JP6089855B2 - Virtualization system, virtual server, file writing method, and file writing program - Google Patents

Description

  The present invention relates to a differential disk type virtual machine, and more particularly to a file storage technique.

  In order to reduce the amount of files used in storage, it is necessary to prepare expensive dedicated deduplication storage. However, in the case of deduplication storage, files that are duplicated among multiple storages cannot be deleted.

  On the other hand, as a method of performing deduplication between a plurality of VMs (virtual machines) in a virtual environment, there is a method called a differential disk type virtual machine. In this method, a master virtual machine is used as a template, a plurality of virtual machines are created from the virtual machines, and only the differences between the virtual machines are stored on a disk.

  An example of background art is disclosed in Patent Document 1. In Patent Document 1, an OS hot fix or application is directly installed on the master.

  Here, FIG. 14 shows a method for accessing a file system of a virtual machine of the differential disk method according to the background art. In the background art method, when a file system is accessed by a virtual machine, an attempt is made to access the image of the virtual machine. If the location to be accessed exists on the difference disk, the difference disk is accessed. If the location to be accessed is consolidated in the master image, the virtual machine master image is accessed.

JP 2011-221597 A

  FIG. 15 is a diagram for explaining a problem in the background art. Reference numeral 900 denotes a master image of a virtual machine using a differential disk method. Reference numerals 901 and 902 denote difference disk images of virtual machines of difference disks. Reference numerals 911 and 912 denote file blocks. In the case of the existing technology, it is uncertain where the file block having the same content is written. That is, in the background art, when the same file is written, the address at which writing is performed in each virtual machine is different. Therefore, even if a file common to the virtual machine exists, it could not be extracted to the master part.

  In this way, in the virtual environment in the background art, since the file system operates independently on each virtual machine, even if the same content file is saved, each virtual machine is saved at different positions. It was. In this case, if you want to reflect the common part from the difference disk of each virtual machine to the master, the common part cannot be reflected to the master because the disk image compared in each virtual machine is different. It was.

  On the other hand, even if a file common to each virtual machine exists, for example, by installing an application that applies a common OS hot fix to each virtual machine, the file is placed on the difference disk of each virtual machine. For this reason, all changes in each virtual machine are accumulated in the difference disk, and there is a problem that the amount of storage used increases.

  Further, the technique disclosed in Patent Document 1 has a problem in that it is necessary to search for an unused area that is common to all virtual machines in order to determine an installation address.

(Object of invention)
An object of the present invention is to provide a virtualization system, a virtual server, a file writing method, and a file writing program that solve the above-described problems and reduce the storage usage.

  The first virtualization system of the present invention creates a plurality of virtual machines using a master virtual machine as a master image, and a virtual machine that operates a differential disk type virtual machine that stores a difference of each virtual machine in a difference disk A system comprising a virtual server including a plurality of virtual machines and a storage for storing a master image and a difference disk, wherein the virtual server has the same file block on each difference disk for each virtual machine. Control means for storing in the address is included.

  The first virtual server of the present invention is a virtual server including a difference disk type virtual machine that creates a plurality of virtual machines using a master virtual machine as a master image and stores differences between the virtual machines in a difference disk. For each virtual machine, control means for storing the same file block at the same address on each difference disk is provided.

  The first file writing method of the present invention uses a virtual server including a virtual machine of a differential disk system that creates a plurality of virtual machines using a master virtual machine as a master image and stores the differences of the virtual machines in a differential disk. In the file writing method, the control unit has a control step of storing a file block having the same contents for each virtual machine at the same address on each difference disk.

  The first file writing program of the present invention creates a plurality of virtual machines using a master virtual machine as a master image, and stores a difference of each virtual machine in a difference disk on a virtual server including a difference disk type virtual machine The control program causes the control unit to execute a control process for storing a file block having the same contents for each virtual machine at the same address on each difference disk.

  According to the present invention, the amount of storage used can be reduced.

