JP5944796B2 - Disaster recovery method and disaster recovery system - Google Patents

Description

  The present invention relates to a disaster recovery method and a disaster recovery system for an IT system on a virtual environment, and more particularly, a plurality of recovery sites, a plurality of RPO (recovery point object: target recovery point) / RTO (recovery time object: target recovery time). It relates to disaster recovery with levels.

  Triggered by the Great East Japan Earthquake that occurred on March 11, 2011, (1) Necessity of social infrastructure to support life continuity, (2) Strengthening business continuity management as a corporate social responsibility, (3) Global premise The issues such as the business development and (4) improvement / strengthening the corporate management structure due to unmanageable management have been recognized again.

  In particular, in the area of continuity, awareness of backup, BC (Business Continuity) / DR (Disaster Recovery) is increasing.

  However, in the backup and BC / DR, specifically, the following points are regarded as problems.

○ Geographic distance between the primary site and the recovery site ○ Distance from the ocean ○ Companies that have not responded to disaster recovery (mainly SMEs) will be separated from the supply chain ○ Japan is an earthquake Concerns about being a major power ○ Concerns about having a single recovery site For this reason, a disaster recovery method and a disaster recovery system with the following features are required.

○ Ability to have recovery sites in multiple geographically separated locations ○ Able to provide a service level plan that meets the budget that SMEs can easily use to maintain the supply chain. Specifically, when backing up to multiple recovery sites, it is possible to specify different RPO / RTO levels for each recovery site. (A) RPO / RTO at minute / time level
・ Hot standby method by near real-time replication including memory disk is required. (B) RPO / RTO at day level
・ Cold standby method by disk replication is required Currently, there are various technologies to realize disaster recovery, but each has the following problems, and disaster recovery method and disaster recovery system with the above features It cannot be realized.

○ CDP (Continuous Data Protection: Continuous data protection technology), various replication technologies ・ To realize RPO / RTO at the minute / hour level, it is necessary to integrate both memory and disk to take a quiesce point Other backup software ・ Cannot transfer data to multiple sites efficiently ・ Requires backup for the number of sites, satisfying RPO / RTO without server processing performance and network bandwidth scale-up Therefore, the object of the present invention is to apply the two methods of hot standby by near real-time replication including a memory disk and cold standby by disk replication to a plurality of recovery sites in an integrated manner. Can It is an object of the present invention to provide a disaster recovery method and a disaster recovery system.

  The above and other objects and novel features of the present invention will be apparent from the description of this specification and the accompanying drawings.

  Of the inventions disclosed in the present application, the outline of typical ones will be briefly described as follows.

  That is, a typical outline is a disaster recovery method for backing up data of a first virtual machine on a data center to second virtual machines on a plurality of other data centers. Thus, the update data of the first virtual machine is extracted and stored in the first update data storage area in the backup source data center while inserting a plurality of levels of synchronization points related to disaster recovery, and the first update data storage The data stored in the area is transferred to a plurality of backup destination data centers, stored in a second update data storage area in the backup destination data center, and stored in the second update data storage area by the backup destination data center. Based on the synchronization point stored in the area, To confirm the backup data corresponding to the level, and stores the backup data to the second virtual machine.

  A disaster recovery system for backing up data of a first virtual machine on a data center to second virtual machines on a plurality of other data centers, the backup source data center being a first virtual machine Update data extraction processing unit for extracting the update data and storing in a first update data storage area in the data center of the backup source while inserting a plurality of synchronization points related to disaster recovery, and a first update data storage area A multi-site data transfer processing unit for transferring the data stored in the data to a plurality of backup destination data centers and storing the data in a second update data storage area in the backup destination data center. The center performs backup based on the synchronization point stored in the second update data storage area. To confirm the backup data corresponding to the level for the previous disaster recovery, and has a backup data determination processing unit for storing the backup data to the second virtual machine.

  The effects obtained by typical ones of the inventions disclosed in the present application will be briefly described as follows.

