WO2011103791A1

WO2011103791A1 - Backup storage system, backup system, data backup method and recovery method

Info

Publication number: WO2011103791A1
Application number: PCT/CN2011/071070
Authority: WO
Inventors: 汪野; 方德辉; 何非
Original assignee: 成都市华为赛门铁克科技有限公司
Priority date: 2010-02-24
Filing date: 2011-02-18
Publication date: 2011-09-01
Also published as: CN101794246B; CN101794246A

Abstract

A backup storage system, a backup system, a data backup method and a recovery method are provided. The backup storage system comprises N storage subsystems, wherein an i(th) storage subsystem is connected with an i+1(th) storage subsystem, the performance of the i(th) storage subsystem is superior to that of the i+1(th) storage subsystem, the capacity of the i(th) storage subsystem is less than that of the i+1(th) storage subsystem, the cost of the unit capacity of the i(th) storage subsystem is higher than that of the i+1(th) storage subsystem, wherein the N is not less than 2, and the i is not less than 1 and not more than N-1. The technical solution can effectively improve the performance of the system without increasing the total cost of the system significantly, thereby improving the cost performance of the system.

Description

Backup storage system, backup system and data backup method and recovery method The application is submitted to the Chinese Patent Office on February 24, 2010, the application number is 201010114868.6, and the invention name is "backup storage system, backup system and data backup method and recovery method" The priority of the Chinese Patent Application, the entire contents of which is incorporated herein by reference. The present invention relates to the field of data backup technologies, and in particular, to a backup storage system, a backup system, and a data backup method and a recovery method.

Background technique

With the development of information technology, the security of data is becoming more and more important. However, hardware failures, software corruption, virus attacks, hacking, erroneous operations, and other unexpected causes are constantly threatening the data in the application. By using backup to save another copy of the data in the application system, when the data in the application system is lost, the data copy is used to restore the data in the application system, thereby effectively avoiding the loss of data loss. Therefore, backup is a very effective way to improve data security.

With the rapid increase of the amount of data in the existing business system, in order to effectively ensure the security of the business data, the business system performs a backup every day, but because the data in the system needs to be read during the backup, it is often required to consume a large number of application systems. Resources, if data backup is performed when the system business is busy, not only seriously affects the system business performance, but also the backup performance is not good. Therefore, in the prior art, the backup window (the backup window is an ideal backup window) is used as the backup time (after the backup window is the ideal backup window) after the first day of work to the next day, to effectively complete the backup of the data in the business system.

Traditional backup storage systems use physical tape libraries to store backup data. The physical tapes in the physical tape library are linear access media. The physical tape library components need to complete a large number of mechanical movements such as robotic arm positioning tapes, tape drive loading tapes, and tape windings to complete a complete write operation to achieve data. Backup. Due to its very low backup performance, backups cannot be completed within the ideal backup window and cannot meet the backup needs of customers. In order to effectively shorten the backup window and improve the performance of the backup, existing In the technology, disk arrays are used instead of physical tape libraries to store backup data. The disk array is a random access medium, the data positioning time is short, and the read and write performance of the disk array is much higher than that of the general physical tape library; the performance of the backup storage system can be improved to meet the requirements of the user for the backup window. However, the use of a disk array as a backup storage system increases the backup performance while causing a significant increase in the cost and power consumption of the backup storage system, making the backup storage system using the disk array less cost-effective.

Summary of the invention

The embodiments of the present invention provide a backup storage system, a backup system, a data backup method, and a recovery method, which can effectively improve the performance of the backup storage system and meet the existing business system without significantly increasing system cost and energy consumption. Demand.

An embodiment of the present invention provides a backup storage system, including:

N storage subsystems, wherein the i-th storage subsystem is connected to the i+1th storage subsystem, and the performance of the i-th storage subsystem is higher than the i+1 storage subsystem, the i-th storage subsystem The capacity of the i-th storage subsystem is higher than that of the i+1th storage subsystem, N > 2, 1 N-1.

The embodiment of the present invention further provides a backup system, including a backup client and a backup server, and the backup storage system. The backup server is configured to receive backup data sent by the client, and store the backup data. In the backup storage system, to complete the storage operation of the backup data; and/or the backup server sends a data request to the backup storage system according to the recovery requirement of the user, to obtain and read from the backup storage system The data requests the corresponding backup data; and sends the corresponding backup data to the client to complete the recovery operation of the backup data.

The embodiment of the invention further provides a data backup method, including:

The first storage subsystem receives and caches the backup data sent by the backup server; sends the copied backup data to the second storage subsystem;

Starting from the second storage subsystem, the i+1th storage subsystem receives and caches the copied backup data sent by the i th storage subsystem, and repeats the step until i=N1; the Nth storage subsystem receives And storing the copied backup data sent by the N-1th storage subsystem, where N is the number of storage subsystems, N > 2, K i < Nl;

Among them, the performance of the i-th storage subsystem is higher than the i+1th storage subsystem, the i-th storage subsystem The capacity of the i-th storage subsystem is higher than that of the i+1th storage subsystem.