It is a block diagram which shows the structure of the virtualization system by the 1st Embodiment of this invention. It is a figure which shows the file writing method by the 1st Embodiment of this invention. It is a flowchart which shows operation | movement of the CA program by the 1st Embodiment of this invention. It is a figure which shows a mode that CA program operate | moves within a disk driver in the 1st Embodiment of this invention. It is a flowchart which shows operation | movement in case the CA program exists in a disk driver in the 1st Embodiment of this invention. It is a figure which shows a mode that the CA program operate | moves within a file system in the 1st Embodiment of this invention. It is a flowchart which shows operation | movement when the CA program exists in a file system in the 1st Embodiment of this invention. It is a figure which shows a mode in case the CA program exists in a rearrangement program in the 1st Embodiment of this invention. It is a flowchart which shows operation | movement in case the CA program exists in a rearrangement program in the 1st Embodiment of this invention. It is a flowchart which shows operation | movement in the case of reflecting the common part on each difference disk in a master in the 1st Embodiment of this invention. It is a figure which shows the file writing method by the 1st Embodiment of this invention. It is a block diagram which shows the minimum structure of the virtualization system of this invention. It is a block diagram which shows the hardware structural example of the virtual server of this invention. It is a figure which shows the access method to the file system of the virtual machine of a difference disk system by background art. It is a figure explaining the subject by background art.

  In order to clarify the above and other objects, features and advantages of the present invention, embodiments of the present invention will be described in detail below with reference to the accompanying drawings. In addition to the above-described object of the present invention, other technical problems, means for solving the technical problems, and operational effects thereof will become apparent from the disclosure of the following embodiments. In all the drawings, the same reference numerals are given to the same components, and the description will be omitted as appropriate.

(First embodiment)
In the present invention, a content address (Contents Address) program (hereinafter referred to as a CA program) for controlling a file having the same content to be written to a disk block having the same address is introduced into the virtual machine. With the CA program, blocks of the same contents can be saved at the same address among multiple virtual machines. In a virtual machine created by the differential disk method, file blocks common to multiple virtual machines are extracted and used as the master. It becomes possible to arrange.

  Hereinafter, a first embodiment of the present invention will be described in detail with reference to the drawings.

  FIG. 1 is a block diagram showing a configuration of a virtualization system 1000 according to the first embodiment of the present invention. The virtualization system 1000 includes a virtual server 10 and a storage 300.

  In the virtual server 10, virtual machines 11 to 13 (virtual machines 1 to 3) are operating. The number of virtual machines is not limited to this, and can be arbitrarily set.

  A disk area on the storage device 300 is allocated to each of the virtual machines 11 to 13. The virtual machines 11 to 13 are created by the differential disk method.

  The storage apparatus 300 includes a master image 200 that is common to each virtual machine and difference disks 201 to 203 that are differences between the master images. A plurality of virtual machines are created using the master virtual machine as a master image, and differences between the virtual machines are stored in a difference disk.

  In each of the virtual machines 11 to 13, a CA program 100 that stores a block of a file having the same content at the same address is operating. The CA program (Contents Address program) determines the storage address of the file block from the CA address ID (ID determined from the file metadata such as the file name and the contents of the file block).

  Three types of methods are conceivable for the component in which the CA program operates.

  The first implementation method is a method that operates in a disk driver. When a file block is written from the file system, the CA program converts the address of the file block. By writing to the virtual disk at the converted address, the same contents are stored in the virtual disk having the same address. In this implementation method, the deduplication storage function is set in the virtual OS, and the physical address on the disk is also determined using the CA ID of each block as a key. However, in this method, it is difficult to obtain file metadata without a disk driver, so address conversion is performed using only the contents of the file block without using metadata.

  The second implementation method is a method that operates in a file system. In the file system, there is a table for managing the block address of the file to be written and the address of the virtual disk. In a normal file system, a free virtual disk address is used, but in this embodiment, a virtual disk address is determined from a CA ID by a CA program. By this method, the same contents are stored in the virtual disk with the same address.

  The third realization method is a method that operates in a file relocation program. When the relocation program relocates the file, the virtual disk address is determined from the CA ID, so that the same contents are stored in the virtual disk with the same address after the relocation of the file is completed. In this implementation method, the CA program causes the file allocation to be distributed in the storage, which is contrary to general defragmentation, and disk access becomes inefficient. However, in recent years, devices with excellent random access performance, such as SSD, have become widespread, and this method can be realized.