  In other words, the effect obtained by the representative one can realize a disaster recovery method and a disaster recovery system having a plurality of recovery sites and a plurality of RPO / RTO levels. Furthermore, the configuration of the disaster recovery system can be simplified as compared with the prior art.

1 is a configuration diagram showing a configuration of a data center network to which a disaster recovery system according to an embodiment of the present invention is applied. FIG. It is explanatory drawing for demonstrating the outline | summary of the backup by the disaster recovery system which concerns on one embodiment of this invention. It is explanatory drawing for demonstrating the detail of the backup by the disaster recovery system which concerns on one embodiment of this invention.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. Note that components having the same function are denoted by the same reference symbols throughout the drawings for describing the embodiment, and the repetitive description thereof will be omitted.

<Configuration of data center network to which disaster recovery system is applied>
A configuration of a data center network to which a disaster recovery system according to an embodiment of the present invention is applied will be described with reference to FIG. FIG. 1 is a configuration diagram showing a configuration of a data center network to which a disaster recovery system according to an embodiment of the present invention is applied.

  In FIG. 1, Kansai data center 100, Chubu data center 110, Kanto data center 120, Kyushu data center 130, China data center 140, and Hokkaido data center 150 are installed.

  Each data center can be connected as necessary, and a data center network is formed.

  In the data center, the virtual machine is activated by the virtualization mechanism. In the example shown in FIG. 1, the data of the virtual machine 101, which is the first virtual machine on the Kansai data center 100, is stored on the central data center 110. Are backed up to the virtual machine 111 which is the second virtual machine and the virtual machine 131 on the Kyushu data center 130 which is the second virtual machine.

  Further, the data of the virtual machine 111 on the central data center 110 is backed up to the virtual machine 121 which is the third virtual machine on the Kanto data center 120.

<Overview of disaster recovery system backup>
Next, an outline of backup by the disaster recovery system according to the embodiment of the present invention will be described with reference to FIG. FIG. 2 is an explanatory diagram for explaining the outline of backup by the disaster recovery system according to the embodiment of the present invention.

  In FIG. 2, a memory 102, a disk (data) 103, and a disk (system) 104 as a virtual machine are allocated to a virtual machine 101 on the Kansai data center 100. The memory 102, the disk (data) 103, Data of the disk (system) 104 is backed up.

  As a backup destination virtual machine, the virtual machine 111 on the central data center 110 is operated as a hot standby type, and the virtual machine 121 on the Kanto data center 120 and the virtual machine 131 on the Kyushu data center 130 are operated as a cold standby type. .

  Further, the virtual machine 121 on the Kanto data center 120 is backed up from the virtual machine 111 on the central data center 110 of the hot standby type.

  Further, from the virtual machine 101 on the Kansai data center 100, a cold standby type synchronization point signal 10 for achieving cold standby type synchronization, a hot standby type synchronization point signal 11 for obtaining hot standby type synchronization, and a memory Alternatively, memory / disk update data 12 that is disk update data is transmitted as data of the memory 102, the disk (data) 103, and the disk (system) 104.

  The cold standby synchronization point signal 10 and the hot standby synchronization point signal 11 are used as multiple levels of synchronization points for disaster recovery.

  Each data transmitted from the virtual machine 101 on the Kansai data center 100 is stored in the virtual machine 101 on the hot standby type central data center 110 on the memory 102, the disk (data) 103, and the disk ( System) 104 data is backed up to the memory 112, the disk (data) 113, and the disk (system) 114.

  Further, in the virtual machine 131 on the cold standby type Kyushu data center 130, the data in the memory 102 is not backed up, and the data in the disk (data) 103 and the disk (system) 104 is stored in the disk (data) 132 and the disk (system). ) 133 is backed up.

  Further, the virtual machine 111 on the hot standby type central data center 110 further receives the received cold standby type synchronization point signal 10, hot standby type synchronization point signal 11, and memory disk update data 12 from the cold standby type Kanto region. It is transmitted to the virtual machine 121 on the data center 120.

