CN106657356A - Data writing method and device for cloud storage system, and cloud storage system - Google Patents

Abstract

The invention discloses a data writing method and device of a cloud storage system. The storage nodes of the cloud storage system include at least a high-speed storage device and a large-capacity storage device. By receiving a write request sent by the client node, the client node is The data to be written after erasure code processing is written to the high-speed storage device; after the writing is completed, a completion response is sent to the client node; the persistence request sent by the client node is received, and the data to be written is persistently written into the mass storage device. The present invention completes data writing by writing high-speed storage devices and writing large-capacity storage devices in two steps, which can not only ensure the data reliability and consistency of the cloud storage system, but also reduce the delay of writing operations while providing large storage capacity . And completing the erasure code processing on the client node can reduce the pressure on the server and reduce the amount of data in network communication. In addition, the present invention also provides a cloud storage system with the above technical advantages.

Description

Data writing method, device and cloud storage system of a cloud storage system

technical field

The present invention relates to the technical field of data storage, in particular to a data writing method and device for a cloud storage system and a cloud storage system.

Background technique

With the rapid development of the mobile Internet and the Internet of Things, personal data, family data, and enterprise data have all shown explosive growth, and the era of massive data has quietly arrived. At the same time, these data have been transformed from purely relational data to non-relational data, such as pictures, audio, video, documents, etc. Traditional storage management methods can no longer meet the storage needs of this type of data. How to effectively store and manage 、 Maintaining this kind of non-relational data has been paid more and more attention by more and more people in the industry.

Cloud storage refers to a system in which a large number of different types of storage devices in the network work together through application software through functions such as cluster applications, grid technology, or distributed file systems, and jointly provide data storage and business access functions to the outside world. Ensure data security and save storage space. Simply put, cloud storage is an emerging solution to put storage resources on the cloud for people to access. Users can connect to the cloud through any network-connected device to access data conveniently at any time and anywhere. Just like the cloud-like wide area network and the Internet, cloud storage does not refer to a specific device for users, but a collection of many storage devices and servers. When users use cloud storage, they do not use a certain storage device, but use a data access service brought by the entire cloud storage system.

With the rapid development of cloud storage technology, cloud storage has become a new data storage solution. In the past two years, domestic cloud storage technology has also achieved vigorous development. Various manufacturers have launched their own cloud storage products, and the field of cloud storage has shown a thriving scene. But generally speaking, domestic cloud storage technology is still in the development stage. In the process of writing big data to existing storage nodes, due to the performance limitation of large-capacity storage devices, it is easy to cause the client to freeze, write operations to be delayed, and affect storage performance.

Contents of the invention

The object of the present invention is to provide a data writing method and device for a cloud storage system, and a cloud storage system, so as to solve the problems in the existing cloud storage system that are prone to jamming, more delays in writing operations, and lower storage performance.

In order to solve the above technical problems, the present invention provides a data writing method of a cloud storage system, wherein the storage nodes of the cloud storage system include at least one high-speed storage device and one large-capacity storage device, and the method includes:

receiving the write request sent by the client node, and writing the data to be written after the client node has performed erasure code processing to the high-speed storage device;

After writing, send a completion response to the client node;

receiving the persistence request sent by the client node, and persistently writing the data to be written to the mass storage device.

Optionally, after the persistent writing of the data to be written to the mass storage device, the method further includes:

Delete corresponding versioned data stored on the high-speed storage device.

Optionally, also include:

Each storage node detects the running status of other storage nodes through the heartbeat mechanism;

When it is detected that the state of one or more storage nodes is abnormal, the version numbers of the data in the high-speed storage device are compared between the nodes in the normal state, and the data with the lowest version is used as the reference data, and the data in other nodes greater than the Delete the data of the version number of the reference data mentioned above.

Optionally, the high-speed storage device is a solid-state hard disk or a non-volatile memory; the large-capacity storage device is a mechanical hard disk or a disk array.

The present invention also provides a data writing device for a cloud storage system, the storage node of the cloud storage system includes at least one high-speed storage device and one large-capacity storage device, and the device includes:

The first write module is configured to receive a write request sent by a client node, and write data to be written after the client node has performed erasure code processing to the high-speed storage device;

A response module, configured to send a completion response to the client node after the writing is completed;

The second writing module is configured to receive the persistence request sent by the client node, and persistently write the data to be written to the mass storage device.