The embodiment of the invention further provides a data recovery method, including:

Receiving, by the first storage subsystem, a data request sent by the backup server, and returning the backup data to the backup server when the backup data corresponding to the data request is cached in the first storage subsystem;

When the backup data corresponding to the data request is not cached in the first storage subsystem, assign i to 1 , where l i N-l , N is the number of storage subsystems, N > 2;

The i+1th storage subsystem receives the data request sent by the i th storage subsystem, and when the backup data corresponding to the data request is cached in the i+1th storage subsystem, the backup is performed The information of the data is returned to the backup server for the backup server to recover data; when the backup data corresponding to the data request is not cached in the i+1th storage subsystem, the value of i is assigned to i+1; Perform this step until i=Nl, returning an error message; where li Nl , N is the number of storage subsystems, N > 2.

The backup storage system, the backup system, the data backup method, and the recovery method of the embodiment of the present invention use the N storage subsystems sequentially connected with the read/write performance and the unit capacity cost, and sequentially increase the capacity, as the backup storage system, and The first storage subsystem with the highest read/write performance is used as an interface with the backup server to implement storage and recovery processing of backup data. By adopting the technical solution of the embodiment of the present invention, the performance of the backup system can be effectively improved without significantly increasing the cost and power consumption of the backup system. Thereby effectively improving the cost performance of the backup storage system.

BRIEF DESCRIPTION OF THE DRAWINGS In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, a brief description of the drawings used in the embodiments or the prior art description will be briefly described below. The drawings are only some of the embodiments of the present invention, and those skilled in the art can obtain other drawings based on these drawings without any inventive labor.

1 is a structural diagram of a backup storage system according to Embodiment 1 of the present invention; FIG. 2 is a flowchart of a data backup method according to Embodiment 4 of the present invention; 3 is a flowchart of a data recovery method according to Embodiment 5 of the present invention; FIG. 4 is a flowchart of a data recovery method according to Embodiment 6 of the present invention; and FIG. 5 is a structural diagram of a backup system according to Embodiment 7 of the present invention.

detailed description

BRIEF DESCRIPTION OF THE DRAWINGS The technical solutions in the embodiments of the present invention will be described in detail below with reference to the accompanying drawings. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments of the present invention without creative work are within the scope of the present invention.

1 is a structural diagram of a backup storage system according to Embodiment 1 of the present invention; as shown in FIG. 1, the backup storage system of this embodiment includes N storage subsystems.

N storage subsystems, wherein the i-th storage subsystem is connected to the i+1th storage subsystem, and the performance of the i-th storage subsystem is higher than the i+1 storage subsystem, the i-th storage subsystem The capacity of the i-th storage subsystem is lower than that of the i+1th storage subsystem, N > 2, K i < Nl , and N and i are positive integers.

Specifically, this embodiment introduces a backup storage system, a first storage subsystem 11, and a second storage subsystem 12 - until the Nth storage subsystem 13 is sequentially connected. That is to say, each storage subsystem is only connected to its previous nearest neighbor storage subsystem and the next nearest neighbor storage subsystem to implement data exchange. It should be noted that the backup storage system further includes a controller 10 for controlling the exchange of backup data between adjacent storage subsystems.

The first storage subsystem 11 of the backup storage system of this embodiment has the highest read/write performance and the highest unit capacity cost, but its capacity is small. For example, the first storage subsystem 11 can select a small-capacity virtual tape library or a disk array, and with the development of high-tech, a storage medium with high read/write performance such as a FLASH chip array may be used. Considering the storage medium with high read/write performance, the cost is generally very high, and the cost per unit storage capacity is high. In this embodiment, the backup storage system can select a high-performance storage medium with a small capacity as the first storage subsystem 11, and mainly uses For high-speed read and write performance. For the capacity of the first storage subsystem 11, it is usually required to accommodate at least one full backup data set, that is, At least one full backup process can be supported, and its capacity is sufficient to store the backup data needed to complete a backup storage operation. In actual use, the first storage subsystem 11 of a suitable capacity size can be selected according to actual conditions.

The first storage subsystem 11 and the second storage subsystem 12, the third storage subsystem, the Nth storage subsystem 13, and the nearest neighbor storage subsystem read and write performance and units The capacity costs are reduced in turn, and the capacity is increased in turn. From the performance point of view, the first storage subsystem 11 is mainly used to reflect the read and write performance of the backup storage system; the second storage subsystem 12 to the N-1th storage subsystem can function as a cache. Moreover, the read and write performance between the nearest neighbor storage subsystems can be matched to meet the smooth read and write. For example, the current storage subsystem has a read/write performance of 1000 mb/s, and the read/write performance of the nearest neighbor storage subsystem is 100 mb/s. There is too much difference between the two, and normal matching cannot be achieved, and efficient backup cannot be achieved. One or more storage subsystems with read/write performance between 1000 mb/s and 100 mb/s can be added in the middle to achieve normal matching. For example, the second storage subsystem 12 to the N-1 storage subsystem in the middle are used to match the read/write performance of the first storage subsystem 11 and the Nth storage subsystem 13. The read/write performance and unit capacity cost between the nearest neighbor storage subsystems are sequentially reduced, and the capacity is sequentially increased, but a storage subsystem with the same read/write performance and/or unit capacity cost is also possible.