  Each data transmitted from the virtual machine 111 on the hot standby type central data center 110 is not backed up in the virtual machine 121 on the cold standby type Kanto data center 120, and the disk (data ) 113 and disk (system) 114 data are backed up to disk (data) 122 and disk (system) 123.

  These backups are performed in the virtual machine 121 on the cold standby type Kanto data center 120 and the virtual machine 131 on the Kyushu data center 130 based on the cold standby type synchronization point signal 10.

  Further, in the virtual machine 111 on the hot standby type central data center 110, backup processing is performed based on the hot standby type synchronization point signal 11.

  As described above, in the present embodiment, the virtual machine 101 on the Kansai data center 100 that is the backup source has the cold standby synchronization point signal 10, hot standby for the virtual machines on the data center that is a plurality of recovery sites. The type synchronization point signal 11 and the memory / disk update data 12 are transmitted asynchronously, and backup is performed at the backup source based on the data.

  Further, the backup is performed based on the cold standby synchronization point signal 10 and the hot standby synchronization point signal 11 depending on the difference between the hot standby type and the cold standby type.

  Further, the received cold standby synchronization point signal 10, hot standby synchronization point signal 11, and memory / disk update data 12 are transmitted as they are from a hot standby virtual machine to another virtual machine. Thus, the backup source data is backed up to another virtual machine.

  Thereby, disaster recovery having a plurality of RPO / RTO levels can be realized at a plurality of recovery sites.

  Furthermore, the configuration of the disaster recovery system can be simplified as compared with the prior art.

<Details of disaster recovery system backup>
Next, details of backup by the disaster recovery system according to the embodiment of the present invention will be described with reference to FIG. FIG. 3 is an explanatory diagram for explaining the details of backup by the disaster recovery system according to the embodiment of the present invention.

  In FIG. 3, a virtual machine 101 on the Kansai data center 100 is assigned a memory 102, a disk (data) 103, and a disk (system) 104 by the virtualization mechanism 200 of the Kansai data center 100, and other CPUs (not shown). Z) etc. are also assigned and operating.

  Similarly, the virtual machine 111 of the central data center 110 is operated by the virtualization mechanism 210, the virtual machine 121 of the Kanto data center 120 is operated by the virtualization mechanism 220, and the virtual machine 131 of the Kyushu data center 130 is virtual It is operated by the conversion mechanism 230.

  The virtualization mechanism 200 also includes a memory / disk update data extraction processing unit 205 that extracts update data from a memory or a disk, a multi-site data transfer processing unit 206 that transfers data to a plurality of backup destinations, and a memory or disk A memory / disk backup data determination processing unit 207 for determining the backup data, and a memory / disk quiesce point acquisition processing unit 208 for acquiring a quiesce point when generating update data for backup of the memory or disk.

  Although not shown, the virtualization mechanisms 210, 220, and 230 similarly have a memory / disk update data extraction processing unit 205, a multi-site data transfer processing unit 206, a memory / disk backup data determination processing unit 207, and a memory / disk. A stationary point acquisition processing unit 208 is provided.

  The virtualization mechanism 200 is a first update data storage area in which the cold standby synchronization point signal 10, the hot standby synchronization point signal 11, and the memory / disk update data 12 are stored when backup is performed. An update data storage area 201 is provided.

  Similarly, the virtualization mechanisms 210, 220, and 230 store the update data storage area 211 that is the second update data storage area, the update data storage area 221 that is the third update data storage area, and the second update data storage. An update data storage area 231 that is an area is provided.

  In this embodiment, the backup data is not transferred directly from the backup source virtual machine, but is stored once in the update data storage area in the virtualization mechanism at the backup source, and the backup destination is also virtual. After each data is stored in the update data storage area in the storage function, the backup data is determined and backed up to the backup destination virtual machine.

  The operation during backup will be described below.

  First, the memory / disk quiesce point acquisition processing unit 208 acquires data as a backup quiesce point, and the memory / disk update data extraction processing unit 205 performs update data extraction processing to extract update data.

  The update data is stored in the update data storage area 201 by inserting a cold standby synchronization point signal 10 and a hot standby synchronization point signal 11 as synchronization points having different RPO / RTO levels.