The present invention also provides a cloud storage system, comprising:

A storage cluster composed of client nodes and multiple storage nodes; each storage node in the storage cluster contains at least one high-speed storage device and one large-capacity storage device;

Wherein, the client node is used to receive a user's write request, perform erasure code processing on the data to be written, and send the data after erasure code processing to a designated storage node;

The storage node receives the write request sent by the client node, and writes the data to be written into the high-speed storage device, and sends a completion response to the client node after the writing is completed;

After receiving the completion response, the client node sends a persistence request to the storage node;

The storage node receives the persistence request, and persistently writes the data to be written to the mass storage device.

Optionally, the storage node is also used to detect the running status of other storage nodes through a heartbeat mechanism; when it is detected that the status of one or more storage nodes is abnormal, compare the high-speed storage For the version number of the data in the device, the data with the lowest version is used as the reference data, and the data in other nodes whose version number is greater than the reference data is deleted.

Optionally, the client node is specifically configured to divide the data to be written into segments to generate an original data block; calculate redundant data in the original data block, and expand and encode the redundant data , and stored in different locations.

Optionally, the storage node is further configured to delete corresponding versioned data stored on the high-speed storage device after persistently writing the data to be written into the large-capacity storage device.

Optionally, the high-speed storage device is a solid-state hard disk or a non-volatile memory; the large-capacity storage device is a mechanical hard disk or a disk array.

In the data writing method and device of the cloud storage system provided by the present invention, the storage nodes of the cloud storage system include at least one high-speed storage device and one large-capacity storage device, by receiving the write request sent by the client node, and sending the client After the node performs erasure code processing, the data to be written is written to the high-speed storage device; after writing, it sends a completion response to the client node; receives the persistence request sent by the client node, and persists the data to be written written to a mass storage device. The present invention completes data writing by writing high-speed storage devices and writing large-capacity storage devices in two steps, which can not only ensure the data reliability and consistency of the cloud storage system, but also reduce the delay of writing operations while providing large storage capacity . And completing the erasure code processing at the client node can reduce the pressure on the server and reduce the amount of data in network communication. In addition, the present invention also provides a cloud storage system with the above technical advantages.

Description of drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions of the prior art, the following will briefly introduce the accompanying drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only For some embodiments of the present invention, those skilled in the art can also obtain other drawings based on these drawings without creative work.

Fig. 1 is a flow chart of a specific embodiment of the data writing method of the cloud storage system provided by the present invention;

Fig. 2 is a flow chart of another specific embodiment of the data writing method of the cloud storage system provided by the present invention;

Fig. 3 is a schematic diagram of erasure code principle;

FIG. 4 is a structural block diagram of a data writing device of a cloud storage system provided by an embodiment of the present invention;

FIG. 5 is a schematic diagram of a cloud storage system provided by an embodiment of the present invention;

FIG. 6 is a schematic diagram of a processing procedure of a write operation of a cloud storage system provided by an embodiment of the present invention.

detailed description

In order to enable those skilled in the art to better understand the solution of the present invention, the present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments. Apparently, the described embodiments are only some of the embodiments of the present invention, but not all of them. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.

A flow chart of a specific implementation of the data writing method of the cloud storage system provided by the present invention is shown in Figure 1, the storage node of the cloud storage system includes at least one high-speed storage device and one large-capacity storage device, The methods include:

Step S101: receiving the write request sent by the client node, and writing the data to be written after the erasure code processing of the client node to the high-speed storage device;

It should be noted that the cloud storage system provided by the present invention may specifically include: one or more client nodes, configured to receive and process user write requests.

A storage cluster composed of multiple storage nodes is used to store data. Each storage node includes at least one high-speed storage device and one large-capacity storage device.

The client node performs erasure code encoding and grouping on the data submitted by the user; for each data packet, the client node encapsulates it into a write request and sends it to the designated storage node. The storage node receives the write request sent by the client, and writes the data to be written to the high-speed storage device in a readable, non-overwritten, and versioned manner.

Step S102: After writing, send a completion response to the client node;

Step S103: receiving the persistence request sent by the client node, and persistently writing the data to be written to the mass storage device.

After receiving the completion responses from all storage nodes, the client node notifies the user that the write operation has been completed, and sends a persistence request to each storage node. Each storage node receives the persistence request of the client node, and writes the data persistence into the mass storage device.

