CN111506453B

CN111506453B - Disk snapshot creation method, device, system and storage medium

Info

Publication number: CN111506453B
Application number: CN201910100438.XA
Authority: CN
Inventors: 廖武钧
Original assignee: Alibaba Group Holding Ltd
Current assignee: Alibaba Cloud Computing Ltd
Priority date: 2019-01-31
Filing date: 2019-01-31
Publication date: 2023-06-16
Anticipated expiration: 2039-01-31
Also published as: CN111506453A

Abstract

The invention discloses a disk snapshot creation method, a device, a system and a storage medium. The method comprises the following steps: sending a snapshot creation command, wherein the snapshot creation command is used for indicating time points of creating snapshots of a plurality of disks; receiving a plurality of blocking time points for blocking disk read-write requests in the process of creating snapshots by a plurality of disks aiming at snapshot creation commands; when the maximum value of the time difference among the plurality of blocking time points is smaller than a set first time difference threshold value, determining that the snapshots created by the plurality of magnetic disks meet the requirement of synchronization precision, and obtaining a plurality of snapshots created by the plurality of magnetic disks synchronously. According to the method provided by the embodiment of the invention, the disk snapshot creation process can be optimized.

Description

Disk snapshot creation method, device, system and storage medium

Technical Field

The present invention relates to the field of computer technologies, and in particular, to a method, an apparatus, a system, and a storage medium for creating a disk snapshot.

Background

In general, disk snapshots can be used for data backup and disaster recovery. In actual business, it is often necessary that a certain number of disks be able to create snapshots at the same point in time. For example, for a disk group formed by multiple disks, there is a data association between the disks, if a snapshot of a group of disks cannot be created at the same point in time, the data recovered by the snapshot of the group of disks may not necessarily meet the requirements of the disk association, and the snapshot of the group is considered to be unavailable in service.

Aiming at the situation, the solution can synchronize the read-write requests of the plurality of disks, so that the data of the plurality of disks are kept in a synchronous state, the snapshots created by the plurality of disks based on the disk data synchronous state can be created at the same moment, and the requirement of synchronously creating the snapshots is met. However, synchronizing read-write requests of multiple disks requires additional machine resources and maintenance costs, and the disk snapshot creation process is complex, which may result in reduced disk performance.

Disclosure of Invention

The embodiment of the invention provides a method, a device, a system and a storage medium for creating a disk snapshot, which can optimize the disk snapshot creation process.

In a first aspect, an embodiment of the present invention provides a method for creating a disk snapshot, including:

sending a snapshot creation command, wherein the snapshot creation command is used for indicating time points of creating snapshots of a plurality of disks; receiving a plurality of blocking time points for blocking disk read-write requests in the process of creating snapshots by a plurality of disks aiming at snapshot creation commands; when the maximum value of the time difference among the plurality of blocking time points is smaller than a set first time difference threshold value, determining that the snapshots created by the plurality of magnetic disks meet the requirement of synchronization precision, and obtaining a plurality of snapshots created by the plurality of magnetic disks synchronously.

In a second aspect, an embodiment of the present invention provides a disk snapshot creation device, including:

the system comprises a creation command sending module, a storage module and a storage module, wherein the creation command sending module is used for sending a snapshot creation command, and the snapshot creation command is used for indicating time points of creating snapshots of a plurality of magnetic disks; the blocking time receiving module is used for receiving a plurality of blocking time points for blocking disk read-write requests in the process of creating snapshots by a plurality of disks aiming at the snapshot creation command; and the synchronization precision determining module is used for determining that the snapshots created by the plurality of magnetic disks meet the synchronization precision requirement when the maximum value of the time difference between the plurality of blocking time points is smaller than a set first time difference threshold value, and obtaining a plurality of snapshots created by the plurality of magnetic disks synchronously.

In a third aspect, an embodiment of the present invention provides a disk snapshot creation system, including: a memory and a processor; the memory is used for storing programs; the processor is configured to read executable program code stored in the memory to perform the disk snapshot creating method of the first aspect described above.

In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having stored therein instructions that, when executed on a computer, cause the computer to perform the disk snapshot creation method of the first aspect described above.

In a fifth aspect, an embodiment of the present invention provides a method for creating a disk snapshot, including:

receiving a snapshot creation command, wherein the snapshot creation command is used for indicating a point in time of creating a snapshot; when waiting until the time point of creating the snapshot, creating disk snapshot data, and transmitting the snapshot data to a snapshot storage system; aiming at a snapshot creation command, recording a blocking time point when a disk read-write request is blocked in the process of creating disk snapshot data; the blocking time points are transmitted so that the management and control node determines whether the created snapshot data meets the synchronization accuracy requirement based on the received plurality of blocking time points.

In a sixth aspect, an embodiment of the present invention provides a disk snapshot creation device, including:

the system comprises a creation command receiving module, a snapshot creating module and a snapshot creating module, wherein the creation command receiving module is used for receiving a snapshot creation command, and the snapshot creation command is used for indicating a time point of creating a snapshot; the snapshot data creation module is used for creating disk snapshot data and transmitting the snapshot data to the snapshot storage system when waiting until a snapshot creation time point; the blocking time recording module is used for recording a blocking time point when a disk read-write request is blocked in the process of creating disk snapshot data according to a snapshot creation command; and the blocking time sending module is used for sending the blocking time points so that the control node can determine whether the created snapshot data meets the requirement of synchronous precision based on the blocking time points.

