CN107508876A - A data-level remote disaster recovery method based on storage virtualization gateway - Google Patents

Abstract

The invention discloses a data-level off-site disaster recovery method based on a storage virtualization gateway. The method aims to solve the problem that an information system can still provide normal services to the greatest extent after experiencing various disasters. The method mainly includes: (1) Integration of heterogeneous storage. Through the storage virtualization gateway SVC, the storage of different brands and different interfaces has the same function, presents the same interface to the application, and integrates multiple storage systems of different brands and models into a single storage pool for unified management; (2) Heterogeneous storage Remote disaster recovery. After integrating the heterogeneous storage, the global mirror technology (Global Mirror) is used to realize remote disaster recovery, and the data of the primary data center is transmitted to the disaster recovery data center through a dedicated network, providing business continuity protection with a maximum distance of 800KM between two sites.

Description

A data-level remote disaster recovery method based on storage virtualization gateway

technical field

The invention relates to a data-level remote disaster recovery method based on a storage virtualization gateway, and belongs to the technical field of information system data disaster recovery.

Background technique

With the rapid development of enterprises, the volume of business data continues to rise, which records the operation of enterprises and provides important information to decision makers. At present, most enterprises store data centrally in the data center for management. While improving management and business efficiency, the data center also has the risk of being destroyed. Therefore, data disaster recovery has become a must for enterprise development. one of the problems.

However, there are many kinds of storage devices in the enterprise: heterogeneous storage devices of different brands and models cause the problem of "SAN islands" of data, for example, storage devices from different manufacturers such as IBM, EMC, SUN, HP, and HDS are not "compatible" .

The following challenges are mainly faced when establishing an enterprise disaster recovery system:

1. Data is separated in different SAN islands, and various heterogeneous storage devices are not "compatible";

2. Various heterogeneous storages have different degrees of intelligence, and data transmission and sharing across heterogeneous storage devices cannot be realized;

3. Since the data on the heterogeneous storage cannot be shared, it is difficult to realize the overall disaster recovery scattered on each device;

4. The disaster recovery software of each storage manufacturer is different and cannot be used universally;

5. Each storage system has a separate management interface, resulting in management complexity;

6. A certain storage device fails, resulting in the loss of important business data.

Contents of the invention

In order to solve the above-mentioned technical problems, the present invention provides a data-level remote disaster recovery method based on a storage virtualization gateway, which takes over heterogeneous storage devices of different brands and models online through a storage virtualization gateway (SVC) and integrates them into a single storage Pool unified management. Then use the global mirror (Globle Mirror) technology to realize the long-distance data asynchronous replication function, so that when the main data center occurs natural disasters such as fires, earthquakes or service interruptions such as equipment failures and software errors, the business system can obtain data from the disaster recovery data center , to ensure the normal and continuous operation of the business system.

The present invention adopts the following technical solutions for solving the problems of the technologies described above:

The present invention provides a data-level remote disaster recovery method based on a storage virtualization gateway. The method takes over heterogeneous storage devices of different brands and models online through the storage virtualization gateway SVC, and integrates them into a single storage pool for unified management; reuse The Global Mirror technology realizes the long-distance data asynchronous replication function, which specifically includes the following steps:

(1) Integration of heterogeneous storage: build a unified and reasonable highly scalable storage architecture, integrate multiple storage systems of different brands into a single storage pool, and simplify management through the graphical interface GUI;

(2) Off-site disaster recovery for heterogeneous storage: realize long-distance asynchronous replication through global mirroring, and provide business continuity protection for two sites.

As a further optimization solution of the present invention, in step (1), the specific method for integrating heterogeneous storage is as follows:

(1-1) The storage virtualization gateway SVC joins the existing heterogeneous storage environment: for the existing storage volume Lun in the user environment, the existing disk configuration is inherited intact through the Image mode operation mode;

(1-2) Virtualize existing data: When the storage virtualization gateway SVC joins the user's existing heterogeneous storage environment, all the original disk storage devices are uniformly allocated and managed by the SVC, and the original disk data is migrated through the SVC transparently striped into virtual volumes;

(1-3) Dynamically create and allocate virtual volumes according to the needs of applications: When the storage space requirements of a host application change, SVC dynamically allocates storage space on multiple storage devices under the condition of non-stop applications.

