WO2015196692A1

WO2015196692A1 - Cloud computing system and processing method and apparatus for cloud computing system

Info

Publication number: WO2015196692A1
Application number: PCT/CN2014/090398
Authority: WO
Inventors: 莫嫣; 高洪; 韩银俊
Original assignee: 中兴通讯股份有限公司
Priority date: 2014-06-24
Filing date: 2014-11-05
Publication date: 2015-12-30
Also published as: CN105323271A; CN105323271B

Abstract

The present invention provides a cloud computing system and a processing method and apparatus for a cloud computing system. The processing method for a cloud computing system comprises: receiving an operation request of a client for a cloud computing system; acquiring a data identifier, on which an operation is to be performed, in the cloud computing system according to the operation request; searching for each disk, which stores data corresponding to the data identifier, in each node of the cloud computing system and the state of each disk according to a node disk state report of the cloud computing system, the node disk state report comprising states of disks in each node of the cloud computing system and data identifiers corresponding to data stored in the disks; and performing a corresponding operation according to the state of each disk, which stores the data corresponding to the data identifier, in each node of the cloud computing system. The present invention can improve the system tolerance of disk failure.

Description

Cloud computing system and processing method and device of cloud computing system

Technical field

The present invention relates to the field of cloud computing technologies, and in particular to a cloud computing system and a processing method and apparatus for a cloud computing system.

Background technique

Currently, Cloud Computing is Grid Computing, Distributed Computing, Parallel Computing, Utility Computing, Network Storage Technologies, and Virtualization. ), load balancing (Load Balance) and other traditional computer technology and network technology development and fusion products. It aims to integrate multiple relatively low-cost computing entities into a system with powerful computing capabilities through the network. Distributed caching is an area in the field of cloud computing. Its role is to provide distributed storage services for massive data and high-speed read and write access.

The distributed cache system is composed of a plurality of server nodes and clients interconnected; the server node is responsible for data storage, and the client can perform operations such as writing, reading, updating, and deleting data to the server. In general, data cannot be stored only on a single server node (hereinafter referred to as "node"), but a copy of the same data is stored on multiple nodes, which are backups of each other. The most common storage mode is the master-slave mode, in which one node acts as the master node and the other nodes act as slaves. The identity of the master node is obtained through election or other algorithms. To simplify the process, data updates generally occur on the primary node. The standby node obtains data from the primary node for synchronization, and the data access can obtain data from the primary node or obtain data from the standby node, depending on the consistency of the access. Strategy.

In the distributed cache system, according to the requirements of consistency and availability, the data storage mode is generally classified according to NRW, where N represents the number of copies of the data, R represents the number of copies of the data obtained in one data access request, and W represents once. The minimum number of participating nodes for the data update request (ie, how many nodes are updated).

When the distributed caching system implements the persistence function, the data distributed on the server is saved on the disk. In the real world, if the disk fails, the server cannot provide read and write services. Since the distributed cache system data has multiple replicas, the system can still provide read and write services through the replicas of other nodes as long as the other servers are in a normal state.

If a distributed cache system node has multiple disks attached, only one or a few of them are damaged for some reason, causing the server to fail to provide services. According to the foregoing, the entire cluster is still available because other servers are normally available. . Assume that during this time, another server has a similar situation, and that node is not able to provide services normally. It is likely that the number of copies cannot satisfy the NRW policy, and the distributed cache cluster is completely unable to provide services. Typically, when the more commonly used NRW is 3/2/2, the two nodes are dropped, only one node is normal, and the read and write operations cannot meet the minimum operation requirements on the two copies.

Summary of the invention

The technical problem to be solved by the present invention is to provide a cloud computing system and a processing method and apparatus for the cloud computing system, which can improve the tolerance of the system to disk failure.

