed, and the number of unused partitions at that time and the number of partitions already used Divide the range up to the next tuning point of the transfer file by the total number of and start the transfer in parallel.

In the example shown in FIG. 8, when the range from tuning point 1 to tuning point 2 is transferred, the three partitions F1, F2 and F3 of the intermediate disk F are already used for transfer at the start of transfer, In addition, since the three sections F4, F5, and F6 are unused at that time, the range of the transfer file is divided into six, and six transfer processes 31 _{4 to} 31 ₉ generated by the transmitting server. Are transferred in parallel. The transfer process 31 _9, after waiting already to a running transfer process 31 ₂ is completed, the relative compartment F2, to transfer data. Thereafter, data transfer similar to the above is repeatedly performed until the end of the file is reached.

  As described above, in the data transfer system according to the third embodiment, by setting a tuning point, a timing for confirming the free state of the partition of the intermediate disk during file transfer is provided. If there is a partition that is in use, by adding this partition to the partition that is used for transfer, the intermediate disk can be efficiently used to perform file transfer.

  Next, the configuration of the data transfer system according to the third embodiment will be described. FIG. 9 is a functional block diagram of the configuration of the data transfer system according to the third embodiment. Here, for convenience of explanation, functional units that play the same functions as the respective units shown in FIG.

  As shown in FIG. 9, this data transfer system includes a transmission server 50 storing business data, a disk array device 60 equipped with an intermediate disk for data transfer, and reception for backup of business data. The side server 70 is connected via a SAN.

  The data transfer program 500 executed by the transmission side server 50 includes a disk information acquisition unit 110, a disk usage status acquisition unit 120, a file division unit 530, a file transfer unit 140, a control unit 550, and a tuning point setting unit. 570.

  The file dividing unit 530 is a processing unit that divides a transfer file in the transmission-side disk of the transmission-side server 50 into units to be transferred. Specifically, the file dividing unit 530 acquires the size of the transfer file for each range delimited by the tuning points set by the tuning point setting unit 570 described later, and is acquired by the disk usage status acquisition unit 120. The range of the transfer file is equally divided into the total number of usable partitions for each intermediate disk and the number of partitions already used for transfer, and a file descriptor (file reading position) is given. Further, the size of the entire file to be transferred and the information of the file descriptor are transferred to the receiving server 70 as transfer file information.

  The control unit 550 is a processing unit that controls the disk information acquisition unit 110, the disk usage status acquisition unit 120, the file division unit 530, the file transfer unit 140, and the tuning point setting unit 570. Specifically, The data transfer program 500 is caused to function as one program by transferring control between the functional units.

  The tuning point setting unit 570 is a processing unit that sets a tuning point for each predetermined size with respect to the transfer file on the transmission side disk of the transmission side server 50.

  The tuning point setting unit 570 sets a tuning point for the transfer file, thereby providing a timing for confirming the usage status of the intermediate disk. A partition released from another file transfer during the file transfer is set. , It can be added to the partition used for the file transfer.

  Next, a processing procedure of the data transfer program 500 according to the third embodiment will be described. FIG. 10 is a flowchart illustrating the processing procedure of the data transfer program 500 according to the third embodiment. As shown in the figure, in the data transfer program 500, the disk information acquisition unit 110 acquires disk information from the disk information storage unit 210 of the receiving server 70 (step S301).

  Further, the tuning point setting unit 570 sets a tuning point for each predetermined size for the transfer file on the disk of the transmission side server (step S302).

  Then, the disk usage status acquisition unit 120 extracts the disk number of the intermediate disk used for transfer from the disk information file (step S303), and uses each intermediate disk of the disk array device 60 based on the extracted disk number. The status is confirmed, and the number of partitions that can be used for transfer is acquired (step S304).

  Then, the file dividing unit 530 acquires the size of the transfer file for each range delimited by the tuning points, and adds the number of usable partitions of each intermediate disk and the number of partitions already used for transfer to the total number. The range is equally divided and a file descriptor is assigned. (Step S305).

  Then, the file transfer unit 140 generates the same number of transfer processes as the number of usable partitions on the intermediate disk, sequentially reads data using each file descriptor of the transfer file as a base point, and transfers it in parallel (step S306).

  If any of the transfer processes being executed is completed, the transfer process being executed is checked (step S307, Yes), and if all the transfer processes have been completed (step S308, Yes), processing is performed. Exit. If there is a transfer process that is still being executed (No in step S308), if the transfer process that has ended is the transfer process that ended first within the range of the transfer file being transferred at that time ( In step S309, Yes), the process returns to step S304 to check the use status of each intermediate disk, and the same transfer processing is performed for the range up to the next tuning point, and this is repeated until the end of the file is reached.

  In this way, the range delimited by each tuning point set by the tuning point setting unit 570 is treated as one file, and there is an unused partition released from other file transfer when starting transfer of each range. In this case, the partition of each intermediate disk can be used efficiently by adding to the partition used for file transfer.