As a further optimization solution of the present invention, in step (2), the specific method of the remote disaster recovery of the heterogeneous storage is as follows:

(2-1) Global mirroring replication is the basis for realizing heterogeneous storage disaster recovery. Through fiber channel, the data on the local storage is asynchronously mirrored to the remote storage device in a unified form with the logical volume as the basic unit;

(2-2) When the server writes to the mirrored volume, the SVC writes the data into two copies at the same time, that is, the data is respectively written into the storage device of the virtualized mirror;

(2-3) When the server reads the data in the mirrored volume, the SVC selects one of the copies for reading;

(2-4) If one of the mirrored volumes is unavailable, the other copy still provides data read and write, and the SVC records the write operation of the virtual volume through the bitmap, and automatically synchronizes the data once the unavailable virtual volume recovers;

(2-5) When the host application data is written into the local SVC cache, the system immediately returns an I/O completion indication to the host application, and the data is forwarded to the remote SVC cache through Global Mirror within a period of time.

As a further optimization scheme of the present invention, in step (2), long-distance asynchronous replication is realized through global mirroring, providing business continuity protection with a maximum distance of 800KM between two sites.

As a further optimization solution of the present invention, the length of a period of time in step (2-5) depends on the bandwidth and distance between the two data centers.

As a further optimization solution of the present invention, the length of the period of time is within one minute.

Compared with the prior art, the present invention adopts the above technical scheme and has the following technical effects:

1. Improve data integration, fully integrate existing heterogeneous storage, lay a solid foundation for data disaster recovery, and achieve high availability of storage to ensure business continuity;

2. Make full use of the old and protect the existing investment, supporting up to more than 120 different brands and types of disk storage systems;

3. Improve the administrator's work efficiency by about 2 times, and the administrator can manage the disaster recovery process through the GUI (graphical user interface);

4. Integrated storage management, reducing system complexity and saving about 50% of management costs;

5. Provide a disaster recovery range with a maximum distance of 800KM between two sites.

detailed description

The technical scheme of the present invention is described in further detail below:

The data-level remote disaster recovery method based on the storage virtualization gateway of the present invention takes over heterogeneous storage devices of different brands and models online through the storage virtualization gateway (SVC), and integrates them into a single storage pool for unified management. Then use the global mirror (Globle Mirror) technology to realize the long-distance data asynchronous replication function, so that when the main data center occurs natural disasters such as fires, earthquakes or service interruptions such as equipment failures and software errors, the business system can obtain data from the disaster recovery data center , to ensure the normal and continuous operation of the business system.

The present invention adopts following technical scheme:

(1) Integration of heterogeneous storage. Through the storage virtualization gateway (SVC), the storage of different brands and different interfaces has the same function, presents the same interface to the application, and integrates multiple storage systems of different brands and models into a single storage pool for unified management. Through the GUI (graphical interface) ) to simplify management;

(2) Remote disaster recovery of heterogeneous storage. After integrating heterogeneous storage, use Global Mirror technology (GlobalMirror) to realize remote disaster recovery, transmit the data of the primary data center to the disaster recovery data center through a dedicated network, and provide business continuity protection with a maximum distance of 800KM between two sites.

In step (1), the specific method for integrating heterogeneous storage is as follows:

(1-1) Storage virtualization gateway (SVC) joins the existing heterogeneous storage environment: For the existing storage volume (Lun) in the user environment, the existing storage volume (Lun) can be transferred through the Image mode operation mode without data migration. The disk configuration is inherited intact, so it is completely transparent to the application on the host;

(1-2) Virtualize the existing data: When the storage virtualization gateway (SVC) joins the user's existing heterogeneous storage environment, all the original disk storage devices will be allocated and managed by the SVC from a separate coup. Data is transparently striped into virtual volumes through the migration function of SVC;

(1-3) Dynamically create and allocate virtual volumes according to the needs of applications: When the storage space requirements of a host application change, SVC can dynamically allocate storage space directly on multiple storage devices without stopping the application .