To solve the above technical problem, an embodiment of the present invention provides a processing method of a cloud computing system, including:

Receiving an operation request of the client to the cloud computing system;

Obtaining, according to the operation request, a data identifier to be operated in the cloud computing system;

And searching, according to the node disk status report of the cloud computing system, each disk in the node of the cloud computing system, where the data corresponding to the data identifier is stored, and a status of each of the disks; the node disk status report includes: a state of a disk in each node of the cloud computing system, and a data identifier corresponding to data stored in the disk;

And performing corresponding operations according to states of each of the disks in which the data identifier corresponding data is stored in each node in the cloud computing system.

According to the state of each of the disks, the steps of performing corresponding operations include:

The operation request is an update request; when the number of the disks in the cloud computing system storing the data and in a normal state is greater than or equal to a minimum number of participating nodes of a predetermined data update request of the cloud computing system, Responding to the update request; otherwise, rejecting the update request; or

The operation request is a data access request; when the number of the disks in the cloud computing system storing the data and in a normal state is greater than or equal to the number of data copies acquired by a predetermined data access request of the cloud computing system Responding to the data access request; otherwise, rejecting the data access request.

Responding to the update request when the number of the disks in the cloud computing system storing the data and in a normal state is greater than or equal to the minimum number of participating nodes of the data update request scheduled by the cloud computing system The steps include:

When the operation request is an update request, and the state of the disk of the primary node storing the data is normal, the primary node of the cloud computing system performs data update to the disk of the data of the primary node; the cloud computing The slave node of the system acquires data to be synchronized from the master node, and the slave node performs data update on the disk where the data of the slave node is located;

When the operation request is an update request, and the state of the disk of the primary node storing the data is a failure, the first slave node of the cloud computing system performs data to the disk where the data of the first slave node is located. Updating; the second slave node of the cloud computing system acquires data to be synchronized from the first slave node; and the second node performs data update to the disk where the data of the second slave node is located; The state of the disk storing the data by a slave node and the second slave node is normal.

Responding to the data access when the number of the disks in the cloud computing system storing the data and in a normal state is greater than or equal to the number of data copies acquired by the cloud computing system for a predetermined data access request The requested steps include:

When the operation request is a data access request, and the state of the disk of the primary node storing the data is normal, obtaining the first copy of the data from the disk where the data of the primary node of the cloud computing system is located Obtaining a second copy of the data from a disk of the at least one slave node of the cloud computing system; selecting, from the first copy and the second copy, a copy of the latest version; Sending a copy of the latest version to the client; the state of the disk storing the data of the second slave node is normal;

When the operation request is a data access request, and the state of the disk of the primary node storing the data is a failure, obtaining the data from the disk where the data of the at least one slave node of the cloud computing system is located Three copies; from at least one of the third copies, selecting a copy of the latest version and transmitting the copy of the latest version to the client; the state of the disk storing the data of the second slave node is normal.

Before the step of receiving an operation request of the client, the method further includes:

The node disk status report of the cloud computing system is obtained from the node.

The embodiment of the invention further provides a processing device of a cloud computing system, including:

a first receiving unit, receiving an operation request of the client for the cloud computing system;

Obtaining, according to the operation request, acquiring a data identifier to be operated in the cloud computing system;

a search unit, configured to search, according to a node disk status report of the cloud computing system, each disk in the node of the cloud computing system, where the data corresponding to the data identifier is stored, and a status of each of the disks; the node disk status report includes : a state of a disk in each node of the cloud computing system, and a data identifier corresponding to data stored in the disk;

The operation unit performs a corresponding operation according to the state of each of the disks in which the data identifier corresponding data is stored in each node in the cloud computing system.

The operating unit includes:

a first response subunit, the operation request being an update request; when the number of the disks in the cloud computing system storing the data and in a normal state is greater than or equal to a predetermined data update request of the cloud computing system Responding to the update request when at least the number of participating nodes;

a first rejecting subunit, rejecting the update when the number of the disks in the cloud computing system storing the data and in a normal state is less than a minimum number of participating nodes of a predetermined data update request of the cloud computing system request;

a second response subunit, the operation request is a data access request; when the number of the disks in the cloud computing system storing the data and in a normal state is greater than or equal to a predetermined data access request of the cloud computing system Responding to the data access request when the number of data copies obtained is obtained;

a second rejecting subunit, rejecting the data when the number of the disks in the cloud computing system storing the data and in a normal state is less than the number of data copies acquired by the cloud computing system for a predetermined data access request Access request.