In a seventh aspect, an embodiment of the present invention provides a disk snapshot creation system, including: a memory and a processor; the memory is used for storing programs; the processor is configured to read executable program code stored in the memory to perform the disk snapshot creating method of the fifth aspect described above.

In an eighth aspect, an embodiment of the present invention provides a computer-readable storage medium having stored therein instructions that, when executed on a computer, cause the computer to perform the disk snapshot creation method of the fifth aspect described above.

According to the disk snapshot creation method, the device, the system and the storage medium, whether the creation time of a plurality of snapshots meets the synchronous precision requirement on service can be analyzed and judged through the IO blocking time in the recorded disk snapshot creation process, the disk snapshot creation process can be optimized, the additionally consumed machine resources and maintenance cost are reduced, and the disk performance is improved to a certain extent.

Drawings

In order to more clearly illustrate the technical solution of the embodiments of the present invention, the drawings that are needed to be used in the embodiments of the present invention will be briefly described, and other drawings can be obtained according to these drawings without inventive effort for a person skilled in the art.

Fig. 1 is a schematic diagram showing an application scenario according to an embodiment of the present invention;

FIG. 2 is a schematic diagram illustrating a disk snapshot creation process according to an embodiment of the present invention;

FIG. 3 is a flow diagram illustrating a method of multiple disk synchronization snapshot creation of one embodiment;

FIG. 4 is a schematic diagram illustrating the architecture of a multiple disk synchronous creation snapshot system in accordance with an embodiment of the present invention;

FIG. 5 is a flowchart illustrating a disk snapshot creation method according to an exemplary embodiment of the present invention;

FIG. 6 is a flow chart illustrating a disk snapshot creation method according to one embodiment of the invention;

FIG. 7 is a diagram illustrating a disk snapshot creation method according to another embodiment of the present invention;

FIG. 8 is a schematic diagram showing the structure of a disk snapshot creating apparatus according to an embodiment of the present invention;

FIG. 9 is a schematic diagram showing the structure of a disk snapshot creating apparatus according to an embodiment of the present invention;

FIG. 10 is a block diagram illustrating an exemplary hardware architecture of a computing device in which disk snapshot creation methods and apparatus according to embodiments of the invention may be implemented.

Detailed Description

Features and exemplary embodiments of various aspects of the present invention will be described in detail below, and in order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be described in further detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are merely configured to illustrate the invention and are not configured to limit the invention. It will be apparent to one skilled in the art that the present invention may be practiced without some of these specific details. The following description of the embodiments is merely intended to provide a better understanding of the invention by showing examples of the invention.

It is noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising … …" does not exclude the presence of other like elements in a process, method, article or apparatus that comprises the element.

For a better understanding of the present invention, a disk snapshot creation method, apparatus, system, and storage medium according to embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be noted that these examples are not intended to limit the scope of the present disclosure.

Fig. 1 is a schematic diagram illustrating a distributed system architecture according to an embodiment of the present invention. In the embodiment of the invention, the disk snapshot creation method can be applied to a distributed computing system. The distributed computing system may provide distributed computing services, distributed storage services, and network connectivity services to users.

In the embodiment of the invention, the distributed system architecture can take a computing device with computing capability as a computing resource through a virtualization technology, wherein the computing device can be a virtual machine on a machine device with computing capability; and may utilize a storage device as a storage resource. The storage device may be a disk array, a hard disk, a virtual hard disk, or the like.

As shown in fig. 1, the computing devices in the distributed system architecture that provide computing resources may be cloud servers. The cloud server may be a computing node implemented based on cloud computing technology that may provide distributed computing services. The cloud server is operated and managed in a remote login mode, and the use mode of the cloud server is basically the same as that of a common remote physical server. The storage device may be a cloud disk. The cloud disk can be a disk instance built on the distributed storage system, and is used for performing read-write operation in the cloud server. The cloud server and the cloud disk can be connected through network equipment.

The Snapshot (snap shot) is a complete record of the storage content in the disk storage device, which is obtained by backing up the disk data of the disk storage device at a certain moment or a certain time point. In one embodiment, incremental backups may be made of data in a disk storage device.

In some embodiments, disk snapshots are used primarily for backup and disaster recovery. A user can create a plurality of snapshots for one cloud disk at different time points, and a snapshot chain formed by the plurality of snapshots is obtained. If the disk data needs to be restored, the disk data rollback can be performed by utilizing the snapshot data, so that the data on the disk is restored to the data content of the snapshot at any time point on the snapshot chain.

In one embodiment, disk snapshot data may be created periodically, i.e., a time interval between two adjacent times of creation of disk snapshot data may be set, and the time interval is taken as a time period for periodically creating snapshots. It should be appreciated that the time period may be, for example, one day, two days or one week, and in an actual application scenario, the user may be determined according to an actual requirement, which is not specifically limited herein.

In one embodiment, since some of the disk data may not change during the creation of disk snapshots in the adjacent two times. That is, there is typically only a small difference in data content between snapshots in a snapshot chain. To save space, the snapshot system may deduplicate the created disk snapshots.

FIG. 2 is a schematic diagram illustrating an incremental backup of disk data by a disk snapshot creation process according to an embodiment of the present invention. As shown in FIG. 2, when creating a disk snapshot, the disk may be divided into a plurality of address intervals according to address offsets, and disk data in each address interval may be stored in a deduplication manner as one data slice (e.g., may also be referred to as a data block) of the snapshot.