  The master update manager 101 operating on the virtual server 10 has a function of searching for a block having the same content in each of the virtual machines 11 to 13, reflecting the block on the master, and eliminating the overlapping portion.

(Description of the operation of the first embodiment)
Next, the operation of the virtualization system 1000 according to the present embodiment will be described in detail with reference to the drawings.

  FIG. 2 is a diagram illustrating a file writing method in the virtual machine according to the present invention. In the following description, the virtual machines 1 and 2 will be described as an example.

  The master image 200 is a master image of the virtual machines 1 and 2 created by the difference disk method. Each of the virtual machines 1 and 2 holds difference disks 201 and 202 that hold data other than the common parts.

  In the present invention, processing is performed by the CA program 100 so that the file blocks 211 and 212 having the same contents are stored in the same positions of the difference disks 201 and 202 of the virtual machines 1 and 2.

  The master update manager 101 periodically searches for a common part in the disks of the virtual machines 1 and 2. Since the blocks having the same contents are stored at the same address by the CA program 100, the search for the common block performed by the master update manager 101 only needs to be performed by comparing the blocks having the same address with each other. Is possible.

  When the master update manager 101 searches for the blocks 211 and 212 as common blocks in the target virtual machine, the contents are copied to the master block 213. Next, the common parts (blocks 211 and 212) copied to the master image 200 are deleted from the difference disks 201 and 202 of each virtual machine. When accessing the location (blocks 211 and 212) deleted from the difference disks 201 and 202, the master is accessed as described with reference to FIG.

  FIG. 3 is a flowchart showing the operation of the CA program 100 in the present embodiment. When there is an instruction to write a file block (step S1101), the CA program 100 obtains a CA ID determined from metadata such as the file name of the target block and the content of the data (step S1102). The CA ID is a kind of hash value.

  Based on the CA ID, the CA program 100 performs conversion to an address for storing the file block (step S1103). Then, the CA program 100 searches the address conversion table to check whether the converted address has already been used (step S1104).

  The CA ID is determined from the contents of the file. However, the uniqueness cannot be guaranteed, and the same value may be obtained even when the contents are different. If the converted address is used, the CA program 100 re-converts the address. A plurality of methods are conceivable for the re-conversion of the address, and one of them is a method of using one block of the converted address (step S1105).

  The CA program 100 performs address conversion until a post-conversion address that is not used can be found. If a post-conversion address that is not used can be found, the CA program 100 stores the pre-conversion address and the post-conversion address in a table, and writes a file block (step S1106).

  FIG. 4 is a diagram showing a case where the CA program 100 operates in the disk driver 112. In the present embodiment, the CA program 100 operates within the disk driver 112. In the case of this embodiment, an address conversion table 111 for managing addresses used in the file system 113 and addresses converted by the CA program 100 is held in the disk driver 112. When writing the file blocks 321 and 322 from the disk driver 112, addresses before and after conversion are added to the address conversion table 111. When reading the file block, the address is converted by the address conversion table 111, and reading is performed by the converted address.

  FIG. 5 is a flowchart showing the operation of the CA program 100 when the CA program 100 exists in the disk driver 112. When the file block write instruction is issued to the file system 113 (step S1201), the CA program 100 operating in the disk driver 112 converts the block address to the virtual disk address (post-conversion address) (step S1202). ). In the file system 113, the contents of the file block are written in the converted address position (step S1203).

  FIG. 6 is a diagram showing a case where the CA program 100 is incorporated in the file system 113. The CA program 100 is incorporated in the file system 113, and the CA program 100 determines the virtual disk address from the block address. The CA program 100 determines a virtual disk address from the CA ID, and file blocks having the same contents are written to the same address.

  FIG. 7 is a flowchart showing an operation when the CA program 100 exists in the file system 113. When a file block write instruction is issued to the file system 113 (step S1301), the CA program 100 operating in the file system 113 converts the block address into a virtual disk address (step S1302). In the file system 113, the contents of the file block are written at the converted address position (step S1303).