From the perspective of capacity, the Nth storage subsystem 13 is used to represent the storage data of the backup storage system. The first N-1 storage subsystems are used to cache backup data, which is equivalent to the backup data being stored in the N-1 level cache and finally stored in the Nth storage subsystem. From a cost perspective, existing backup storage systems typically contain only one Nth storage subsystem 13 of the backup storage system of the present embodiment. Preferably, the capacity of the first N-1 storage subsystems is small compared to the capacity of the Nth storage subsystem. For example, preferably, the storage capacity of the Nth storage subsystem 13 is tens of thousands of times the total capacity of the first N-1 storage subsystems, and the cost per unit capacity is the first N-1 storage subsystem units. The cost of the capacity is a fraction of the total cost. In this case, the cost of the backup storage system in this embodiment is less than one percent of the total cost due to the addition of the first N-1 storage subsystems. It can be seen that the backup storage system of this embodiment hardly increases the cost compared to the existing backup storage system. In addition, since the capacity of the first N-1 storage subsystems is not large, the total power consumption is not high.

When the first storage subsystem of the backup storage system of this embodiment adopts a disk array, the Nth storage subsystem uses a physical tape library. The backup storage system of this embodiment is compared with the backup storage system of the present embodiment and the backup storage system using the disk array having the same capacity as the Nth storage subsystem. The read and write performance of the disk array can also be achieved, but the cost and power consumption are much smaller than the backup storage system using the disk array of the same capacity.

The backup storage system of this embodiment is configured by sequentially adopting N storage subsystems in which the read/write performance is sequentially decreased, the capacity is sequentially increased, and the unit capacity cost is sequentially increased. It can meet the system performance and capacity requirements, and can effectively improve the performance of the backup storage system without significantly increasing the cost and energy consumption, thereby effectively improving the cost performance of the backup storage system.

Embodiment 2 of the present invention provides a backup storage system. The backup storage system of this embodiment further describes the performance of the backup storage of the backup storage system of the first embodiment on the basis of the backup storage system of the first embodiment.

The first storage subsystem 11 in the backup storage system of the embodiment is configured to receive and cache backup data sent by the backup server; when i+l<N, the i+1th storage subsystem is configured to receive and cache the first The backup backup data sent by the i storage subsystems; when i+l=N, the i+1th storage subsystem is configured to receive and store the copied backup data sent by the i th storage subsystem.

Specifically, the first storage system 11 has the highest read and write performance, and is used to reflect the read and write performance of the entire backup storage system. The first storage subsystem 11 is configured to receive and cache backup data sent by the backup server as an interface between the entire backup storage system and the backup server. Because the capacity of the first backup storage system 11 is small, after receiving the backup data sent by the backup server, the backup data is cached, then the backup data is copied and sent to the second storage subsystem 12, and then After the two storage subsystems 12 receive and cache the backup data, the backup data is copied and sent to the next nearest neighbor backup storage subsystem, and the above operations are performed in sequence until the backup data is finally sent to the Nth storage subsystem.

13. The backup data is stored by the Nth storage subsystem 13. The transmission and reception of backup data between the nearest neighbor storage subsystems is completed under the scheduling of the controller 10.

From the perspective of the entire backup storage system, only the first storage subsystem is connected to the backup server.

11, the read and write performance of the first storage subsystem 11 determines the read and write performance of the entire backup storage system. After the first storage subsystem 11 receives the backup data, the backup data is copied and cached one level at a time, and belongs to the operation inside the backup storage system, and does not affect the overall read and write performance of the backup storage system.

The backup storage system of the present implementation adopts the first storage subsystem with the highest read and write performance as the interface for communicating with the backup server, and then sequentially caches the backup data to the subsequent storage subsystems, and finally stores the backup data. At the Nth storage subsystem. The backup storage system in this embodiment has high backup performance and short backup time; it can meet the requirements of existing business systems. It should be noted that, on the basis of the second embodiment, when the Li N-2, JLN > 3, the i+1th storage subsystem is specifically configured to receive the replication sent by the i th storage subsystem. Backing up the data, overwriting the backup data that is first stored in the i+1th storage subsystem; specifically, when the first storage subsystem 11 to the N-1 storage subsystem When one of the storage subsystems has no free storage space, the storage subsystem directly receives the backup data sent by the last nearest neighbor storage subsystem to directly overwrite the backup data first stored in the storage subsystem. Normally, the first stored backup data has been copied to the next nearest neighbor storage subsystem, meaning no storage, so it can be overwritten. However, the backup data in the storage subsystem, if it has not been copied to the next storage subsystem, must be replicated before it is allowed to be overwritten. By adopting the above solution, it is ensured that the latest backup data is stored in each storage subsystem, thereby effectively ensuring the security of the stored backup data.

When i=N-1, the i+1th storage subsystem is specifically configured to receive the copied backup data sent by the i th storage subsystem, and overwrite the duplicate backup data with the i+1th Backup data that has expired in the storage subsystem.

Specifically, when the Nth storage subsystem 13 has no free storage space, the copied backup data sent by the received N-1th storage subsystem directly overwrites the expired backup in the Nth storage subsystem 13. data. The expired backup data described here refers to data that exceeds the backup storage time. The backup data of the backup storage system has a certain storage time, for example, three months, half a year, or one year. When the backup storage time is exceeded, the backup data is no longer useful, and there is no need to save it. These backup data are called backup data that has expired. With this technical solution, the security of all backup data stored in the backup storage system can be guaranteed.