  Thereafter, the data transfer processing is performed by the multi-site data transfer processing unit 206 and transferred to the update data storage areas of the virtualization mechanisms of the plurality of data centers.

  In the example shown in FIG. 3, each data in the update data storage area 201 of the virtualization mechanism 200 of the Kansai data center 100 is stored in the update data storage area 211 of the virtualization mechanism 210 of the central data center 110 and the virtualization of the Kyushu data center 130. It has been transferred to the update data storage area 231 of the mechanism 230.

  In the virtualization mechanism 210 of the central data center 110, the data stored in the update data storage area 211 is subjected to backup data confirmation processing by the memory / disk backup data confirmation processing unit 207, confirmed as backup data, and the virtual machine 111 memory 112, disk (data) 113, and disk (system) 114.

  Since the virtual machine 111 of the central data center 110 operates as a hot standby type, the cold standby synchronization point signal 10 and the hot standby synchronization point signal 11 are processed as synchronization points, and the memory between the processed synchronization points is processed. Disk update data 12 is stored as backup data.

  Further, in the virtualization mechanism 230 of the Kyushu data center 130, the data stored in the update data storage area 231 is subjected to backup data confirmation processing by the memory / disk backup data confirmation processing unit 207 and confirmed as backup data. And stored in the disk (data) 132 and the disk (system) 133 of the virtual machine 131.

  Since the virtual machine 131 of the Kyushu data center 130 operates as a cold standby type, the memory update data is ignored, the hot standby synchronization point signal 11 is ignored, and the cold standby synchronization point signal 10 is synchronized. The memory disk update data 12 between the processed synchronization points is stored as backup data.

  Further, in the virtualization mechanism 210 of the central data center 110, the data transfer process is performed by the multi-site data transfer processing unit 206, and each data transferred from the virtualization mechanism 200 of the Kansai data center 100 is used as it is. The update data storage area 221 of the 120 virtualization mechanism 220 is transferred.

  In the virtualization mechanism 220 of the Kanto data center 120, the data stored in the update data storage area 221 is subjected to backup data confirmation processing by the memory / disk backup data confirmation processing unit 207, and confirmed as backup data. And stored in the disk (data) 122 and the disk (system) 123 of the virtual machine 121.

  Since the virtual machine 121 of the Kanto data center 120 operates as a cold standby type, the memory update data is ignored, the hot standby synchronization point signal 11 is ignored, and the cold standby synchronization point signal 10 is synchronized. The memory disk update data 12 between the processed synchronization points is stored as backup data.

  As described above, in this embodiment, an update data storage area is provided in the virtualization mechanism of the data center, and update data is not transferred between virtual machines, but update data is transferred between update data storage areas. So it can be a simple configuration.

  In addition, it will be possible to eliminate geographical concerns and prevent disruption of the supply chain against widespread disasters such as the Great East Japan Earthquake, contributing to customer satisfaction in terms of business continuity.

  As mentioned above, the invention made by the present inventor has been specifically described based on the embodiment. However, the present invention is not limited to the embodiment, and various modifications can be made without departing from the scope of the invention. Needless to say.

  The present invention relates to a disaster recovery method and a disaster recovery system for an IT system in a virtual environment, and can be widely applied to a plurality of recovery sites, systems that require disaster recovery having a plurality of RPO / RTO levels, and the like.

  DESCRIPTION OF SYMBOLS 10 ... Cold standby type synchronization point signal, 11 ... Hot standby type synchronization point signal, 12 ... Memory disk update data, 100 ... Kansai data center, 101 ... Virtual machine, 102 ... Memory, 103 ... Disk (data), 104 ... Disk (system), 110 ... Central data center, 111 ... Virtual machine, 112 ... Memory, 113 ... Disk (data), 114 ... Disk (system), 120 ... Kanto data center, 121 ... Virtual machine, 122 ... Disk (data) ), 123 ... disk (system), 130 ... Kyushu data center, 131 ... virtual machine, 132 ... disk (data), 133 ... disk (system), 140 ... China data center, 150 ... Hokkaido data center, 200 ... virtualization Mechanism, 201 ... Update data storage area 205 ... Memory / disk update data extraction processing unit, 206 ... Multiple site data transfer processing unit, 207 ... Memory / disk backup data confirmation processing unit, 208 ... Memory / disk quiesce point acquisition processing unit, 210 ... Virtualization mechanism, 211 ... update data storage area, 220 ... virtualization mechanism, 221 ... update data storage area, 230 ... virtualization mechanism, 231 ... update data storage area.