The device further includes:

The second receiving unit receives the node disk status report of the cloud computing system from the node.

An embodiment of the present invention further provides a cloud computing system, including: a client, a processing device, a node, and a disk corresponding to the node;

Receiving, by the processing device, an operation request of the client to the cloud computing system; acquiring, according to the operation request, a data identifier to be operated in the cloud computing system; searching according to a node disk status report of the cloud computing system Each of the nodes of the cloud computing system stores a respective disk of the data identifier corresponding data and a status of each of the disks; the node disk status report includes: the nodes in the cloud computing system The state of the disk, the data identifier corresponding to the data stored in the disk; and the corresponding operation according to the state of each of the disks storing the data identifier corresponding data in each node in the cloud computing system.

The node sends a node disk status report to the processing device.

The beneficial effects of the above technical solutions provided by the embodiments of the present invention are as follows:

For distributed cache systems, in the case of disk corruption, you can make full use of available resources, consolidate replica resources that meet the requirements of consistency and availability, maximize system availability, and improve system tolerance to failure.

DRAWINGS

1 is a schematic flowchart of a processing method of a cloud computing system according to an embodiment of the present invention;

2 is a schematic structural diagram of a processing apparatus of a cloud computing system according to an embodiment of the present invention;

3 is a schematic structural diagram of a cloud computing system according to an embodiment of the present invention;

4 and FIG. 5 are schematic structural diagrams of an application scenario of a cloud computing system according to an embodiment of the present invention.

detailed description

The technical problems, the technical solutions, and the advantages of the present invention will be more clearly described in the following description.

As shown in FIG. 1 , a processing method of a cloud computing system according to an embodiment of the present invention includes:

Step 11: Receive an operation request of the client for the cloud computing system; the operation request may be a data update request or a data access request.

Step 12: Acquire, according to the operation request, a data identifier to be operated in the cloud computing system; for example, the operation request is to update the copy 1 in FIG. 4, and the copy 1 is a data identifier.

Step 13: Search, according to the node disk status report of the cloud computing system, the status of each disk in which the data identifier corresponding data is stored in each node of the cloud computing system and the status of each of the disks; the node disk status report includes The state of the disk in each node of the cloud computing system, the data identifier corresponding to the data stored in the disk; the state of the disk is normal or faulty; in FIG. 4, the disk state report of node A is: (node A : Disk I, Replica 1, Failure; Disk II, Replica 2, Normal; Disk III, Replica 3, Normal).

Step 14: Perform corresponding operations according to states of each of the disks in which the data identifier corresponding data is stored in each node in the cloud computing system.

Before step 14, the method further includes:

Step 10: Obtain a node disk status report of the cloud computing system from a node. The node sends a report if it detects that the disk storing the data is corrupted or fails, or sends a report based on the request.

Wherein step 14 includes:

Specifically:

When the operation request is a data access request, and the state of the disk of the primary node storing the data is normal, obtaining the first copy of the data from the disk where the data of the primary node of the cloud computing system is located Obtaining, from at least one (or two or three, according to actual conditions) of the cloud computing system, a second copy of the data from a disk on which the data of the node is located; from the first copy And selecting, in the second copy, a copy of the latest version; and sending the copy of the latest version to the client; the state of the disk storing the data of the second slave node is normal;

For example, FIG. 5 is a distributed cache storage system composed of three nodes. The storage system has three copies of each data, and updates and accesses data in a manner of 322. The number of read request access copies specified by the cloud computing system is 2. When one disk is broken, it can still respond to the update or data access operation request. When two disks are broken, the operation request cannot be responded to.

In the present invention, when a node disk failure occurs, and even multiple nodes simultaneously fail the disk, as long as the number of available disks on the cluster can satisfy the NRW policy, the system can ensure consistency and availability, and may even affect all data. The service, and the fact that the system is completely unable to provide services, will provide services as much as possible.