In one embodiment, the size of the address interval may be, for example, 2MB, and in practical application, the user may be determined according to the practical requirement, which is not limited herein specifically.

As shown in fig. 2, assuming that the disk may be divided into 4 slices (e.g., slice 1, slice 2, slice 3, and slice 4) by address offset, a disk snapshot is created at a first point in time, resulting in snapshot a.

As one example, snapshot a may include, for example: slice 1-A, slice 2-A, slice 3-A, and slice 4-A.

With continued reference to FIG. 2, a disk snapshot is created at a second point in time, resulting in snapshot B. It is assumed that only the data of the disk sections corresponding to slice 1 and slice 3 are modified during the time intervals of the first time point and the second time point.

Thus, the slice of snapshot B includes newly created slice 1-B for slice 1 and slice 3-B for slice 3. While slices 2 and 4 continue to use slices 2-a and 4-a in snapshot a.

As can be seen from fig. 2, when a snapshot is created, each address interval of the current disk is checked: if the slice corresponding to the data content in the address interval is changed compared with the slice corresponding to the data content in the address interval when the snapshot is created last time, the snapshot creates a new slice by using the updated data in the address interval; otherwise, if not changed, the new snapshot will continue to use the slice corresponding to the data content in the address space last time the snapshot was created. That is, an incremental backup mechanism may be used to create disk snapshots, and only the data slices of the address intervals where the data changes are backed up between two times of creating disk snapshots, so as to improve the backup efficiency and save the storage space.

In the embodiment of the invention, the snapshot creation process specifically comprises the following two steps:

first, metadata of disk data is constructed.

In the step, the address interval of the disk is analyzed, the address interval with data modification relative to the existing snapshot data is determined, the address interval with data modification is marked as the address interval needing backup, and the address interval needing backup is recorded in the snapshot metadata.

In general, the construction of metadata may take a time scale of seconds, for example, 1-5 seconds, in order to avoid that address intervals that need to be backed up during the creation process are written with data, during the construction of metadata of disk data, a disk may block a read-write request for the disk until the metadata creation is completed.

Second, the snapshot data is transferred to the snapshot storage system.

In the step, when the metadata construction of the disk data is completed, the snapshot data of the disk can be determined, and the snapshot data needs to be transmitted to a snapshot storage system for backup.

In the metadata construction process, the data slices of the address intervals which need to be backed up to the snapshot storage system can be determined, and the data slices of the address intervals are transmitted to the snapshot storage system one by one and stored as the data blocks of the new snapshot.

By the steps, the snapshot can be created on the disk, the incremental backup can be carried out on the data in the disk, the data safety and the data storage efficiency are improved, and the storage space is saved.

In a practical application scenario, there is a scenario that requires several disks to be able to create a snapshot at the same point in time. As an example, when multiple disks form a disk group, there is a data association between the disks, and a snapshot of the disk group cannot be created at the same point in time, the data recovered from the snapshot of the disk group may not necessarily meet the requirement of the data association between the disks, and the snapshot of the disk group is not available in service.

As an example, in service, if a database spans two disks, i.e., the database needs to use two disks, if two disks cannot create snapshots very accurately at the same point in time, then using the created snapshots to recover data, it is possible that disk a recovers data at one point in time, e.g., 12 minutes and 10 seconds, and disk B recovers data at another point in time, e.g., 12 minutes and 25 seconds, and the database cannot be completed and maintained consistent by recovering the data.

Therefore, the higher the time precision of synchronously creating the snapshot, the greater the success rate of recovering the complete data with the created snapshot.

In one embodiment, a plurality of snapshots may be considered to be available when a maximum value of a time difference between a plurality of creation time points corresponding to the plurality of snapshots is within a preset time difference threshold. As one example, the predetermined time difference threshold range may be 0.5 seconds or less. It should be understood that the preset time difference threshold range may be set by itself according to a requirement for time accuracy of creating the snapshot synchronously, and embodiments of the present invention are not limited in particular.

FIG. 3 is a flow diagram of a method for creating snapshots synchronously across multiple disks in one embodiment. As shown in fig. 3, in one embodiment, an existing synchronization creation disk snapshot system may include a management module, an input/output IO synchronization relay module (which may also be simply referred to as a synchronization module in the following description of the embodiment), and a plurality of disks. When the synchronous creation disk snapshot system is used for carrying out multi-disk synchronous creation snapshot, the method can comprise the following steps:

In step S310, as shown in S1 in fig. 3, the operating states of the plurality of disks are determined, for example, the operating states of the plurality of disks are determined to be normal states, for example, a disk state a, that is, an operating state in which the disk initiates a read/write request to the storage system.

In step S320, as shown in S2 in fig. 3, when the management and control module triggers the synchronous creation of a disk snapshot, the IO synchronous transfer module synchronizes the read-write requests of a plurality of disks to the storage system. At this time, the working states of the plurality of magnetic disks are the magnetic disk state B, namely the read-write synchronous state.

As shown in fig. 3, in the read-write synchronous state, the IO synchronous transfer module receives a plurality of read-write requests initiated by a plurality of disks to the storage system, queues the plurality of read-write requests, and sequentially issues the queued plurality of read-write requests.

In step S330, as shown in S3 in fig. 3, the IO synchronization transfer module stops issuing the read-write requests from the plurality of disks to the storage system. At this time, the working states of the plurality of magnetic disks are the data synchronization state of the magnetic disk state C.

As shown in fig. 3, the IO synchronous transfer module stops issuing read-write requests from the plurality of disks to the storage system at a certain moment, so that the plurality of disks stop data updating at the moment, and the plurality of disks reach a data synchronous state.