Embodiment 3 of the present invention provides a backup storage system. The backup storage system of the second embodiment is mainly used for receiving backup data of the backup server and storing the data. This embodiment describes the backup storage system from the perspective of data recovery on the basis of the backup storage system of the first embodiment. The technical solution of the present embodiment will be described in detail below.

The first storage subsystem 11 is configured to receive a data request sent by the backup server. When the backup data corresponding to the data request is cached in the first storage subsystem 11, the first storage subsystem 11 returns the backup data to the backup server.

The i+1th storage subsystem is further configured to receive the data request sent by the i th storage subsystem when the backup data corresponding to the data request is not cached in the first storage subsystem , And when the backup data corresponding to the data request is cached in the (i+1)th storage subsystem, returning the backup data to the backup server.

Specifically, when the backup data corresponding to the data request is not cached in the first storage subsystem 11, the data request is sent to the second storage subsystem 12 to be in the second storage subsystem 12. Obtaining backup data corresponding to the data request. If the backup data corresponding to the data request is still not cached in the second storage subsystem 12, the data request is sent to the next nearest neighbor storage subsystem; and so on, until the data request is sent to the A target storage subsystem that caches the backup data corresponding to the data request; and then the target storage subsystem returns the backup data corresponding to the data request to the backup server. For convenience of description, the target storage subsystem in the embodiment of the present invention is a storage subsystem that caches backup data corresponding to the data request.

It should be noted that, in the embodiment, the target storage subsystem returns the backup data corresponding to the data request to the backup server, which is a level-1 return, that is, the target storage subsystem first returns the backup data corresponding to the data request. The previous nearest neighbor storage subsystem of the target storage subsystem is then forwarded back by the previous nearest neighbor storage subsystem of the target storage subsystem. That is, each storage subsystem returns the backup data corresponding to the data request sent by the next nearest neighbor storage subsystem to its previous nearest neighbor storage subsystem until the backup data corresponding to the data request is returned to the first The storage subsystem is finally returned to the backup server by the first storage subsystem.

If the data request is sent to the Nth storage subsystem 13 by the N-1th storage subsystem, but the Nth storage subsystem 13 still does not store the backup data corresponding to the data request, N storage subsystems return error information to the backup server. The return of the error message is the same as the return of the backup data corresponding to the above data request, and is also a level one return, and finally returned by the first storage subsystem 11 to the backup server, telling the backup server that the data corresponding to the recovery request cannot be found.

The transmission of the data request during the data recovery process and the return of the backup data corresponding to the data request are all performed under the scheduling of the controller 10. Since the write data operation is performed in the storage process, the storage record of the backup data is saved in the controller 10. When the data read operation is performed during the process of restoring data, the storage record of the backup data saved in the controller 10 can be referred to, so as to more conveniently find the backup data corresponding to the data request. Preferably, however, the backup data corresponding to the data request is returned to the first level, until the first storage subsystem 11 is returned to the backup server by the first storage subsystem. In the disaster recovery application, the storage subsystem storing the backup data corresponding to the data request may be found according to the storage record of the backup data saved in the controller 10, and the storage subsystem may be directly transported to the storage subsystem. The backup server is localized and docked with the backup server to implement backup data recovery.

In the backup storage system of the embodiment of the present invention, the first storage subsystem 11 communicates with the backup server, receives a data request sent by the server, or returns the requested backup data to the server, so as to achieve efficient recovery of the backup data and satisfy the existing backup. The storage system's demand for business.

It should be noted that, based on the foregoing technical solution of the embodiment, the i+1th storage subsystem is further configured to: when the backup data corresponding to the data request is not cached in the first storage subsystem 11 Receiving, in the middle, the data request sent by the i th storage subsystem, when the backup data corresponding to the data request is cached in the i+1th storage subsystem, storing the i+1th storage The subsystem stores the address of the backup data back to the backup server.

Specifically, when the backup data corresponding to the data request is not cached in the first storage subsystem 11, the first storage subsystem 11 sends the data request to the second storage subsystem 12, if the second storage subsystem 12, the backup data corresponding to the data request is cached, and the second storage subsystem 12 returns the address of the backup data corresponding to the data request (that is, the pointer of the backup data corresponding to the data request) to the backup server. If the backup data corresponding to the data request is still not cached in the second storage subsystem 12, the data request is sent to the next nearest neighbor storage subsystem; and so on, until the data request is sent to the a target storage subsystem that caches the backup data corresponding to the data request; and then returns an address of the backup data corresponding to the data request in the target storage subsystem (ie, a pointer of the backup data corresponding to the data request) to the backup server. Correspondingly, the backup server returns the channel formed in the process of returning to the backup server according to the address of the corresponding backup data (that is, the pointer of the backup data corresponding to the data request), and restores the data.