  FIG. 8 is a diagram illustrating a case where the CA program 100 exists in the relocation program 115 operating on the virtual machine 11. When the file defragmentation is executed by the relocation program 115, the virtual disk address is determined from the CA ID, and the block is moved. By this relocation program 115, file blocks having the same contents among a plurality of virtual machines are arranged at the same address positions.

  FIG. 9 is a flowchart showing an operation when the CA program 100 exists in the relocation program 115 operating on the virtual machine 11. When the rearrangement program 115 is executed, the file block is moved. The CA program 100 determines the destination address of the file block based on the CA ID (step S1401). The rearrangement program 115 moves the file block to the converted address (step S1402).

  FIG. 10 is a flowchart showing the operation when the common part is reflected on the master in this embodiment. The master update manager 101 refers to the difference disk of each virtual machine and searches for a block in which the contents of the file block match (step S1501). Next, the master update manager 101 copies the file blocks whose contents match to the master image (step S1502). Finally, the master update manager 101 deletes the file block copied to the master image from the difference disk of the virtual machine (step S1503).

  By the way, in the CA program 100 described with reference to FIG. 3, if the converted address is already used, other free addresses are set as candidates. This is because if the CA ID is the same, it cannot be guaranteed that the stored contents are the same.

  However, in reality, the possibility that the CA ID is the same and the contents are different is sufficiently low. Therefore, as shown in the flowchart of FIG. 11, if the post-conversion address is already used, it is possible to check whether the contents of the file blocks match. In the flowcharts described with reference to FIGS. 3 and 11, when the converted address cannot be used, the address is shifted by one and the adjacent block is used, but again with an algorithm different from the CA ID used first. A method for calculating the CA ID is conceivable.

(Effects of the first embodiment)
According to the present embodiment, in a virtual environment, by combining a technology for storing file blocks having the same contents in a virtual machine at the same address and a technology for reflecting a common part existing in a difference disk of each virtual machine on a master. The common part existing in each difference disk can be moved to the master image, and the amount of disk usage in the entire system can be reduced. In other words, it is possible to reduce the amount of storage used by moving a common part from the difference disk of the virtual machine to the master virtual machine.

  Further, according to the present embodiment, when writing a file from a virtual machine, if the same contents are stored, it is stored at the same address. Processing can be performed.

  In addition, according to the present embodiment, since a common part such as an OS hot fix is reflected on the master from each virtual machine, it is not necessary to recreate the virtual machine by updating the master, and the operation time can be reduced. It becomes. Furthermore, in addition to OS hot fixes and applications, files shared by each virtual machine are also found at the same time and transferred to the master, thereby reducing the amount of disk usage.

  In the present embodiment, it is possible to delete duplicate files in a plurality of storages without selecting the type of storage.

  Here, FIG. 12 shows a minimum configuration capable of solving the problems of the present invention. The virtualization system 1000 is a virtualization system 1000 in which a virtual machine using a differential disk system that creates a plurality of virtual machines using a master virtual machine as a master image and stores differences between the virtual machines in a differential disk operates. A virtual server 10 including a plurality of virtual machines, and a storage 300 for storing a master image and a difference disk, and the virtual server 10 assigns the same file block for each virtual machine to the same address on each difference disk. By including the control means stored in the above, the above-described problems of the present invention can be solved.

  Next, a hardware configuration example of the virtual server 10 of the present invention will be described with reference to FIG.

  Referring to FIG. 13, the virtual server 10 of the present invention has a hardware configuration similar to that of a general computer device, and includes data such as a CPU (Central Processing Unit) 401 and a RAM (Random Access Memory). The main storage unit 402 used for the work area and the temporary data saving area, the communication unit 403 that transmits / receives data via the network, the input device 405, the output device 406, and the storage device 407 are connected to transmit / receive data. An output interface unit 404 and a system bus 408 for interconnecting the above components are provided. The storage device 407 is realized by, for example, a hard disk device including a non-volatile memory such as a ROM (Read Only Memory), a magnetic disk, and a semiconductor memory.