In step S340, as shown in S4 in fig. 3, in the data synchronization state, each disk creates snapshot metadata respectively, so as to obtain a plurality of snapshot metadata.

In step S350, as shown in S5 in fig. 3, the operating states of the plurality of disks are switched to the normal state, and the read-write requests of the plurality of disks to the storage system are recovered. And after the snapshot metadata is created, recovering the normal read-write states of the plurality of disks to the storage system.

According to the steps, an IO synchronous transfer module can be utilized in the prior art, and read-write requests of a plurality of disks can be synchronized in advance. That is, the write paths of the plurality of disks may pass through a common port; when the disk data is synchronously created, the read-write requests of the plurality of disks are cut off at the common access, so that the data writing of the plurality of disks can be cut off simultaneously, and the plurality of disks are frozen at the same time at the moment, so that the snapshots created by the plurality of disks can be regarded as being created at the same time, and the requirement of synchronously creating the snapshots is met.

Because the synchronous module itself needs to occupy certain machine and network bandwidth resources, and the synchronous transfer module is used as a service module, the manpower maintenance cost is higher. Thus, in the prior art, to create a snapshot for synchronization, a relatively high additional cost and resource is required.

In addition, in the process of creating the snapshot synchronously, when the plurality of disks are switched to be synchronous modules, the transmission performance of the N disks depends on the performance of the synchronous modules, so that the synchronous modules generally adopt serial issuing of synchronous requests of the plurality of disks for ensuring the sequential writing of requests, and when the number of the disks is large, the disk capacity is greatly affected. As an example, for example, 5 disks, each with a throughput of 100MB/s, but the synchronization module itself may have only 150MB/s of transfer capability, resulting in an average capacity per disk that may temporarily drop to 30MB/s.

And, before creating the snapshot, the synchronization module blocks all read-write requests and resumes the read-write requests after all disks have completed creating the snapshot metadata. That is, the metadata creates a fast disk, and needs to wait for the slow metadata creation disk to complete the creation before recovering the read-write request, so that the blocking time of the read-write request of the whole plurality of disks is increased.

From the above description, it is clear that, when disk data is created in the prior art, additional machine resources and maintenance costs are required; in the process of creating, an additional synchronous module is utilized to transfer the disk to a mode of synchronously converting a read-write request, so that the performance of the disk is obviously reduced; in the whole creation process, a plurality of disks are required to stop reading and writing requests together, snapshot metadata is created together, reading and writing requests are restored together, and a disk with a lower creation speed can enable a disk with a higher creation speed to not immediately restore the reading and writing requests to a storage system, so that the limited time of reading and writing of the whole batch of disks is prolonged, for example, the limited time of a single disk reading and writing request is longer than that of asynchronous snapshot creation.

In order to solve the defects of the scheme, the embodiment of the invention provides a disk snapshot creation method, which removes an IO synchronous transfer module set in a multi-disk synchronous snapshot creation system, thereby simplifying the system architecture, avoiding generating extra resource consumption and maintenance cost and improving the disk performance; each disk independently creates the snapshot, so that the condition that the limited duration of disk reading and writing is prolonged because the snapshot can be created by all the disks can be restored only is avoided, and the efficiency of disk snapshot creation is improved.

The disk snapshot creation method according to the embodiment of the present invention is described in detail below with reference to fig. 4 to 6.

FIG. 4 is a schematic diagram of a disk snapshot creation system according to an embodiment of the present invention. As shown in fig. 4, in one embodiment, the disk snapshot creation system may include a management module and a plurality of disks, e.g., cloud disk 1, cloud disk 2, … …, yun Cipan N.

In one embodiment, the management and control module may be configured to trigger a request to synchronously create a snapshot; multiple disks respond to the request of creating the snapshot synchronously to realize the creation of the snapshot synchronously.

In one embodiment, the management module may be a compute node with independent computing and processing capabilities. As one example, the management and control module may include a central processor, and the management and control module may operate in the same server as the snapshot storage system, or may operate in a different server from the snapshot storage system, independent of the snapshot storage system.

In one embodiment, the management module may be coupled to the plurality of disks using a communications network. And in the description of the embodiment of the invention, plural may represent 1 or more.

FIG. 5 illustrates a flowchart of a disk snapshot creation method according to an exemplary embodiment of the present invention. As shown in fig. 5, in one embodiment, the disk snapshot creation method may include:

in step S501, the management module starts to create a synchronized snapshot for a plurality of disks, such as disk 1 through disk N.

Step S502 sets a synchronization initiation waiting time for a plurality of disks.

In this step, the synchronization initiation wait time may be empirically set, assuming that the duration of the synchronization initiation wait time is t seconds for descriptive convenience.

Step S503, the control module obtains a current time stamp T;

in step S504, the management and control module sends a snapshot creation command to the plurality of disks, so that the plurality of disks start to create snapshots at a time point t+t.

In step S505, the plurality of disks, upon receiving the creation command, instruct each disk to start creating a snapshot at a time point of t+t.

In this step, the management module may decide to have multiple disks begin creating a snapshot at a point in time by formulating the point in time, such as the current time plus 5 seconds. Issuing a creation command to a plurality of disks, and waiting for each disk to create a snapshot according to a planned time point.

In step S506, each disk waits to the t+t time point according to the time stamp of the day.

In step S507, the disk blocks the disk read/write request, and records the time point of blocking the disk read/write request.