The process of returning the address of the backup data corresponding to the data request corresponding to the data request in the target storage subsystem (that is, the pointer of the backup data corresponding to the data request) to the backup server is returned by each storage subsystem level first level, because each A storage subsystem is only connected to the nearest storage subsystem and the nearest nearest neighbor storage subsystem of its nearest neighbor for data exchange. Therefore, after each storage subsystem receives the address sent by the next nearest neighbor storage subsystem, the storage subsystem caches the address and returns the address in the storage subsystem for caching the address to the previous nearest neighbor storage. Subsystem. Such a level of return until the second storage subsystem 12 returns an address to the first storage subsystem 11, and then the first storage subsystem 11 will store the address of the address returned by the second storage subsystem 12. Returning to the backup server, the process of returning the address forms a channel from the target storage subsystem to the backup server, and the backup server can find the backup data corresponding to the recovery request according to the channel. The backup storage system of the embodiment can quickly and efficiently find the backup data corresponding to the backup server recovery request, and return the pointer corresponding to the target storage subsystem of the backup data corresponding to the recovery request to the backup server, thereby directly The data is restored in the target storage subsystem that stores the backup data corresponding to the recovery request. The backup data corresponding to the recovery request does not need to be copied and returned step by step, and the data recovery requirement of the existing service system can be satisfactorily satisfied.

The backup storage system embodiments described above are merely illustrative, wherein the subsystems described as separate components may or may not be physically separate, that is, may be located in one place, or may be distributed in multiple locations. Place. Some or all of the subsystems may be selected according to actual needs to achieve the objectives of the solution of the embodiment. Those of ordinary skill in the art can understand and implement without undue labor.

2 is a flowchart of a data backup method according to Embodiment 4 of the present invention; as shown in FIG. 2, the data backup method in this embodiment may specifically include the following steps:

Step 100: The first storage subsystem receives and caches backup data sent by the backup server, and sends the copied backup data to the second storage subsystem.

Step 101: Starting from i=l, the i+1th storage subsystem receives and caches the copied backup data sent by the i th storage subsystem, and repeats the step until i=N1, and the Nth storage subsystem receives And storing the copied backup data sent by the N-1th storage subsystem; N is the number of storage subsystems, N > 2, K i Nl , N and i are positive integers;

The performance of the i-th storage subsystem is higher than that of the i+1th storage subsystem, and the capacity of the i-th storage subsystem is smaller than the i+1th storage subsystem, and the unit capacity of the i-th storage subsystem is high. On the i+1th storage subsystem.

Specifically, the data backup method of the embodiment is a technical solution for implementing the backup storage performance of the second embodiment in the backup storage system provided in the first embodiment. The first storage subsystem 11 is configured to receive backup data sent by the backup server, and cache the backup data, and then copy the backup data to the second storage subsystem 12, starting from the second storage subsystem 12, each storage sub- The system receives and caches the replicated backup data sent by the previous nearest neighbor storage subsystem, and then copies the backup data and sends it to the next nearest neighbor backup storage subsystem until it is finally backed up by the N-1th storage subsystem. The data is sent to the Nth storage subsystem 13, and the Nth storage subsystem 13 stores the backup data.

It can be seen from the above that the first storage subsystem 11 to the N-1 storage subsystem are used to cache the backup data; the Nth storage subsystem 13 is configured to store the backup data. Because of the first storage subsystem 11 As an interface between the entire backup storage system and the backup server, the read and write performance of the first storage subsystem 11 is equivalent to the read and write performance of the entire backup storage system. Starting from the second storage subsystem 12, the performance and unit capacity costs of each storage subsystem are sequentially reduced, but the capacity is sequentially increased. Until the Nth storage subsystem 13 mainly reflects the function of the memory of the backup storage system, the performance and unit capacity cost are the lowest, but the storage capacity is the largest. For the backup storage system, reference may be made to the foregoing embodiment 1, and details are not described herein again.

With the storage method of this embodiment, the first storage subsystem 11 is used as an interface between the backup storage system and the backup server, and the backup data sent by the backup server is received and cached. Because the first storage subsystem 11 has high read and write performance. Therefore, the entire backup storage system also has high backup performance.

The mechanism used in the data backup method of this embodiment is the same as that of the first embodiment and the second embodiment. For details, refer to the descriptions of the first embodiment and the second embodiment.

The data backup method of the embodiment is used to store backup data for a backup storage system composed of N storage subsystems in which the read/write performance and the unit capacity cost are sequentially decreased, the storage capacity is sequentially increased, and the logically connected N storage subsystems are sequentially stored; The first storage subsystem with write performance serves as an interface for communication with the backup server, improving the performance of backup processing, effectively shortening the backup window, and meeting the needs of existing business systems.

It should be noted that, in the foundation of the fourth embodiment, it may also include:

When l<i<N-2, N>3, the i+1th storage subsystem receives the copied backup data sent by the i th storage subsystem; when the i+1th storage subsystem has no free storage space And the i+1th storage subsystem directly overwrites the backup data stored in the i+1th storage subsystem.

Specifically, when one of the first storage subsystem 11 to the N-1 storage subsystem has no free storage space, the storage subsystem will receive the replication sent by the last nearest neighbor storage subsystem. The backup data directly covers the backup data that was first stored in the storage subsystem. Normally, the first backup data stored has been copied to the next nearest neighbor storage subsystem, meaning no storage, so it can be overwritten. However, the backup data in the storage subsystem, if it has not been copied to the next storage subsystem, must be replicated before it is allowed to be overwritten. With the above solution, it is ensured that the latest backup data is stored in each storage subsystem, and the security of the stored backup data is effectively ensured.