  Each function of the virtual server 10 of the present invention is realized by implementing circuit operations, which are hardware components such as LSI (Large Scale Integration), in which a program is incorporated, thereby realizing the operation in hardware. A program that provides the function can be realized in software by storing the program in the storage device 407, loading the program into the main storage unit 402, and executing it by the CPU 401.

  The present invention has been described above with reference to preferred embodiments. However, the present invention is not necessarily limited to the above embodiments, and various modifications can be made within the scope of the technical idea. it can.

  It should be noted that any combination of the above-described constituent elements and a conversion of the expression of the present invention between a method, an apparatus, a system, a recording medium, a computer program, and the like are also effective as an aspect of the present invention.

  The various components of the present invention do not necessarily have to be independent of each other. A plurality of components are formed as a single member, and a single component is formed of a plurality of members. It may be that a certain component is a part of another component, a part of a certain component overlaps with a part of another component, or the like.

  Moreover, although the several procedure is described in order in the method and computer program of this invention, the order of the description does not limit the order which performs a several procedure. For this reason, when implementing the method and computer program of this invention, the order of the several procedure can be changed in the range which does not interfere in content.

  The plurality of procedures of the method and the computer program of the present invention are not limited to being executed at different timings. For this reason, another procedure may occur during the execution of a certain procedure, or some or all of the execution timing of a certain procedure and the execution timing of another procedure may overlap.

  Further, a part or all of the above-described embodiment can be described as in the following supplementary notes, but is not limited thereto.

(Appendix 1)
A virtualization system in which a virtual machine of a differential disk type that creates a plurality of virtual machines using a master virtual machine as a master image and stores differences between the virtual machines in a differential disk operates.
A virtual server containing multiple virtual machines;
A storage for storing the master image and the difference disk;
The virtual server is
A virtualization system comprising: a control means for storing a file block having the same contents for each virtual machine at the same address on each difference disk.

(Appendix 2)
Periodically, each differential disk is searched for the presence of a file block having the same content, and if it exists, the file block is stored at the same address as the address on the differential disk on the master image, and the differential The virtualization system according to claim 1, further comprising a master update manager that deletes the file block on the disk.

(Appendix 3)
It has an address conversion table that stores the address before conversion and the address after conversion.
The control means includes
When there is an instruction to write a file block, a CA ID that is a hash value is obtained based on the metadata and data contents of the block, and an address is converted based on the CA ID.
Write the file block to the converted address,
The virtualization system according to appendix 1 or appendix 2, wherein an address before translation and an address after translation are stored in an address translation table.

(Appendix 4)
The control means includes
The virtualization system according to appendix 3, wherein the address is re-converted when the converted address is used.

(Appendix 5)
The control means includes
If the converted address is used, check whether the file block to be written and the contents of the already written file block match, and if they match, the process is terminated without writing,
The virtualization system according to appendix 3, wherein the address is reconverted when the contents of the file block do not match.

(Appendix 6)
The control means includes
6. The virtualization system according to appendix 4 or appendix 5, wherein when the converted address is used, re-conversion is performed using one block adjacent to the translated address as the translated address.

(Appendix 7)
Relocation means for relocating file blocks is provided,
The control means is
When the file is defragmented on the virtual machine, the file block destination address is determined based on the CA ID which is a hash value based on the metadata and data contents of the file block;
The relocation means comprises:
The virtualization system according to any one of appendix 1 to appendix 6, wherein the file block is moved to a destination address determined by the control means.

(Appendix 8)
A virtual server including a virtual machine of a differential disk method that creates a plurality of virtual machines using a master virtual machine as a master image and stores the difference of each virtual machine in a differential disk,
A virtual server comprising control means for storing a file block having the same contents for each of the virtual machines at the same address on each difference disk.

(Appendix 9)
Periodically, each differential disk is searched for the presence of a file block having the same content, and if it exists, the file block is stored at the same address as the address on the differential disk on the master image, and the differential The virtual server according to appendix 8, further comprising a master update manager that deletes the file block on the disk.