  As described above, in the third embodiment, the tuning point setting unit 570 sets a tuning point for the transfer file on the transmission side disk of the transmission side server 50, and the disk is being transferred while the file is being transferred. The usage status acquisition unit 120 checks the usage status of each intermediate disk of the disk array device 60, and if there are partitions freed from other file transfers, the partitions used by the file transfer that is being executed. Therefore, it is possible to reduce the time for which unused sections are left unattended, efficiently use the intermediate disk, and perform file transfer, and perform file transfer at higher speed.

  Next, a computer (server) that executes the data transfer program 100 according to the first embodiment will be described. Here, the computer that executes the data transfer program 100 according to the first embodiment will be described, but the same applies to the data transfer program 300 according to the second embodiment and the data transfer program 500 according to the third embodiment. It can be executed on a computer having a configuration.

  FIG. 11 is a functional block diagram of the configuration of the computer that executes the data transfer program according to the first embodiment. As shown in the figure, the computer 600 includes a RAM 610, a CPU 620, an HDD 630, a LAN interface 640, an input / output interface 650, a DVD drive 660, and a SCSI interface 670.

  The RAM 610 is a memory that stores a program and a program execution result, and the CPU 620 is a central processing unit that reads and executes the program from the RAM 610.

  The HDD 630 is a disk device that stores programs and data, and the LAN interface 640 is an interface for connecting the computer 600 to other computers via the LAN.

  The input / output interface 650 is an interface for connecting an input device such as a mouse or a keyboard and a display device, and the DVD drive 660 is a device for reading / writing a DVD.

  The SCSI interface 670 is an interface for connecting the computer 600 to the disk array device via the SAN.

  The data transfer program 611 executed in the computer 600 is stored in the DVD, read from the DVD by the DVD drive 660, and installed in the computer 600.

  Alternatively, the data transfer program 611 is stored in a database or the like of another computer system connected via the LAN interface 640, read from these databases, and installed in the computer 600.

  The installed data transfer program 611 is stored in the HDD 630, read into the RAM 610, and executed as the data transfer process 621 by the CPU 620.

  In this embodiment, the multiplicity of transfer is determined according to the use status of the intermediate disk according to the day and time of the file transfer and the transfer performance between the intermediate disks (second embodiment), and the file transfer In the middle, a tuning point is provided to check the usage status of the intermediate disk for each tuning point. If there are unused partitions, those partitions are also used to transfer files (Example 3) separately. However, the present invention is not limited to this, and can be similarly applied to a case where the components are combined. In this case, the configuration is such that the transfer multiplicity described in the second embodiment is adjusted at the timing of checking the usage status of the intermediate disk for each tuning point described in the third embodiment.

(Supplementary Note 1) A data transfer program for transferring data from the first server to the disk when transferring data from the first server to the second server via a disk of a disk device connected in a SAN environment There,
Transfer data division procedure to divide the data to be transferred into multiple,
A parallel transfer procedure for generating a plurality of transfer processes and transferring the transfer data divided by the transfer data dividing procedure to the disk in parallel;
A data transfer program characterized by causing the first server to execute.

(Supplementary Note 2) The first server is further caused to execute a transfer multiplicity determination procedure for obtaining the number of usable partitions from the disk used for data transfer and determining the transfer multiplicity of the data,
The data transfer program according to appendix 1, wherein the transfer data division procedure divides data to be transferred based on the transfer multiplicity determined by the multiplicity determination procedure.

(Additional remark 3) The said transfer multiplicity determination procedure determines the time slot | zone in which a data transfer tends to concentrate based on the transfer time for every predetermined time slot | zone of the data transfer performed in the past, and the determined time slot | zone The data transfer program according to appendix 2, wherein the data transfer multiplicity is reduced.

(Supplementary Note 4) If the transfer multiplicity determination procedure is such that when the transfer time in a time zone in which the data transfer tends to concentrate exceeds a predetermined reference time, the transfer multiplicity is reduced in the time zone. The data transfer program according to Supplementary Note 3, which is a feature.

(Supplementary Note 5) The transfer multiplicity determination procedure obtains the number of usable partitions from a plurality of disks and determines the data transfer multiplicity for each disk. The data transfer program according to any one of appendices 2 to 4, wherein a part of the transfer multiplicity determined to be a bad disk is transferred to a disk with good performance.

(Supplementary Note 6) The transfer multiplicity determination procedure is such that if the difference in transfer performance between the disks exceeds a predetermined reference value, a part of the transfer multiplicity determined to be a disk with poor performance is used as a disk with good performance. The data transfer program according to appendix 5, wherein the program is transferred to

(Additional remark 7) Let the 1st server further perform the tuning point determination procedure which determines a tuning point at predetermined intervals with respect to the data to transfer,
The transfer multiplicity determination procedure obtains the number of usable partitions of the disk used for data transfer for each tuning point determined by the tuning point determination procedure to determine the data transfer multiplicity,
The data division procedure divides the range delimited by the tuning point of the transfer file into a plurality based on the transfer multiplicity determined by the transfer multiplicity determination procedure,
The parallel transfer procedure generates a plurality of transfer processes for each tuning point determined by the tuning point determination procedure, and transfers the data divided by the transfer data division procedure to the disk in parallel. The data transfer program according to any one of appendices 2 to 6.