When i=N1, the i+1th storage subsystem receives and caches the copy sent by the i th storage subsystem Backing up data; when the i+1th storage subsystem has no free storage space, the i+1th storage subsystem overwrites the duplicated backup data in the i+1th storage subsystem .

Specifically, when the Nth storage subsystem 13 has no free storage space, the copied backup data sent by the received N-1th storage subsystem directly overwrites the backup data that has expired in the Nth storage subsystem 13. . The expired backup data described here refers to backup data that exceeds the backup storage time. The backup data of the backup storage system has a certain backup storage time, such as three months, six months, or one year, when the backup storage is exceeded. At the time, the corresponding backup data is no longer useful, and there is no need to save it. These backup data are called backup data that has expired. With this technical solution, the security of all backup data stored in the backup storage system can be guaranteed.

FIG. 3 is a flowchart of a data recovery method according to Embodiment 5 of the present invention. The present embodiment provides a technical solution for implementing data recovery in the backup storage system of Embodiment 3 in detail in the foregoing embodiment. As shown in FIG. 3, the data recovery method in this embodiment may specifically include the following steps:

Step 200: The first storage subsystem receives a data request sent by the backup server, and returns the backup data to the backup server when the backup data corresponding to the data request is cached in the first storage subsystem. ;

Specifically, the first storage subsystem 11 is configured to receive a data request sent by the backup server. When the backup data corresponding to the data request is cached in the first storage subsystem, the first storage subsystem 11 returns the backup data to the backup server.

According to the storage scheme of the embodiment of the present invention described in the second embodiment, when data is stored, it is sequentially cached from front to back in the N storage subsystems. However, the capacity of the storage subsystem is incremented. When a storage subsystem has no storage space, it uses the backup data stored first to ensure the security of the stored backup data. It can be seen that the storage subsystem with the N storage subsystems has a smaller capacity, and generally stores the latest backup data, because some backup data with a longer storage time has been overwritten. The lower the storage subsystem, the larger the capacity, the smaller the probability of overwriting the backup data, and the more commonly saved backup data. Therefore, when data recovery is performed, when the backup data corresponding to the recovery request is the latest backup data, it may be found in the first storage subsystem 11, and then the backup data corresponding to the recovery request may be returned to the backup server. When the backup data corresponding to the recovery request is not in the first storage subsystem 11, the backup data may be stored long ago, that is, the storage time is long. At this time, the following steps 201 and 202 are performed to recover the data.

Step 201: When the backup data corresponding to the data request is not cached in the first storage subsystem In the system, assign i to 1;

Where: l i N-l , N is the number of storage subsystems, N > 2;

Step 202: The i+1th storage subsystem receives the data request sent by the i th storage subsystem, and when the backup data corresponding to the data request is cached in the i+1th storage subsystem, Returning the backup data to the backup server, ending; when the backup data corresponding to the data request is not cached in the i+1th storage subsystem, assigning i to i+1; repeating step 202 until i=N, returns an error message, and ends.

Specifically, in step 201 and step 202, when the backup data corresponding to the recovery request is not in the first storage subsystem 11, the value of i is first set to 1, and the execution of the loop is performed to send the data request to the first level. Until the backup data corresponding to the data request is found or until the Nth storage subsystem is found, it is still not found. Here, i is the same as Embodiment 3 above, and N is used to indicate the number of storage subsystems, N > 2; i represents the label of one of the storage subsystems, where 1 N-1.

After the value of i is set to 1, the second storage subsystem 12 receives the data request sent by the first storage subsystem 11, and when the backup data corresponding to the data request is cached in the second storage subsystem 12, the backup is performed. The data is returned to the backup server. Here, the backup data return process is specifically performed by the second storage subsystem 12 to return the backup data to the first storage subsystem 11, and then returned to the backup server by the first storage subsystem 11.

If there is still no backup data corresponding to the cache data request in the second storage subsystem 12, increment i by 1, and continue to execute the "i+1th storage subsystem receives the data request sent by the i th storage subsystem. When the backup data corresponding to the data request is cached in the i+1th storage subsystem, the backup data is returned to the backup server; and the process of returning backup data is similar to the foregoing embodiment 3, The level returns forward until the backup data is returned to the first storage subsystem 11, and then returned to the backup server by the first storage subsystem 11. If the i+1th storage subsystem still does not have the corresponding backup data of the data request, continue to perform i incrementing by 1, and so on, and continue to send a data request to the next nearest neighbor storage subsystem until a data request is corresponding to the storage. The target backup storage subsystem of the backup data receives the data request and finds the backup data to be restored. Then, the target backup storage system returns the backup data to the backup server, and the process of returning the backup data is similar to the foregoing embodiment 3, and the first level returns forward until the backup data is returned to the first storage subsystem 11. It is returned to the backup server by the first storage subsystem 11.