(Appendix 10)
It has an address conversion table that stores the address before conversion and the address after conversion.
The control means includes
When there is an instruction to write a file block, a CA ID that is a hash value is obtained based on the metadata and data contents of the block, and an address is converted based on the CA ID.
Write the file block to the converted address,
The virtual server according to appendix 8 or appendix 9, wherein the address before translation and the address after translation are stored in an address translation table.

(Appendix 11)
The control means includes
The virtual server according to appendix 10, wherein when the address after conversion is used, the address is re-converted.

(Appendix 12)
The control means includes
If the converted address is used, check whether the file block to be written and the contents of the already written file block match, and if they match, the process is terminated without writing,
The virtual server according to appendix 10, wherein the address is reconverted when the contents of the file block do not match.

(Appendix 13)
The control means includes
13. The virtual server according to appendix 11 or appendix 12, wherein when the translated address is used, reconversion is performed using a block adjacent to the translated address as the translated address.

(Appendix 14)
Relocation means for relocating file blocks is provided,
The control means is
When the file is defragmented on the virtual machine, the file block destination address is determined based on the CA ID which is a hash value based on the metadata and data contents of the file block;
The relocation means comprises:
The virtual server according to any one of appendix 8 to appendix 13, wherein the file block is moved to the destination address determined by the control means.

(Appendix 15)
A method of writing a file by a virtual server including a virtual machine of a differential disk method that creates a plurality of virtual machines using a master virtual machine as a master image and stores a difference of each virtual machine in a differential disk,
A file writing method, wherein the control means includes a control step of storing a file block having the same content for each virtual machine at the same address on each difference disk.

(Appendix 16)
The master update manager periodically searches for a file block having the same contents on each difference disk, and if it exists, saves the file block at the same address on the master image as the address on the difference disk. And a master updating step of deleting the file block on the difference disk.

(Appendix 17)
In the control step,
When there is an instruction to write a file block, a CA ID that is a hash value is obtained based on the metadata and data contents of the block, and an address is converted based on the CA ID.
Write the file block to the converted address,
The file writing method according to appendix 15 or appendix 16, wherein the address before translation and the address after translation are stored in an address translation table.

(Appendix 18)
In the control step,
The file writing method according to appendix 17, wherein when the converted address is used, the address is re-converted.

(Appendix 19)
In the control step,
If the converted address is used, check whether the file block to be written and the contents of the already written file block match, and if they match, the process is terminated without writing,
18. The file writing method according to appendix 17, wherein the address is reconverted when the contents of the file block do not match.

(Appendix 20)
In the control step,
20. The file writing method according to appendix 18 or appendix 19, wherein when the converted address is used, reconversion is performed using one block adjacent to the converted address as the converted address.

(Appendix 21)
The rearrangement means includes a rearrangement step for rearranging the file blocks;
In the control step,
When the file is defragmented on the virtual machine, the file block destination address is determined based on the CA ID which is a hash value based on the metadata and data contents of the file block;
In the rearrangement step,
21. The file writing method according to any one of appendix 15 to appendix 20, wherein the file block is moved to a destination address determined by the control step.

(Appendix 22)
A file writing program that operates on a virtual server including a virtual machine of a differential disk method that creates a plurality of virtual machines using a master virtual machine as a master image and stores the differences of each virtual machine in a differential disk,
A file writing program characterized by causing a control unit to execute a control process for storing a file block having the same contents at the same address on each difference disk for each of the virtual machines.

(Appendix 23)
The master update manager periodically searches for a file block with the same contents on each difference disk, and if it exists, saves the file block at the same address on the master image as the address on the difference disk. 23. The file writing program according to appendix 22, wherein a master update process for deleting the file block on the difference disk is executed.

(Appendix 24)
In the control process,
When there is an instruction to write a file block, a CA ID that is a hash value is obtained based on the metadata and data contents of the block, and an address is converted based on the CA ID.
Write the file block to the converted address,
The file writing program according to appendix 22 or appendix 23, wherein the address before translation and the address after translation are stored in an address translation table.