In the embodiment of the present invention, after persistently writing the data to be written to the large-capacity storage device, it may further include: deleting corresponding versioned data on the high-speed storage device.

In the data writing method of the cloud storage system provided by the present invention, the storage node of the cloud storage system includes at least one high-speed storage device and one large-capacity storage device, and receives the write request sent by the client node, and the client node performs The data to be written after erasure code processing is written to the high-speed storage device; after the writing is completed, a completion response is sent to the client node; the persistence request sent by the client node is received, and the data to be written is persistently written into the mass storage device. The present invention completes data writing by writing high-speed storage devices and writing large-capacity storage devices in two steps, which can not only ensure the data reliability and consistency of the cloud storage system, but also reduce the delay of writing operations while providing large storage capacity . And completing the erasure code processing at the client node can reduce the pressure on the server and reduce the amount of data in network communication.

The flow chart of another specific embodiment of the data writing method of the cloud storage system provided by the present invention is shown in Figure 2, and the method includes:

Step S201: the client node performs erasure code encoding and grouping on the data submitted by the user; for each data group, the client node encapsulates it into a write request and sends it to the designated storage node.

Step S202: Each storage node receives the write request from the client node, and writes the data in it to the high-speed storage device in a readable, non-overwritten, and versioned manner; after the writing is completed, the storage node sends Complete the response.

Step S203: After receiving the completion responses from all storage nodes, the client node notifies the user that the write operation has been completed, and sends a persistence request to each storage node.

Step S204: Each storage node receives the persistence request from the client node, writes the persistent data into the mass storage device, and deletes the corresponding versioned data on the high-speed storage device.

Erasure coding (EC) is a data protection method that divides data into fragments, expands and encodes redundant data blocks, and stores them in different locations, such as disks, storage nodes, or other geographic locations . A common use case for erasure coding today is object-based cloud storage.

Referring to Figure 3, the basic idea of erasure codes is to calculate k blocks of original data elements through certain encoding calculations to obtain m blocks of check elements. The system stores the k+m blocks of data on k+m storage nodes respectively. When any storage node less than or equal to m fails, the system can restore the original k blocks of data through the data on the remaining nodes.

On the basis of any of the above embodiments, the data writing method of the cloud storage system provided by the present invention may further include: each storage node detects the running status of other storage nodes through a heartbeat mechanism; when one or more When the state of the storage node is abnormal, the version numbers of the data in the high-speed storage device are compared between the nodes in the normal state, and the data with the lowest version is used as the reference data, and the version numbers of the other nodes that are greater than the version number of the reference data are compared. The data is deleted.

Abnormal phenomena can be specifically node failure or node restart, etc. Through the above detection and recovery operations, the data between all storage nodes can be at the same version, thus ensuring the consistency of data in the system.

As a specific implementation manner, the high-speed storage device may specifically be a solid-state hard disk or a non-volatile memory; the large-capacity storage device may specifically be a mechanical hard disk or a disk array. High-speed storage devices have the advantages of high performance and low latency, so they are used as caches; while large-capacity storage devices can achieve large capacity and high cost performance, so they are used for persistent storage of data.

The data writing device of the cloud storage system provided by the embodiment of the present invention is introduced below. The data writing device of the cloud storage system described below and the data writing method of the cloud storage system described above can be referred to in correspondence.

Fig. 4 is a structural block diagram of the data writing device of the cloud storage system provided by the embodiment of the present invention, the storage node of the cloud storage system includes at least one high-speed storage device and one large-capacity storage device, referring to the data writing device of the cloud storage system in Fig. Input devices can include:

The first write module 100 is configured to receive a write request sent by a client node, and write data to be written after the client node has performed erasure code processing to the high-speed storage device;

The response module 200 is configured to send a completion response to the client node after the writing is completed;

The second writing module 300 is configured to receive the persistence request sent by the client node, and persistently write the data to be written to the mass storage device.

The data writing device of the cloud storage system in this embodiment is used to realize the aforementioned data writing method of the cloud storage system, so the specific implementation of the data writing device of the cloud storage system can be seen in the data writing of the cloud storage system mentioned above. The embodiment part of the input method, for example, the first writing module 100, the response module 200, and the second writing module 300 are respectively used to implement steps S101, S102, and S103 in the data writing method of the above-mentioned cloud storage system, so, For specific implementation manners, reference may be made to the descriptions of the corresponding partial embodiments, and details are not repeated here.