If the data request is not sent to the Nth storage subsystem by the N-1th storage subsystem 13 . However, when the Nth storage subsystem 13 still does not store the backup data corresponding to the data request, an error message is returned to the backup server. The return of the error message is the same as the return of the backup data described above, and is also a return of the first level, and finally returned by the first storage subsystem 11 to the backup server, telling the backup server that the data corresponding to the recovery request cannot be found.

The implementation mechanism of the data recovery method in this embodiment is the same as that in the foregoing embodiment 3. For details, refer to the description of the third embodiment.

The data recovery method of the embodiment of the present invention can support the high-efficiency recovery of the backup data by using the first storage subsystem 11 to communicate with the backup server, receiving the data request sent by the server or returning the requested backup data to the server. There is a need for a business system.

4 is a flowchart of a data recovery method according to Embodiment 6 of the present invention. The data recovery method of this embodiment is different from the data recovery method of the foregoing Embodiment 5 in that the return pointer is used in this embodiment. As shown in FIG. 4, the data recovery method in this embodiment may specifically include the following steps:

Step 300: The first storage subsystem receives a data request sent by the backup server, and when the backup data corresponding to the data request is cached in the first storage subsystem, returning the address of the backup data to the Backup server

Specifically, after the first storage subsystem receives the data request sent by the backup server, and determines that the backup data corresponding to the data request is cached in the first storage subsystem, the first storage subsystem addresses the backup data. Return to the backup server for the backup server to recover data based on this address.

It should be noted that, here, the pointer corresponding to the first storage subsystem may be returned to the backup server. The data is restored by the backup server based on the channel established by the return pointer.

Step 301: When the backup data corresponding to the data request is not cached in the first storage subsystem, assign i to 1;

Where l i N-l , N is the number of storage subsystems, N > 2;

Step 302: The i+1th storage subsystem receives the data request sent by the i th storage subsystem, and when the backup data corresponding to the data request is cached in the i+1th storage subsystem, The address of the backup data corresponding to the data request is returned to the backup server, for the backup server to recover data, and ends; when the backup data corresponding to the data request is not cached in the i+1th storage subsystem, Assign i to i+1 and repeat step 302; until i=N, return an error message and end.

Specifically, the steps 301 and 302 are different from the steps 201 and 202 in the foregoing embodiment, in which, if the backup data corresponding to the recovery request is stored in a target storage subsystem, when the recovery is found, When the corresponding backup data is obtained, the backup data address corresponding to the data request is returned to the backup server. Similar to the return of the aforementioned backup data, the address return here is also a level one return. Each storage subsystem receives the address sent by the nearest nearest neighbor storage subsystem and caches the address, and then sends the address of the address sent by the cached nearest neighbor storage subsystem to the previous nearest neighbor storage subsystem; until the first The storage subsystem returns the address used to store the address sent by the second storage subsystem to the backup server.

From the target storage subsystem that stores the corresponding backup data to the backup server, a channel is formed according to the return of the address. The backup server returns the backup data corresponding to the data request according to the received address and returns the channel formed by the process.

If the data request is not sent to the Nth storage subsystem by the N-1th storage subsystem

13 , but the Nth storage subsystem 13 still does not store the backup data corresponding to the data request, at which time the Nth storage subsystem returns an error message to the backup server. The return of the error information is the same as that of the third embodiment described above, and details are not described herein again.

The data recovery method of the embodiment can quickly and efficiently find the data requested by the backup server, and return the pointer corresponding to the target storage subsystem of the backup data corresponding to the recovery request to the backup server, thereby directly recovering from the storage. The data is restored in the target storage subsystem of the corresponding backup data; the backup data corresponding to the recovery request does not need to be copied and returned step by step, which can well meet the data recovery requirements of the existing service system.

FIG. 5 is a structural diagram of a backup system according to Embodiment 7 of the present invention; as shown in FIG. 5, the backup system of this embodiment includes a backup client 20, a backup server 21, and a backup storage storage system 22.

The backup server 21 is configured to receive the backup data sent by the client 20, and store the backup data in the backup storage system 22 to complete the storage operation of the backup data; and/or the backup server 21 restores the request according to the data of the user. And sending a data request to the backup storage system 22 to obtain backup data corresponding to the data request from the backup storage system 22, and send the corresponding backup data to the client 20 to complete the recovery operation of the backup data.

Specifically, when a write operation, that is, backup data is performed, the backup client 20 reads data from the business system, and transmits the read data to the backup server 21. After receiving the data sent by the client, the backup server 21 stores the data in the backup storage system 22 to complete the operation of storing the backup data. When When the row read operation is to restore the data, the backup server 21 sends a data request to the backup storage system 22 according to the user's recovery request. After receiving the data request sent by the backup server 21, the backup storage system 22 performs a data recovery operation in each storage subsystem to acquire backup data corresponding to the recovery request. The backup storage system 22 then returns the acquired backup data to the backup server 21, which in turn sends the backup data to the client 20. Finally, the backup client 20 writes the backup data into the service system to complete the operation of restoring the backup data.

The backup storage system 22 of the present embodiment may be configured with the backup storage system of any one of the foregoing implementations to the third embodiment, and the data backup method may be implemented by using the data backup method described in the fourth embodiment. The data recovery method described in Embodiment 6 implements recovery of backup data. For details, refer to the related descriptions of the first embodiment to the sixth embodiment, and details are not described herein again.