As one example, a point in time X at which a disk read-write request is blocked, for example, is recorded.

In step S508, after each disk blocks the disk read/write request, snapshot metadata is created.

In step S509, the snapshot metadata creation is completed, and each disk returns the recorded time point X of blocking the disk read-write request to the management module.

In this step, a plurality of disks, for example, N disks, report the point in time when the IO starts to be blocked when the snapshot is created, respectively.

In step S510, each disk resistor transmits the created disk snapshot metadata to the snapshot storage system.

In step S511, the management and control module waits for the completion of the creation of the plurality of disk snapshots, and waits for a time point of the plurality of blocked disk read/write requests for the current creation command returned by the plurality of disks.

In step S512, it is determined whether the maximum time difference between the time points of the multiple blocked disk read/write requests is smaller than the set time difference threshold M.

In step S513, when the maximum time difference between the time points of the multiple blocking disk read/write requests is greater than or equal to the set time difference threshold M, it is determined that the synchronization accuracy requirement is not satisfied, and the multiple snapshot data transmitted to the snapshot storage system are deleted.

In step S514, when the maximum time difference between the time points of the multiple blocking disk read-write requests is smaller than the set time difference threshold M, it is determined that the synchronization precision requirement is met, and the synchronization snapshot is successfully created.

As an example, assume that the number of disks is 5, i.e., N has a value of 5, and there are 5 disks to create the snapshot synchronously. Controlling to find that the current time point is 13 hours, 05 minutes and 03 seconds, and starting to create a snapshot after 5 seconds are planned; and sending a creation command to the magnetic disks 1 to 5 according to the planned time length.

In the example, each of the disk 1 to the disk 5 waits for 13 hours for 05 minutes and 08 seconds, starts to block IO requests, and creates a snapshot; and after the 5 snapshots of the disk 1-5 are successfully created, respectively informing the management and control module of the time stamp of blocking the disk IO request. Assume that the time stamps of blocking disk IO requests from disk 1-disk 5 include:

13:05:08.145；

13:05:08.487；

13:05:08.098；

13:05:08.253；

13:05:08.313；

the difference between the maximum timestamp and the minimum timestamp is 487-98=389 milliseconds, and if the set time difference m=500 milliseconds of the synchronization precision requirement is set, the 5 snapshots meet the requirement, and the creation of the synchronization snapshot is considered to be successful.

Compared with the prior art, the method for creating the disk snapshot simplifies the architecture module of the disk snapshot creating system, omits the IO synchronous transfer module set in the disk snapshot creating system in the prior art, and avoids extra resource consumption and cost.

In this embodiment, each disk creates a snapshot independently, and it is not necessary to transfer the read-write request to the synchronization module first, and there is naturally no performance bottleneck of the synchronization module.

Because each disk independently creates the snapshot, the disk can restore the disk read-write of itself by itself after the creation is completed, and the situation that the disk with fast snapshot creation is required to be completed by the disk with slow snapshot creation speed, and the read-write can be restored can not occur.

According to the disk snapshot creation method provided by the embodiment of the invention, the management and control module compares a plurality of blocking time points of a plurality of received blocking disk read-write requests, and if the time difference among the blocking time points meets the set threshold value (for example, the time difference is smaller than 0.5 seconds) of the synchronous snapshot, the created snapshots are considered to meet the requirement of synchronous precision, and the creation of the synchronous snapshot is considered to be successful. Otherwise, deleting the N snapshots, and then retrying creation again.

In the embodiment of the present invention, public cloud systems corresponding to a cloud server, a cloud storage system, such as a cloud disk, a snapshot storage system, and the like, generally use a network time protocol (Network Time Protocol, NTP) to maintain time clock consistency with each other. And under the normal state, the time consumption of blocking the read-write request of each disk is basically consistent.

Therefore, the disk snapshot creating method in the embodiment of the invention not only can simplify the engineering realization of the disk snapshot creating system, but also can achieve better synchronization precision. If the time difference between the plurality of blocking time points is 0.5 seconds, which can be achieved by one synchronous creation, the new creation time point can be repeatedly formulated a plurality of times to further reduce the time difference between the plurality of blocking time points, for example, 0.1 seconds, thereby improving the synchronous creation accuracy.

FIG. 6 illustrates a flow chart of a disk snapshot creation method according to one embodiment of the invention. As shown in FIG. 6, in one embodiment, a disk snapshot creation method 600 may include:

in step S610, a snapshot creation command is sent, where the snapshot creation command is used to indicate a point in time when a plurality of disks create snapshots.

Step S620, for the snapshot creation command, receives a plurality of blocking time points from a plurality of disk creating snapshots in which the disk read/write requests are blocked.

In step S630, when the maximum value of the time differences between the multiple blocking time points is smaller than the set first time difference threshold, it is determined that the snapshots created by the multiple disks meet the synchronization precision requirement, and multiple snapshots created by the multiple disks in synchronization are obtained.

According to the disk snapshot creation method provided by the embodiment of the invention, the management and control module can analyze and judge whether the creation time of a plurality of snapshots meets the synchronous precision requirement on the service through the received IO blocking time in the snapshot creation process. The whole creation process does not need to rely on a synchronization module in advance to ensure read-write synchronization, but only needs to check the synchronization precision according to the IO blocking time difference among multiple disks after the snapshot is created, so that the system architecture is simpler, a new module is not needed to be referenced, the problem of extra performance degradation for the disks is avoided, the problem that the IO blocking time for creating the snapshot is more than that of a single disk snapshot is avoided, and the IO performance of the disks is ensured.