The client may specifically include an erasure code module and a message processing module.

Wherein, the erasure code module is used to divide the data to be written into K blocks of original data according to the erasure code algorithm, and calculate and obtain M blocks of redundant data based on the K blocks of original data.

The message processing module is used for communication between the client and the storage node. The client first sends K blocks of original data and M blocks of redundant data to (K+M) storage nodes respectively, and then waits for the completion response of each storage node; after receiving the completion response of all (K+M) storage nodes After that, the client confirms that the write operation is completed, and then sends a persistence request to each storage node.

After receiving the write request from the client, the first writing module of the storage node writes the data into the high-speed storage device, and the response module sends a completion response to the client after the writing is completed; the second writing module receives the persistent data from the client Data is written to mass storage devices and deleted from high-speed storage devices after processing requests.

Wherein, the first writing module is specifically used to write data to the high-speed storage device in a readable, non-overwriting, and versioned form; after the persistent writing is completed, the cache module deletes the corresponding data.

The second writing module is specifically configured to write data into the large-capacity device, and notify the cache module to delete corresponding data.

Further, the data writing device of the cloud storage system provided by the present invention may further include: a status detection and recovery module: each storage node detects the running status of other storage nodes through a heartbeat mechanism. When the state of one or more nodes changes (node failure, node restart, etc.), the nodes in the normal state compare the version numbers of the data in the high-speed storage device with each other, and take the data of the node with the lowest version number as the benchmark data. Other nodes restore the data to a state consistent with the baseline by deleting data greater than this version number.

In addition, the present invention also provides a cloud storage system, as shown in Figure 5, the system includes:

A storage cluster composed of client nodes and multiple storage nodes; each storage node in the storage cluster contains at least one high-speed storage device and one large-capacity storage device;

Wherein, the client node is used to receive a user's write request, perform erasure code processing on the data to be written, and send the data after erasure code processing to a designated storage node;

The storage node receives the write request sent by the client node, and writes the data to be written into the high-speed storage device, and sends a completion response to the client node after the writing is completed;

After receiving the completion response, the client node sends a persistence request to the storage node;

The storage node receives the persistence request, and persistently writes the data to be written to the mass storage device.

Wherein, the storage node is also used to detect the running status of other storage nodes through the heartbeat mechanism; when it is detected that the status of one or more storage nodes is abnormal, each node in the normal status compares the status of the high-speed storage device The version number of the data, the data with the lowest version is used as the reference data, and the data in other nodes whose version number is greater than the reference data is deleted.

As a specific implementation manner, the client node is specifically configured to divide the data to be written into segments to generate original data blocks; calculate redundant data in the original data blocks, and perform redundant data processing on the redundant data Expanded, encoded, and stored in a different location.

The storage node is further configured to delete corresponding versioned data stored on the high-speed storage device after persistently writing the data to be written into the large-capacity storage device.

As a specific implementation, the high-speed storage device is a solid-state hard disk or a non-volatile memory; the large-capacity storage device is a mechanical hard disk or a disk array.

Referring to the schematic diagram of the processing process of the write operation in FIG. 6 , the data writing process will be further described in detail below. Specifically, the process includes:

The client node performs erasure code encoding and grouping on the data submitted by the user; for each data packet, the client node encapsulates it into a write request and sends it to the designated storage node.

Each storage node receives the write request from the client node, and writes the data in it to the high-speed storage device in a readable, non-overwritten, and versioned manner; after the writing is completed, the storage node sends a completion response to the client node.

After receiving the completion responses from all storage nodes, the client node notifies the user that the write operation has been completed, and sends a persistence request to each storage node.

Each storage node receives the persistence request of the client node, writes the persistent data into the mass storage device, and deletes the corresponding versioned data on the high-speed storage device.

The present invention completes data writing by writing high-speed storage devices and writing large-capacity storage devices in two steps, which can not only ensure the data reliability and consistency of the cloud storage system, but also reduce the delay of writing operations while providing large storage capacity . And completing the erasure code processing on the client node can reduce the pressure on the server and reduce the amount of data in network communication.

Each embodiment in this specification is described in a progressive manner, each embodiment focuses on the difference from other embodiments, and the same or similar parts of each embodiment can be referred to each other. As for the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and for the related information, please refer to the description of the method part.