Of course, in the case that part of the disk damage continues to provide services, the problem of data recovery after disk recovery is brought about by the distributed cache data recovery function, that is, the copy data is obtained from other nodes to be repaired.

As shown in FIG. 2, a processing device of a cloud computing system according to an embodiment of the present invention includes:

The first receiving unit 21 receives an operation request of the client for the cloud computing system;

The obtaining unit 22 is configured to acquire, according to the operation request, a data identifier to be operated in the cloud computing system;

The searching unit 23 is configured to search, according to the node disk status report of the cloud computing system, each disk in the node of the cloud computing system, where the data corresponding to the data identifier is stored, and a status of each of the disks; the node disk status report The method includes: a state of a disk in each node of the cloud computing system, and a data identifier corresponding to data stored in the disk;

The operation unit 24 performs a corresponding operation according to the state of each of the disks in which the data identifier corresponding data is stored in each node in the cloud computing system.

The operating unit 24 includes:

The device further includes:

The second receiving unit 25 receives the node disk status report of the cloud computing system from the node.

As shown in FIG. 3, a cloud computing system according to the present invention includes: a client 31, a processing device 32, a node 33, and a disk 34 corresponding to the node 33;

The processing device 32 receives an operation request of the client 31 for the cloud computing system; and according to the operation request, acquires a data identifier to be operated in the cloud computing system; and reports a node disk status according to the cloud computing system. And searching, in each of the nodes 33 of the cloud computing system, a disk that stores the data identifier corresponding data and a status of each of the disks 34; the node disk status report includes: each node 33 of the cloud computing system a state of the disk, a data identifier corresponding to the data stored in the disk; performing a corresponding operation according to a state of each of the disks 34 storing the data identifier corresponding data in each node in the cloud computing system .

The node 33 sends a node disk status report to the processing device 32.

Two application scenarios of the embodiment of the present invention are described below.

The first application scenario describes a method for implementing availability when a disk is damaged in a multi-disk path in a cloud computing distributed cache system.

Pre-step: The client establishes a connection with multiple server nodes in the distributed cache system. The server nodes establish connections and run normally. Each server has several disks for data persistence. Different data points. Slices are persisted on different disks. The number of copies of the data is N, and the number of copies of the read request access is R, write The minimum number of successful requests for copying is W, and the system's single maximum fault tolerance is O (indicating that the request on O nodes is allowed to fail, such as single point failure, O=1, O<W), consistency requirement W+ R>N.

Step A: Under normal circumstances, all the disks on each node work normally, and the data has N copies in the system. When the client initiates a data update request, the master performs data update processing on the disk where the data is located, the slave synchronizes the data from the master, and updates the data to the disk where the data is on the slave. When the data update is successfully completed on the W nodes, the return is performed. Client data update success message;

When the client initiates a data access request, the master/Slave processes the request, and after obtaining the accessed data copy from the disk where the R node data is located, the latest copy is selected from the R data copies and returned to the client.

Step B: When the node A starts, it finds that a certain disk failure is inaccessible, but the other disks are still normal. Or, during the running of the node A, it is found that the disk fails to access multiple times and it is determined that the disk is faulty. Node A does not switch to a node failure, but continues to provide read and write services, while recording the identity of the failed disk and the corresponding data copy on that disk.

Step C: When the client initiates a data update request, and the data is exactly distributed on the fault disk of node A described in step B, when updating data to the node, node A directly returns failure; when the data is updated in W After the node is successfully completed (the node A is not included in the W nodes), the data update success message is returned to the client;

When the client initiates a data access request, node A directly returns a failure, and the master/Slave processes the request, and obtains the accessed data copy from the disk where the data of the R nodes (the R nodes do not include node A) is from the R. The latest copy is selected from the data copies and returned to the client.

Step D: When the client initiates a data update and an access request, and the data is not distributed on the fault disk of node A described in step B, the processing manner is the same as step A.

Step E: When Node B is running, multiple access failures to a disk determine that the disk is faulty. Node B does not switch to a node failure, but continues to provide read and write services, while recording the identity of the failed disk and the corresponding data copy on the disk.