In one embodiment, step S610 may specifically include:

step S611, setting a synchronous waiting time length, wherein the synchronous waiting time length is used for indicating waiting time lengths before the plurality of disks create the snapshot;

step S612, generating a snapshot creation command according to the waiting time before creating the snapshot;

step S613, a snapshot creation command is sent.

In this embodiment, the management and control module may set a synchronization waiting duration for creating a snapshot for the disk based on a clock consistent with the disk, so that the disk starts to create the snapshot after the synchronization waiting duration, and time accuracy of synchronization creation is ensured.

In one embodiment, step S610 may specifically include:

step S614, setting a synchronous creation time point, wherein the synchronous creation time point is later than the current time point;

step S615, generating a snapshot creation command according to the synchronous creation time point;

step S616, a snapshot creation command is sent.

In this embodiment, the management and control module may directly set the synchronization creation time point, so that each disk creates the snapshot at the synchronization creation time point at respective timings according to the synchronization creation time point. Since a plurality of disks can be based on clocks consistent with the disks, the time accuracy of synchronous creation is ensured.

In one embodiment, step S630 may specifically include:

step S631, calculating a time difference maximum value between the plurality of blocking time points by using the blocking time point maximum value and the blocking time point minimum value among the plurality of blocking time points;

in step S632, when the maximum time difference is smaller than the set first time difference threshold, it is determined that the snapshots created by the multiple disks synchronously meet the synchronization precision requirement.

In the embodiment, by recording IO blocking time in the process of creating the snapshots, whether the creation time of a plurality of snapshots meets the synchronous precision requirement on the service or not is analyzed and judged, the synchronous precision requirement is met, and a plurality of disk snapshots are obtained.

In one embodiment, disk snapshot creation method 600 may further include:

in step S640, when the maximum time difference is greater than or equal to the set first time difference threshold, a new snapshot creation command is sent, where the new snapshot creation command is used to instruct the multiple disks to synchronously create a new time point of the snapshot after receiving the snapshot creation command.

In one embodiment, disk snapshot creation method 600 may further include:

in step S650, when the maximum value of the time difference is smaller than the first time difference threshold, a new snapshot creation command is sent until the maximum value of the new time difference is smaller than a set second time difference threshold, so as to obtain a plurality of new snapshots created by a plurality of disks synchronously, wherein the second time difference threshold is smaller than the first time difference threshold.

In this embodiment, since the clock consistency used between the disks is higher, the time consumption of blocking the read-write request between the disks is basically consistent, and the time difference between a plurality of blocking time points can be gradually reduced by creating the snapshots at multiple times, so that the creation time difference between the disks is smaller and smaller, and the consistency between a plurality of snapshots obtained by synchronous creation is higher and higher.

According to the disk snapshot creation method provided by the embodiment of the invention, on the basis of not adding an additional module, by recording IO blocking time in the snapshot creation process, whether the creation time of a plurality of snapshots meets the synchronous precision requirement on business is judged, and the disk snapshot creation method has higher calculation efficiency on the premise of lower resource consumption and cost and no influence on the disk read-write performance.

FIG. 7 illustrates a disk snapshot creation method according to another embodiment of the present invention. As shown in FIG. 7, in one embodiment, the disk snapshot creation method 700 may specifically include:

step S710, receiving a snapshot creation command, wherein the snapshot creation command is used for indicating a time point of creating a snapshot;

step S720, when waiting until a time point of creating the snapshot, creating disk snapshot data, and transmitting the snapshot data to a snapshot storage system;

step S730, recording a blocking time point when a disk read-write request is blocked in the process of creating disk snapshot data according to the snapshot creation command;

step S740, sending the blocking time points, so that the management node determines whether the created snapshot data meets the synchronization accuracy requirement based on the received multiple blocking time points.

In the embodiment, each disk responds to the snapshot creation command, creates snapshots independently, and can restore the IO request of itself after the creation is completed, so that the condition that the disk with fast snapshot creation speed needs to be completed to restore the IO is avoided, and the IO capability of the disk is guaranteed to a great extent.

In one embodiment, step S720 may specifically include:

when the snapshot creation command comprises the synchronous waiting time, determining the time point of creating the snapshot of the disk according to the current time point of the disk and the synchronous waiting time; wait until a point in time when a snapshot of the disk is created.

In this embodiment, the disk may wait according to the synchronization waiting duration, and after the synchronization waiting duration is reached, the snapshot is started to be created, so as to realize the synchronous snapshot creation between the disks.

In one embodiment, step S720 may specifically include:

when the snapshot creation command comprises a set synchronous creation time point, waiting from the current time point of the disk to the set synchronous creation time point.

In the embodiment, the disk can automatically and regularly create the snapshot according to the synchronous creation time point, so that the snapshot can be synchronously created among the disks. And the time clocks among the disks are maintained consistent by using a network time protocol, so that the snapshot is synchronously created among the disks.

Fig. 8 shows a schematic structural diagram of a disk snapshot creating apparatus according to an embodiment of the present invention. As shown in fig. 8, the disk snapshot creation device 800 may include:

the creation command sending module 810 is configured to send a snapshot creation command, where the snapshot creation command is used to indicate a point in time when a snapshot is created by a plurality of disks.

The blocking time receiving module 820 is configured to receive, for a snapshot creation command, a plurality of blocking time points from blocking disk read/write requests during a process of creating snapshots by a plurality of disks.