Claims (8)

A disaster recovery method for backing up data of a first virtual machine on a data center to second virtual machines on a plurality of other data centers,
The update data of the first virtual machine is extracted by the data center of the backup source, and stored in the first update data storage area in the data center of the backup source while inserting a plurality of synchronization points regarding disaster recovery. Then, the data stored in the first update data storage area is transferred to the plurality of backup destination data centers and stored in the second update data storage area in the backup destination data center,
Based on the synchronization point stored in the second update data storage area, the backup data center determines backup data according to the level related to the disaster recovery of the backup destination, and the second virtual data center A disaster recovery method, wherein the backup data is stored in a machine. The disaster recovery method according to claim 1,
The data stored in the second update data storage area is transferred as it is to the data center of the other backup destination by the data center of the backup destination, and the third update data storage in the data center of the backup destination is stored. A disaster recovery method characterized by storing in an area. In the disaster recovery method according to claim 2,
The disaster recovery method, wherein the synchronization point is a hot standby type or a cold standby type synchronization point. In the disaster recovery method according to claim 3,
Update data of the memory and disk of the first virtual machine is extracted by the data center of the backup source, and a plurality of levels of synchronization points related to the disaster recovery are inserted into each of the memory and the disk. Storing in the first update data storage area in the data center;
When the second virtual machine is a hot standby backup destination, the backup data is determined by the data center of the backup destination based on the hot standby type and the cold standby type synchronization point, and The backup data is stored in a memory and a disk of a second virtual machine, and when the second virtual machine is a cold standby backup destination, the backup data is stored based on the cold standby synchronization point. A disaster recovery method comprising: confirming and storing the backup data in the disk of the second virtual machine. A disaster recovery system that backs up data of a first virtual machine on a data center to second virtual machines on a plurality of other data centers,
The data center of the backup source
An update data extraction processing unit that extracts update data of the first virtual machine and stores it in a first update data storage area in the data center of the backup source while inserting a plurality of levels of synchronization points related to disaster recovery;
Multi-site data transfer processing for transferring data stored in the first update data storage area to a plurality of backup destination data centers and storing the data in a second update data storage area in the backup destination data center And
The backup data center is
Based on the synchronization point stored in the second update data storage area, the backup data corresponding to the level related to the disaster recovery of the backup destination is determined, and the backup data is stored in the second virtual machine. A disaster recovery system comprising a backup data confirmation processing unit. In the disaster recovery system according to claim 5,
The backup data center is
The multi-site data transfer processing unit;
The multi-site data transfer processing unit transfers the data stored in the second update data storage area as it is to the data center of another backup destination, and the third update data in the data center of the backup destination A disaster recovery system characterized by being stored in a storage area. The disaster recovery system according to claim 6,
The disaster recovery system according to claim 1, wherein the synchronization point is a hot standby type or a cold standby type synchronization point. In the disaster recovery system according to claim 7,
The update data extraction processing unit extracts the update data of the memory and the disk of the first virtual machine, and inserts a plurality of synchronization points related to the disaster recovery for each of the memory and the disk while the backup source Storing in the first update data storage area in the data center;
When the second virtual machine is a hot standby type backup destination, the backup data determination processing unit determines the backup data based on the hot standby type and cold standby type synchronization points, and The backup data is stored in a memory and a disk of a second virtual machine, and when the second virtual machine is a cold standby backup destination, the backup data is stored based on the cold standby synchronization point. A disaster recovery system characterized by confirming and storing the backup data in the disk of the second virtual machine.

Images