Assume that the failed disk of Node B and the copy saved on the failed disk of Node A do not coincide. Continue to the next step.

Step F: When the client initiates a data update and access request, and the data is exactly distributed on the fault disk of the node B described in step E, based on the above assumption, not on the fault disk of the node A in step B, then When the node updates the data, the Node B directly returns the failure; when the data update is successfully completed on the W nodes (the W nodes do not include the Node B), the data update success message is returned to the client;

When the client initiates a data access request, the Node B directly returns a failure, and the Master/Slave processes the request, and obtains the accessed data copy from the disk where the R nodes (the R nodes do not include the Node B) data, from the R The latest copy is selected from the data copy and returned to the client.

Step G: When the client initiates a data update request, and the data is exactly distributed on the fault disk of the node A described in step B, based on the above assumption, if it is not on the fault disk of the node B described in step E, then When the node updates and accesses the data, the process is the same as step C, and the result is that it can be updated and accessed normally.

In the embodiment of the present invention, when a node disk failure occurs, and even multiple nodes simultaneously fail the disk, as long as the number of available copies of the remaining disk on the cluster can satisfy the NRW policy, the system can ensure consistency and availability, and may even have no effect. The service of all data, and the fact that the system is completely unable to provide services, will provide services as much as possible.

The embodiment of the invention provides an implementation method for improving the availability of a distributed cache system in the case of multiple disk corruption. The consistency of the system is enhanced, and the availability of the system is enhanced, thereby optimizing the application experience.

The second application scenario will be described below with reference to FIGS. 4 and 5.

Specifically: for the 322 mode primary and secondary storage systems, a detailed description of the availability of a single node disk damage and multi-node disk corruption, availability implementation.

The server node and the client form a distributed cache system. For a specific data, there is a master node responsible for processing client updates and access requests. There are several standby nodes for synchronizing the master data and receiving client data. Access request (slave does not process data update request).

Environment: A distributed cache storage system consisting of three nodes. The storage system has three copies of each data and updates and accesses data in 322 mode.

Embodiments of the present invention include the following steps:

Step 1, in the initial normal phase, the system receives the client request, assuming that the data is located on the disk I of the node A, copy 1 (corresponding to the above data identification), the copy of the disk I of the node B, and the copy of the disk III of the node C. 1 on. For the sake of simplicity of description, assume that replica 1 on node B is master, on the other two nodes. The copy is a slave. The copy 2 on node A is the master, and the copy on the other two nodes is the slave. The copy 3 on node A is the master, and the copy on the other two nodes is the slave.

Step 2: When the client initiates a data update request, the Node B master updates the data to the copy 1 on the disk 1. The slave synchronizes the data from the master and updates the data to the disk where the data is on the slave. When the data is updated at W=2 After successful completion on each node, the data update success message is returned to the client. Since all the disks are normal, all the actual copies are successfully updated; when the client initiates a data access request, all three nodes process the request, and obtain the accessed data copy from the disk where the R=2 node data is located, and then return to the client. All actual node copies are read successfully.

Step 3. As shown in FIG. 4, it is assumed that the disk I on the node A is damaged, resulting in the copy 1 being unavailable. When the data of the update request initiated by the client is located on the replica 1 of the node A, the data of the copy 1 of the disk I is updated by the master of the node B, the slave of the node C synchronizes the data from the master, and the data is copied to the copy of the disk C of the node C. Data update is performed. At this time, after the data update is successfully completed on W=2 nodes, the data update success message is returned to the client;

When the data of the client initiating the data access request is located on the replica 1 of the node A, the node A directly returns the failure, and after obtaining the data from the replica 1 of the node B and the node C, (satisfying R=2) is returned to the client.

Step 4: In the case of step 3, when the update and access request initiated by the client is located on the copy 2 or the copy 3 of the node A, since the copies of the three nodes are available, the processing flow is the same as the step 2.