The synchronization precision determining module 830 is configured to determine that the snapshots created by the plurality of disks meet the synchronization precision requirement when a maximum value of a time difference between the plurality of blocking time points is smaller than a set first time difference threshold, and obtain a plurality of snapshots created by the plurality of disks synchronously.

In one embodiment, the create command sending module 810 may include:

the waiting time length setting unit is used for setting synchronous waiting time length which is used for indicating waiting time length before the plurality of disks create the snapshot;

the first command generating unit is used for generating a snapshot creating command according to the waiting time before creating the snapshot;

and the first command sending unit is used for sending the snapshot creation command.

In one embodiment, the create command sending module 810 may include:

the creation time setting unit is used for setting a synchronous creation time point which is later than the current time point;

a second command generating unit for creating a command generating unit for generating a snapshot creating command according to the synchronous creating time point;

and the second command sending unit is used for sending the snapshot creation command.

In one embodiment, the synchronization accuracy determination module 830 includes:

a time difference calculation unit for calculating a time difference maximum value between a plurality of congestion time points using a congestion time point maximum value and a congestion time point minimum value among the plurality of congestion time points;

and the synchronization precision judging unit is used for determining that the snapshots created by the plurality of disks synchronously meet the synchronization precision requirement when the maximum time difference value is smaller than the set first time difference threshold value.

In one embodiment, the create command sending module 810 may also be specifically configured to:

and when the maximum value of the time difference is greater than or equal to a set first time difference threshold value, a new snapshot creation command is sent, wherein the new snapshot creation command is used for indicating a new time point for synchronously creating the snapshot after the plurality of magnetic disks receive the snapshot creation command.

and when the maximum value of the time difference is smaller than the first time difference threshold, sending a new snapshot creation command until the maximum value of the new time difference is smaller than a set second time difference threshold, and obtaining a plurality of new snapshots created by a plurality of disks synchronously, wherein the second time difference threshold is smaller than the first time difference threshold.

Fig. 9 shows a schematic structural diagram of a disk snapshot creating apparatus according to an embodiment of the present invention. As shown in fig. 9, the disk snapshot creation device 900 may include:

the creation command receiving module 910 is configured to receive a snapshot creation command, where the snapshot creation command is used to indicate a point in time when a snapshot is created.

The snapshot data creation module 920 is configured to create disk snapshot data and transmit the snapshot data to the snapshot storage system when waiting until a point in time when the snapshot is created.

And the blocking time recording module 930 is configured to record, for the snapshot creation command, a blocking time point when the disk read-write request is blocked in the process of creating the disk snapshot data.

And the blocking time sending module 940 is configured to send blocking time points, so that the management and control node determines whether the created snapshot data meets the synchronization accuracy requirement based on the multiple blocking time points.

In one embodiment, the snapshot data creation module 920, when used to wait until the point in time when the snapshot is created, is specifically further used to:

It should be clear that the invention is not limited to the specific arrangements and processes described in the foregoing embodiments and shown in the drawings. For convenience and brevity of description, detailed descriptions of known methods are omitted herein, and specific working processes of the systems, modules and units described above may refer to corresponding processes in the foregoing method embodiments, which are not repeated herein.

FIG. 10 is a block diagram illustrating an exemplary hardware architecture of a computing device capable of implementing disk snapshot creation methods and apparatus in accordance with embodiments of the present invention.

As shown in fig. 10, the computing device 1000 includes an input device 1001, an input interface 1002, a central processor 1003, a memory 1004, an output interface 1005, and an output device 1006. The input interface 1002, the central processing unit 1003, the memory 1004, and the output interface 1005 are connected to each other via a bus 1010, and the input device 1001 and the output device 1006 are connected to the bus 1010 via the input interface 1002 and the output interface 1005, respectively, and further connected to other components of the computing device 1000.

Specifically, the input device 1001 receives input information from the outside, and transmits the input information to the central processor 1003 through the input interface 1002; the central processor 1003 processes the input information based on computer executable instructions stored in the memory 1004 to generate output information, temporarily or permanently stores the output information in the memory 1004, and then transmits the output information to the output device 1006 through the output interface 1005; output device 1006 outputs output information to the outside of computing device 1000 for use by a user.

In one embodiment, the computing device 1000 shown in FIG. 10 may be implemented as a disk snapshot creation system that may include: a memory configured to store a program; and a processor configured to run a program stored in the memory to execute the disk snapshot creating method described in the above embodiment.

The processes described above with reference to flowcharts may be implemented as computer software programs according to embodiments of the present invention. For example, embodiments of the present invention include a computer program product comprising a computer program tangibly embodied on a machine-readable medium, the computer program comprising program code for performing the method shown in the flowchart. In such embodiments, the computer program may be downloaded and installed from a network, and/or installed from a removable storage medium.

In the above embodiments, it may be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product comprises one or more computer instructions which, when run on a computer, cause the computer to perform the methods described in the various embodiments above. When loaded and executed on a computer, produces a flow or function in accordance with embodiments of the present invention, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a computer network, or other programmable apparatus. The computer instructions may be stored in or transmitted from one computer-readable storage medium to another, for example, by wired (e.g., coaxial cable, optical fiber, digital Subscriber Line (DSL)), or wireless (e.g., infrared, wireless, microwave, etc.). The computer readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that contains an integration of one or more available media. The usable medium may be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid state disk), etc.

The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.

Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some or all of the technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A disk snapshot creation method, comprising:

Sending a snapshot creation command, wherein the snapshot creation command is used for indicating time points of creating snapshots of a plurality of disks;

receiving a plurality of blocking time points for blocking disk read-write requests in the process of creating snapshots by a plurality of disks aiming at the snapshot creation command;

and when the maximum value of the time difference among the plurality of blocking time points is smaller than a set first time difference threshold value, determining that the snapshots created by the plurality of magnetic disks meet the requirement of synchronization precision, and obtaining a plurality of snapshots created by the plurality of magnetic disks synchronously.

2. The disk snapshot creation method of claim 1, wherein the sending a snapshot creation command includes:

setting synchronous waiting time length, wherein the synchronous waiting time length is used for indicating waiting time length before the plurality of disks create the snapshot;

generating a snapshot creation command according to the waiting time before creating the snapshot;

and sending the snapshot creation command.

3. The disk snapshot creation method of claim 1, wherein the sending a snapshot creation command includes:

setting a synchronous creation time point, wherein the synchronous creation time point is later than the current time point;

generating a snapshot creation command according to the synchronous creation time point;

And sending the snapshot creation command.

4. The disk snapshot creation method of claim 1, wherein the determining that the snapshots synchronously created by the plurality of disks satisfy a synchronization accuracy requirement when a maximum value of time differences between the plurality of blocking time points is less than a set first time difference threshold value comprises:

calculating a time difference maximum value between the plurality of blocking time points using a blocking time point maximum value and a blocking time point minimum value of the plurality of blocking time points;

and when the maximum time difference value is smaller than a set first time difference threshold value, determining that the snapshots synchronously created by the plurality of magnetic disks meet the requirement of synchronous precision.

5. The disk snapshot creation method of claim 1, further comprising:

and when the maximum time difference value is greater than or equal to a set first time difference threshold value, a new snapshot creation command is sent, wherein the new snapshot creation command is used for indicating a new time point for synchronously creating the snapshot after a plurality of magnetic disks receive the snapshot creation command.

6. The disk snapshot creation method of claim 1, further comprising:

when the maximum value of the time difference is smaller than the first time difference threshold value, a new snapshot creation command is sent until the maximum value of the new time difference is smaller than a set second time difference threshold value, a plurality of new snapshots which are created by the plurality of disks synchronously are obtained, wherein,

The second time difference threshold is less than the first time difference threshold.

7. A disk snapshot creation method, comprising:

receiving a snapshot creation command, wherein the snapshot creation command is used for indicating time points of creating snapshots of a plurality of disks;

when waiting until the time point of creating the snapshot, creating disk snapshot data, and transmitting the snapshot data to a snapshot storage system;

recording a blocking time point when a disk read-write request is blocked in the process of creating the disk snapshot data aiming at the snapshot creation command;

and sending the blocking time points so that the control node determines whether the created snapshot data meets the requirement of synchronous precision based on the received multiple blocking time points.

8. The disk snapshot creation method of claim 7, wherein the waiting to the point in time of creating the snapshot includes:

when the snapshot creation command comprises synchronous waiting time, determining a time point of creating the snapshot of the magnetic disk according to the current time point of the magnetic disk and the synchronous waiting time;

waiting until a point in time when a snapshot of the disk is created.

9. The disk snapshot creation method of claim 7, wherein the waiting to the point in time of creating the snapshot includes:

10. A disk snapshot creation device, comprising:

the system comprises a creation command sending module, a storage module and a storage module, wherein the creation command sending module is used for sending a snapshot creation command, and the snapshot creation command is used for indicating time points of creating snapshots of a plurality of magnetic disks;

the blocking time receiving module is used for receiving a plurality of blocking time points for blocking disk read-write requests in the process of creating snapshots by a plurality of disks aiming at the snapshot creation command;

and the synchronization precision determining module is used for determining that the snapshots created by the plurality of magnetic disks meet the synchronization precision requirement when the maximum value of the time difference between the plurality of blocking time points is smaller than a set first time difference threshold value, and obtaining the plurality of snapshots created by the plurality of magnetic disks synchronously.

11. A disk snapshot creation device, comprising:

the system comprises a creation command receiving module, a storage module and a storage module, wherein the creation command receiving module is used for receiving a snapshot creation command, and the snapshot creation command is used for indicating time points of creating snapshots of a plurality of magnetic disks;

the snapshot data creation module is used for creating disk snapshot data and transmitting the snapshot data to the snapshot storage system when waiting until the snapshot creation time point;

The blocking time recording module is used for recording a blocking time point when a disk read-write request is blocked in the process of creating the disk snapshot data aiming at the snapshot creation command;

and the blocking time sending module is used for sending the blocking time points so that the control node can determine whether the created snapshot data meets the synchronous precision requirement or not based on a plurality of blocking time points.

12. A disk snapshot creation system, comprising a memory and a processor;

the memory is used for storing executable program codes;

the processor is configured to read executable program code stored in the memory to perform the disk snapshot creation method of any of claims 1-6.

13. A disk snapshot creation system, comprising a memory and a processor;

the memory is used for storing executable program codes;

the processor is configured to read executable program code stored in the memory to perform the disk snapshot creation method of any of claims 7 to 9.

14. A computer readable storage medium comprising instructions that when run on a computer cause the computer to perform the disk snapshot creation method of any of claims 1 to 6.

15. A computer readable storage medium comprising instructions which, when run on a computer, cause the computer to perform the disk snapshot creation method of any of claims 7 to 9.