Professionals can further realize that the units and algorithm steps of the examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, computer software or a combination of the two. In order to clearly illustrate the possible For interchangeability, in the above description, the composition and steps of each example have been generally described according to their functions. Whether these functions are executed by hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art may use different methods to implement the described functions for each specific application, but such implementation should not be regarded as exceeding the scope of the present invention.

The steps of the methods or algorithms described in connection with the embodiments disclosed herein may be directly implemented by hardware, software modules executed by a processor, or a combination of both. Software modules can be placed in random access memory (RAM), internal memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any Any other known storage medium.

The data writing method and device of the cloud storage system and the cloud storage system provided by the present invention have been introduced in detail above. In this paper, specific examples are used to illustrate the principle and implementation of the present invention, and the descriptions of the above embodiments are only used to help understand the method and core idea of the present invention. It should be pointed out that for those skilled in the art, without departing from the principle of the present invention, some improvements and modifications can be made to the present invention, and these improvements and modifications also fall within the protection scope of the claims of the present invention.

Claims (10)

1. A data writing method of a cloud storage system, wherein the storage node of the cloud storage system includes at least one high-speed storage device and a large-capacity storage device, and the method comprises: receiving the write request sent by the client node, and writing the data to be written after the client node has performed erasure code processing to the high-speed storage device; After writing, send a completion response to the client node; receiving the persistence request sent by the client node, and persistently writing the data to be written to the mass storage device. 2. The data writing method of the cloud storage system according to claim 1, further comprising: after said persistently writing said data to be written into said mass storage device: Delete corresponding versioned data stored on the high-speed storage device. 3. The data writing method of the cloud storage system according to claim 1 or 2, further comprising: Each storage node detects the running status of other storage nodes through the heartbeat mechanism; When it is detected that the state of one or more storage nodes is abnormal, the version numbers of the data in the high-speed storage device are compared between the nodes in the normal state, and the data with the lowest version is used as the reference data, and the data in other nodes greater than the Delete the data of the version number of the reference data mentioned above. 4. The data writing method of the cloud storage system according to claim 3, wherein the high-speed storage device is a solid-state hard disk or a non-volatile memory; and the large-capacity storage device is a mechanical hard disk or a disk array. 5. A data writing device for a cloud storage system, characterized in that the storage node of the cloud storage system includes at least one high-speed storage device and a large-capacity storage device, and the device includes: The first write module is configured to receive a write request sent by a client node, and write data to be written after the client node has performed erasure code processing to the high-speed storage device; A response module, configured to send a completion response to the client node after writing; The second writing module is configured to receive the persistence request sent by the client node, and persistently write the data to be written to the mass storage device. 6. A cloud storage system, characterized in that, comprising: A storage cluster composed of client nodes and multiple storage nodes; each storage node in the storage cluster contains at least one high-speed storage device and one large-capacity storage device; Wherein, the client node is used to receive a user's write request, perform erasure code processing on the data to be written, and send the data after erasure code processing to a designated storage node; The storage node receives the write request sent by the client node, and writes the data to be written into the high-speed storage device, and sends a completion response to the client node after the writing is completed; After receiving the completion response, the client node sends a persistence request to the storage node; The storage node receives the persistence request, and persistently writes the data to be written to the mass storage device. 7. The cloud storage system according to claim 6, wherein the storage node is also used to detect the operating status of other storage nodes through a heartbeat mechanism; when it is detected that the status of one or more storage nodes is abnormal, The nodes in the normal state compare the version numbers of the data in the high-speed storage device, use the data with the lowest version as the reference data, and delete the data in other nodes whose version numbers are greater than the reference data. 8. The cloud storage system according to claim 6 or 7, wherein the client node is specifically configured to divide the data to be written into segments to generate original data blocks; The redundant data is expanded, coded, and stored in different locations. 9. The cloud storage system according to claim 8, wherein the storage node is further configured to delete the data stored in the large-capacity storage device after persistently writing the data to be written into the large-capacity storage device. Corresponding versioned data on the above-mentioned high-speed storage device. 10. The cloud storage system according to claim 9, wherein the high-speed storage device is a solid-state hard disk or a non-volatile memory; the large-capacity storage device is a mechanical hard disk or a disk array.