(Appendix 25)
In the control process,
25. The file writing program according to appendix 24, wherein when the converted address is used, the address is re-converted.

(Appendix 26)
In the control process,
If the converted address is used, check whether the file block to be written and the contents of the already written file block match, and if they match, the process is terminated without writing,
25. The file writing program according to appendix 24, wherein when the contents of the file block do not match, the address is re-converted.

(Appendix 27)
In the control process,
27. The file writing program according to appendix 25 or appendix 26, wherein when the converted address is used, reconversion is performed with one block adjacent to the converted address as the converted address.

(Appendix 28)
Have the relocation means execute a relocation process to relocate the file block,
In the control process,
When the file is defragmented on the virtual machine, the file block destination address is determined based on the CA ID which is a hash value based on the metadata and data contents of the file block;
In the rearrangement process,
The file writing program according to any one of appendix 22 to appendix 27, wherein the file block is moved to the destination address determined by the control process.

10: Virtual server 11-13: Virtual machine 100: CA program 101: Master update manager 111: Address conversion table 112: Disk driver 113: File system 114: Virtual disk address conversion table 115: Relocation program 200: Master image 201 203: Difference disk 211-213, 321, 322, 331, 332: File block 300: Storage 401: CPU
402: Main storage unit 403: Communication unit 404: Input / output interface unit 405: Input device 406: Output device 407: Storage device 408: System bus 1000: Virtualization system

Claims (10)

A virtualization system in which a virtual machine of a differential disk type that creates a plurality of virtual machines using a master virtual machine as a master image and stores differences between the virtual machines in a differential disk operates.
A virtual server containing multiple virtual machines;
A storage for storing the master image and the difference disk;
The virtual server is
A virtualization system comprising: a control means for storing a file block having the same contents for each virtual machine at the same address on each difference disk. Periodically, each differential disk is searched for the presence of a file block having the same content, and if it exists, the file block is stored at the same address as the address on the differential disk on the master image, and the differential The virtualization system according to claim 1, further comprising a master update manager that deletes the file block on the disk. It has an address conversion table that stores the address before conversion and the address after conversion.
The control means includes
When there is an instruction to write a file block, a CA ID that is a hash value is obtained based on the metadata and data contents of the block, and an address is converted based on the CA ID.
Write the file block to the converted address,
The virtualization system according to claim 1 or 2, wherein an address before conversion and an address after conversion are stored in an address conversion table. The control means includes
The virtualization system according to claim 3, wherein when the address after conversion is used, the address is re-converted. The control means includes
If the converted address is used, check whether the file block to be written and the contents of the already written file block match, and if they match, the process is terminated without writing,
4. The virtualization system according to claim 3, wherein when the contents of the file block do not match, address re-conversion is performed. The control means includes
6. The virtualization system according to claim 4, wherein, when the converted address is used, re-conversion is performed by using one block adjacent to the converted address as a converted address. 6. Relocation means for relocating file blocks is provided,
The control means is
When the file is defragmented on the virtual machine, the file block destination address is determined based on the CA ID which is a hash value based on the metadata and data contents of the file block;
The relocation means comprises:
The virtualization system according to claim 1, wherein the file block is moved to a destination address determined by the control unit. A virtual server including a virtual machine of a differential disk method that creates a plurality of virtual machines using a master virtual machine as a master image and stores the difference of each virtual machine in a differential disk,
A virtual server comprising control means for storing a file block having the same contents for each of the virtual machines at the same address on each difference disk. A method of writing a file by a virtual server including a virtual machine of a differential disk method that creates a plurality of virtual machines using a master virtual machine as a master image and stores a difference of each virtual machine in a differential disk,
A file writing method, wherein the control means includes a control step of storing a file block having the same content for each virtual machine at the same address on each difference disk. A file writing program that operates on a virtual server including a virtual machine of a differential disk method that creates a plurality of virtual machines using a master virtual machine as a master image and stores the differences of each virtual machine in a differential disk,
A file writing program characterized by causing a control unit to execute a control process for storing a file block having the same contents at the same address on each difference disk for each of the virtual machines.

Images