The backup system of this embodiment can effectively improve the performance of the backup system, shorten the backup window, and meet the requirements of the service system by adopting the above backup storage system, without significantly increasing the system cost and energy consumption.

Through the description of the above embodiments, those skilled in the art can clearly understand that the present invention can be implemented by means of software plus a necessary hardware platform, and of course, all can be implemented by hardware, but in many cases, the former is better. Implementation. Based on such understanding, all or part of the technical solution of the present invention contributing to the background art may be embodied in the form of a software product, which may be stored in a storage medium such as a ROM/RAM, a magnetic disk, an optical disk, or the like. A number of instructions are included to cause a computer device (which may be a personal computer, server, or network device, etc.) to perform the methods described in various embodiments of the present invention or portions of the embodiments. It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to be limiting, although the present invention will be described in detail with reference to the preferred embodiments. The technical solutions of the present invention may be modified or equivalently substituted, and the modified technical solutions may not deviate from the spirit and scope of the technical solutions of the present invention.

Claims

Rights request

A backup storage system, comprising:

2. The backup storage system of claim 1 wherein:

The first storage subsystem is configured to receive and cache backup data sent by the backup server; when i+l<N, the i+1th storage subsystem is configured to receive and cache the replication sent by the i th storage subsystem Backup data;

When i+l=N, the i+1th storage subsystem is configured to receive and store the copied backup data sent by the i th storage subsystem.

3. The backup storage system of claim 2, wherein

When li N-2, N > 3, the i+1th storage subsystem is specifically configured to receive the copied backup data sent by the i th storage subsystem, and overwrite the duplicate backup data first. Backup data in the i+1th storage subsystem;

When i=N1, the i+1th storage subsystem is specifically configured to receive the copied backup data sent by the i th storage subsystem, and overwrite the duplicated backup data with the i+1th storage. Backup data that has expired in the subsystem.

4. The backup storage system according to claim 2 or 3, characterized in that

The first storage subsystem is configured to receive a data request sent by the backup server, and when the backup data corresponding to the data request is cached in the first storage subsystem, return the backup data to The backup server;

The i+1th storage subsystem is configured to receive the data request sent by the i th storage subsystem when the backup data corresponding to the data request is not cached in the first storage subsystem, Returning the backup data to the backup server or storing the (i+1)th storage subsystem when the backup data corresponding to the data request is cached in the (i+1)th storage subsystem The address of the backup data is returned to the backup server;

An Nth storage subsystem, configured to receive the N-1th storage subsystem when the Nth storage subsystem receives Sending the data request, when the backup data corresponding to the data request is cached in the Nth storage subsystem, returning the backup data to the backup server or the Nth storage subsystem The address storing the backup data is returned to the backup server.

5. The backup storage system of claim 4, wherein

When the Nth storage subsystem does not have backup data corresponding to the data request, an error message is returned to the backup server.

6. A backup system comprising a backup client, a backup server, and a backup storage system according to claim 5;

The backup server is configured to receive backup data sent by the client, and store the backup data in the backup storage system to complete a storage operation of backup data; and/or according to a recovery requirement of the user The backup storage system sends a data request to obtain backup data corresponding to the data request from the backup storage system; and sends the corresponding backup data to the client to complete the recovery operation of the backup data.

The backup storage system according to claim 6, wherein the backup system further comprises:

And a controller, configured to control sending and receiving of backup data between adjacent storage subsystems in the backup storage system.

8. A data backup method, comprising:

Starting from i=l, the i+1th storage subsystem receives and caches the copied backup data sent by the i th storage subsystem, and repeats the step until i=N1, and the Nth storage subsystem receives and stores the first The replicated backup data sent by N-1 storage subsystems, N is the number of storage subsystems, N > 2, K i < N-1 , N and i are positive integers;

The data backup method according to claim 8, wherein the i+1th storage sub-system receives and caches the copied backup data sent by the i-th storage subsystem, including:

When li N-2, N > 3, the i+1th storage subsystem receives the ith storage subsystem to send The copied backup data; when the i+1th storage subsystem has no free storage space, the i+1th storage subsystem stores the copied backup data coverage first in the i+1th Backup data in the storage subsystem;

When i=N-1, the i+1th storage subsystem receives and caches the copied backup data sent by the i th storage subsystem; when the i+1th storage subsystem has no free storage space, the i+ A storage subsystem overwrites the duplicated data in the i+1th storage subsystem that has expired.

10. A data recovery method, comprising:

When the backup data corresponding to the data request is not cached in the first storage subsystem, assign i to 1;

The i+1th storage subsystem receives the data request sent by the i th storage subsystem, and when the backup data corresponding to the data request is cached in the i+1th storage subsystem, the data is backed up. The information is returned to the backup server for the backup server to recover data; when the backup data corresponding to the data request is not cached in the i+1th storage subsystem, i is updated to i+1, and the execution is repeated. Steps, until i=N, return an error message; where li Nl , N is the number of storage subsystems, N > 2.

The data recovery method according to claim 10, wherein the information of the backup data comprises: an address of the backup data or the backup data.