(Supplementary note 8) The transfer multiplicity determination procedure is characterized in that the transfer multiplicity of the current data is determined by adding the number of partitions acquired at the current tuning point to the number of partitions acquired at the previous tuning point. The data transfer program according to appendix 7.

(Supplementary note 9) A data transfer method for transferring data from a first server to a second server via a disk of a disk device connected in a SAN environment,
A data transmission step of dividing the data to be transferred into a plurality of pieces and transmitting in parallel from the first server to the disk of the disk device;
A data reception step of transferring the data transmitted to the disk of the disk device by the data transmission step in parallel to a second server, and combining the transferred data;
A data transfer method comprising:

(Supplementary Note 10) Data transfer device for transferring data from the first server to the disk when transferring data from the first server to the second server via the disk of the disk device connected in the SAN environment Because
Transfer data dividing means for dividing the data to be transferred into a plurality of data;
Parallel transfer means for generating a plurality of transfer processes and transferring the transfer data divided by the transfer data dividing means to the disk in parallel;
A data transfer device comprising:

  As described above, the data transfer program, the data transfer method, and the data transfer apparatus according to the present invention are useful when transferring data stored in a business server or the like to a backup server. This is suitable for backing up a large amount of accumulated data at high speed.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is an explanatory diagram for explaining the concept of a data transfer system according to a first embodiment; 1 is a functional block diagram illustrating a configuration of a data transfer system according to a first embodiment. 3 is a flowchart illustrating a processing procedure of a data transfer program according to the first embodiment. FIG. 10 is an explanatory diagram for explaining a multiplicity determination method of the data transfer system according to the second embodiment. FIG. 6 is a functional block diagram illustrating a configuration of a data transfer system according to a second embodiment. It is a figure which shows an example of the disc information which concerns on the present Example 2. 12 is a flowchart (1) illustrating a processing procedure of a data transfer program according to the second embodiment. It is a flowchart (2) which shows the process sequence of the data transfer program which concerns on the present Example 2. It is explanatory drawing for demonstrating the concept of the data transfer system which concerns on the present Example 3. FIG. FIG. 10 is a functional block diagram illustrating a configuration of a data transfer system according to a third embodiment. 14 is a flowchart illustrating a processing procedure of a data transfer program according to the third embodiment. FIG. 2 is a functional block diagram illustrating a configuration of a computer that executes a data transfer program according to the first embodiment.

Explanation of symbols

50 Transmission side server 60 Disk array device 70 Reception side server 100, 300, 500 Data transfer program 110, 310 Disk information acquisition unit 120 Disk usage status acquisition unit 130, 530 File division unit 140, 340 File transfer unit 150, 350, 550 Control unit 200, 400 Data reception program 210, 410 Disk information storage unit 220, 420 File reception unit 360 Transfer multiplicity setting unit 570 Tuning point setting unit 600 Computer 610 RAM
611 Data transfer program 620 CPU
621 Data transfer process 630 HDD
640 LAN interface 650 I / O interface 660 DVD drive 670 SCSI interface

Claims (5)

A data transfer program for transferring data from the first server to the disk when transferring data from the first server to the second server via a disk of a disk device connected in a SAN environment,
Transfer data division procedure to divide the data to be transferred into multiple,
A parallel transfer procedure for generating a plurality of transfer processes and transferring the transfer data divided by the transfer data dividing procedure to the disk in parallel;
A data transfer program characterized by causing the first server to execute. The first server further executes a transfer multiplicity determination procedure for determining the transfer multiplicity of data by obtaining the number of usable partitions from the disk used for data transfer,
The data transfer program according to claim 1, wherein the transfer data division procedure divides data to be transferred into a plurality of data based on the transfer multiplicity determined by the multiplicity determination procedure.   The transfer multiplicity determination procedure obtains the number of usable partitions from a plurality of disks, determines the data transfer multiplicity for each disk, and if there is a difference in the transfer performance of each disk, determines the disk with poor performance. 3. The data transfer program according to claim 2, wherein a part of the transferred multiplicity is transferred to a disk having good performance. A data transfer method for transferring data from a first server to a second server via a disk of a disk device connected in a SAN environment,
A data transmission step of dividing the data to be transferred into a plurality of pieces and transmitting in parallel from the first server to the disk of the disk device;
A data reception step of transferring the data transmitted to the disk of the disk device by the data transmission step in parallel to a second server, and combining the transferred data;
A data transfer method comprising: A data transfer device for transferring data from the first server to the disk when transferring data from the first server to the second server via a disk of a disk device connected in a SAN environment,
Transfer data dividing means for dividing the data to be transferred into a plurality of data;
Parallel transfer means for generating a plurality of transfer processes and transferring the transfer data divided by the transfer data dividing means to the disk in parallel;
A data transfer device comprising:

Images