In step (2), the specific method of the remote disaster recovery of the heterogeneous storage is as follows:

(2-1) Global Mirror (Global Mirror) replication is the basis for realizing heterogeneous storage disaster recovery. Through Fiber Channel, with logical volumes as the basic unit, the data on the local storage is asynchronously mirrored to the remote storage device in a unified form. ;

(2-2) When the server writes to the mirrored volume, the SVC writes the data into two copies at the same time. That is, the data is respectively written into the storage device of the virtualized image;

(2-3) When the server reads the data in the mirror volume, SVC will select one of the copies to read;

(2-4) If one of the mirrored volumes is unavailable, the other copy still provides data read and write, and the SVC records the write operation of the virtual volume through the bitmap, and automatically synchronizes the data once the unavailable virtual volume recovers;

(2-5) When the host application data is written into the local SVC cache, the system immediately returns an I/O completion indication to the host application, and the data will be processed within a period of time (usually a few seconds to a minute in practice, depending on the two data centers) The bandwidth and distance between) forward the data to the remote SVC cache through the global mirror (Global Mirror).

The above is only a specific implementation mode in the present invention, but the scope of protection of the present invention is not limited thereto. Anyone familiar with the technology can understand the conceivable transformation or replacement within the technical scope disclosed in the present invention. All should be covered within the scope of the present invention, therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims (6)

1. A data-level remote disaster recovery method based on a storage virtualization gateway, characterized in that the method takes over heterogeneous storage devices of different brands and models online through the storage virtualization gateway SVC, and integrates them into a single storage pool for unified management ; Then use the global mirror technology to realize the long-distance data asynchronous replication function, which specifically includes the following steps: (1) Integration of heterogeneous storage: build a unified and reasonable highly scalable storage architecture, integrate multiple storage systems of different brands into a single storage pool, and simplify management through the graphical interface GUI; (2) Off-site disaster recovery for heterogeneous storage: realize long-distance asynchronous replication through global mirroring, and provide business continuity protection for two sites. 2. A data-level remote disaster recovery method based on storage virtualization gateway according to claim 1, characterized in that, in step (1), the specific method of integrating heterogeneous storage is as follows: (1-1) The storage virtualization gateway SVC joins the existing heterogeneous storage environment: for the existing storage volume Lun in the user environment, the existing disk configuration is inherited intact through the Image mode operation mode; (1-2) Virtualize existing data: When the storage virtualization gateway SVC joins the user's existing heterogeneous storage environment, all the original disk storage devices are uniformly allocated and managed by the SVC, and the original disk data is migrated through the SVC transparently striped into virtual volumes; (1-3) Dynamically create and allocate virtual volumes according to the needs of applications: When the storage space requirements of a host application change, SVC dynamically allocates storage space on multiple storage devices under the condition of non-stop applications. 3. A data-level off-site disaster recovery method based on storage virtualization gateway according to claim 1, characterized in that, in step (2), the specific method of off-site disaster recovery for heterogeneous storage is as follows: (2-1) Global mirroring replication is the basis for realizing heterogeneous storage disaster recovery. Through fiber channel, the data on the local storage is asynchronously mirrored to the remote storage device in a unified form with the logical volume as the basic unit; (2-2) When the server writes to the mirrored volume, the SVC writes the data into two copies at the same time, that is, the data is respectively written into the storage device of the virtualized mirror; (2-3) When the server reads the data in the mirrored volume, the SVC selects one of the copies for reading; (2-4) If one of the mirrored volumes is unavailable, the other copy still provides data read and write, and the SVC records the write operation of the virtual volume through the bitmap, and automatically synchronizes the data once the unavailable virtual volume recovers; (2-5) When the host application data is written into the local SVC cache, the system immediately returns an I/O completion indication to the host application, and the data is forwarded to the remote SVC cache through Global Mirror within a period of time. 4. A kind of data-level off-site disaster recovery method based on storage virtualization gateway according to claim 1, characterized in that, in step (2), long-distance asynchronous replication is realized through global mirroring, providing a maximum two-site distance of 800KM Business continuity protection. 5. A data-level off-site disaster recovery method based on storage virtualization gateway according to claim 3, characterized in that the length of a period of time described in step (2-5) depends on the distance between two data centers bandwidth and distance. 6. The data-level remote disaster recovery method based on storage virtualization gateway according to claim 5, characterized in that, the length of the period of time is within one minute.