Step 5, as shown in FIG. 5, when the disk II on the node B is damaged, the copy 3 of the node B is unavailable. When the data of the update and access request initiated by the client is located on the replica 1 of the node A, the replicas on the node B and the node C are available, and the NRW policy is satisfied, and the processing flow is the same as step 3.

Step 6. In the case of step 5, when the update and access request initiated by the client is located on the copy 2 of the node A, since the copy 2 of the three nodes is available, the processing flow is the same as the step 2.

Step 7. In the case of step 5, when the data of the update request initiated by the client is located on the copy 3 of the node A, the copy 3 of the node B is damaged, and the copy 3 of the node C is available. The data is updated by the A-Node master to the replica 3 on the disk III, the slave of the node C synchronizes the data from the master, and updates the data on the replica 3 of the disk C on the node C. At this time, the data update succeeds on W=2 nodes. After completion, return the client data update success message;

When the data of the client initiating the data access request is located on the replica 3 of the node A, the node B directly returns the failure, and after obtaining the data from the replica 3 of the node A and the node C, the data is returned to the client (satisfying R=2).

As can be seen from the above, even if both node A and node B have disk corruption, the distributed cache cluster can provide read and write services for all data as long as the damaged disk copy is not duplicated.

In the above application scenario, if there are two faulty nodes, each node is actually partially damaged. In a more optimistic situation, if the damaged disk does not store a copy of the same data, the actual available disk of the entire system. At the top, it still holds at least two copies of all the data, and is fully qualified to provide all services normally. Even if a copy of the same data is stored on the damaged disk, the available data on other disks can still meet the consistency and availability, and can provide read and write services, which cannot be provided only for the data that is damaged at the same time. Read and write access.

In summary, the beneficial effects of the present invention are as follows:

The present invention is directed to a distributed cache system. In the case of disk corruption, the available resources can be fully utilized, and the copy resources conforming to the requirements of consistency and availability can be integrated to improve the availability of the system and improve the tolerance of the system to faults. That is to say, in the distributed cache system of the cloud computing field, a disk and data management mechanism is provided, and even in the case of a failure of a node part disk, the data on the available disk can be utilized as much as possible, and the ability to provide the service is maintained. Enables the server to provide consistent and available storage services with fewer disks or data resources.

The above is a preferred embodiment of the present invention, and it should be noted that those skilled in the art can also make several improvements and retouchings without departing from the principles of the present invention. It should be considered as the scope of protection of the present invention.

Industrial applicability

The above technical solution adopted by the embodiment of the present invention is directed to a distributed cache system, in the case of disk corruption, the available resources can be fully utilized, and the copy resources conforming to the consistency and availability requirements are integrated to improve the system availability as much as possible. Improve the system's tolerance to failure.

Claims

A method for processing a cloud computing system, comprising:

Receiving an operation request of the client to the cloud computing system;

Obtaining, according to the operation request, a data identifier to be operated in the cloud computing system;

And searching, according to the node disk status report of the cloud computing system, each disk in the node of the cloud computing system, where the data corresponding to the data identifier is stored, and a status of each of the disks; the node disk status report includes: a state of a disk in each node of the cloud computing system, and a data identifier corresponding to data stored in the disk;

And performing corresponding operations according to states of each of the disks in which the data identifier corresponding data is stored in each node in the cloud computing system.
The method of claim 1, wherein the step of performing a corresponding operation according to the state of each of the disks comprises:

And when the operation request is an update request, and the number of the disks in the cloud computing system storing the data and in a normal state is greater than or equal to a minimum number of participating nodes of a data update request scheduled by the cloud computing system Responding to the update request; otherwise, rejecting the update request; or

The operation request is a data access request, and the number of the disks in the cloud computing system storing the data and in a normal state is greater than or equal to the number of data copies acquired by the cloud computing system for a predetermined data access request. Responding to the data access request; otherwise, rejecting the data access request.
The method of claim 2, wherein said number of said disks in said cloud computing system storing said data and in a normal state is greater than or equal to a minimum of a predetermined data update request of said cloud computing system When participating in the number of nodes, the steps of responding to the update request include:

When the operation request is an update request, and the state of the disk of the primary node storing the data is normal, the primary node of the cloud computing system performs data update to the disk of the data of the primary node; the cloud computing The slave node of the system acquires data to be synchronized from the master node, and the slave node performs data update on the disk where the data of the slave node is located;

When the operation request is an update request, and the state of the disk of the primary node storing the data is a fault, the data of the first slave node of the cloud computing system to the data of the first slave node is located Performing data update; the second slave node of the cloud computing system acquires data to be synchronized from the first slave node; and the second node performs data update to the disk where the data of the second slave node is located; The state of the disk storing the data of the first slave node and the second slave node is normal.
The method according to claim 2, wherein said number of said disks in said cloud computing system storing said data and in a normal state is greater than or equal to a predetermined data access request acquisition by said cloud computing system The number of copies of the data, the steps of responding to the data access request include:

When the operation request is a data access request, and the state of the disk of the primary node storing the data is normal, obtaining the first copy of the data from the disk where the data of the primary node of the cloud computing system is located Obtaining a second copy of the data from a disk of the at least one slave node of the cloud computing system; selecting, from the first copy and the second copy, a copy of the latest version; Sending a copy of the latest version to the client; the state of the disk storing the data of the second slave node is normal;

When the operation request is a data access request, and the state of the disk of the primary node storing the data is a failure, obtaining the data from the disk where the data of the at least one slave node of the cloud computing system is located Three copies; from at least one of the third copies, selecting a copy of the latest version and transmitting the copy of the latest version to the client; the state of the disk storing the data of the second slave node is normal.
The method of claim 1, wherein before the step of receiving an operation request from a client, the method further comprises:

The node disk status report of the cloud computing system is obtained from the node.
A processing device for a cloud computing system, comprising:

a first receiving unit, receiving an operation request of the client for the cloud computing system;

Obtaining, according to the operation request, acquiring a data identifier to be operated in the cloud computing system;

a search unit, configured to search, according to a node disk status report of the cloud computing system, each disk in the node of the cloud computing system, where the data corresponding to the data identifier is stored, and a status of each of the disks; the node disk status report includes : a state of a disk in each node of the cloud computing system, and a data identifier corresponding to data stored in the disk;

The operation unit performs a corresponding operation according to the state of each of the disks in which the data identifier corresponding data is stored in each node in the cloud computing system.
The apparatus of claim 6, wherein the operating unit comprises:

a first response subunit, wherein the operation request is an update request, and the number of the disks in the cloud computing system storing the data and in a normal state is greater than or equal to a predetermined data update of the cloud computing system Responding to the update request when the minimum number of participating nodes is requested;

a first rejecting subunit, rejecting the update when the number of the disks in the cloud computing system storing the data and in a normal state is less than a minimum number of participating nodes of a predetermined data update request of the cloud computing system request;

a second response subunit, wherein the operation request is a data access request, and the number of the disks in the cloud computing system storing the data and in a normal state is greater than or equal to a predetermined data of the cloud computing system Responding to the data access request when accessing the number of data copies requested to be obtained;

a second rejecting subunit, rejecting the data when the number of the disks in the cloud computing system storing the data and in a normal state is less than the number of data copies acquired by the cloud computing system for a predetermined data access request Access request.
The apparatus of claim 6 further comprising:

The second receiving unit receives the node disk status report of the cloud computing system from the node.
A cloud computing system includes: a client, a processing device, a node, and a disk corresponding to the node;

Receiving, by the processing device, an operation request of the client to the cloud computing system; acquiring, according to the operation request, a data identifier to be operated in the cloud computing system; searching according to a node disk status report of the cloud computing system And storing, in each of the nodes of the cloud computing system, a disk of the data identifier corresponding data and a status of each of the disks; the node disk status report includes: a status of the disk in each node of the cloud computing system And the data identifier corresponding to the data stored in the disk; performing the corresponding operation according to the state of each of the disks in which the data identifier corresponding data is stored in each node in the cloud computing system.
The system of claim 9 wherein said node sends a node disk status report to said processing device.