+

WO2018153251A1 - 一种快照的处理方法及分布式块存储系统 - Google Patents

一种快照的处理方法及分布式块存储系统 Download PDF

Info

Publication number
WO2018153251A1
WO2018153251A1 PCT/CN2018/075458 CN2018075458W WO2018153251A1 WO 2018153251 A1 WO2018153251 A1 WO 2018153251A1 CN 2018075458 W CN2018075458 W CN 2018075458W WO 2018153251 A1 WO2018153251 A1 WO 2018153251A1
Authority
WO
WIPO (PCT)
Prior art keywords
read
write
operations
snapshot mode
preset
Prior art date
Application number
PCT/CN2018/075458
Other languages
English (en)
French (fr)
Inventor
张广艳
罗圣美
都涛
杨洪章
Original Assignee
中兴通讯股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Priority to EP18757805.9A priority Critical patent/EP3588330A4/en
Priority to AU2018223401A priority patent/AU2018223401A1/en
Priority to US16/487,820 priority patent/US11455276B2/en
Publication of WO2018153251A1 publication Critical patent/WO2018153251A1/zh
Priority to AU2021204579A priority patent/AU2021204579A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/14—Error detection or correction of the data by redundancy in operation
    • G06F11/1402—Saving, restoring, recovering or retrying
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/10—File systems; File servers
    • G06F16/11—File system administration, e.g. details of archiving or snapshots
    • G06F16/128—Details of file system snapshots on the file-level, e.g. snapshot creation, administration, deletion
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/10—File systems; File servers
    • G06F16/18—File system types
    • G06F16/182—Distributed file systems
    • G06F16/184—Distributed file systems implemented as replicated file system
    • G06F16/1844—Management specifically adapted to replicated file systems

Definitions

  • This document relates to, but is not limited to, the field of communication technologies, and in particular, to a snapshot processing method and a distributed block storage system.
  • a typical distributed storage system usually consists of three parts: a client, a metadata server, and a data server. Snapshot technology provides storage systems with the ability to perform online backup, data recovery, and data reuse. It is an indispensable part of distributed block storage systems.
  • Embodiments of the present invention provide a snapshot processing method and a distributed block storage system.
  • An embodiment of the present invention provides a method for processing a snapshot, including:
  • the embodiment of the invention further provides a distributed block storage system, comprising:
  • a data acquisition module configured to acquire a number of read operations and a number of read and write operations of the data block in the distributed block storage system
  • the snapshot mode determining module is configured to determine a target snapshot mode based on the number of read operations and the number of read and write operations, wherein the target snapshot mode includes copy-on-write or write-time redirection;
  • a processing module configured to perform a snapshot operation on the data block in the distributed block storage system based on the target snapshot mode.
  • the embodiment of the present invention further provides a computer storage medium, where the computer storage medium stores one or more programs executable by a computer, and when the one or more programs are executed by the computer, the computer is executed as described above.
  • a method of processing a snapshot provided.
  • the snapshot of the distributed storage system when the data block is mainly read, it is determined that the data block is snapshotted by copy-on-write; when the data block is write-based, it is determined that the data block is redirected by writing. Snapshot operations, so that write performance, rollback performance, and read performance are both good.
  • Figure 1 is a schematic diagram of a copy-on-write snapshot
  • Figure 2 is a schematic diagram of re-scheduled snapshots during writing
  • FIG. 3 is a schematic flowchart of a method for processing a snapshot according to an embodiment of the present disclosure
  • FIG. 4 is a schematic flowchart diagram of another method for processing a snapshot according to an embodiment of the present disclosure
  • FIG. 5 is a schematic structural diagram of a part of a distributed block storage system according to an embodiment of the present disclosure
  • FIG. 6 is a schematic structural diagram of a distributed block storage system according to an embodiment of the present disclosure.
  • FIG. 7 is a schematic structural diagram of a snapshot mode determining module in a distributed block storage system according to an embodiment of the present disclosure
  • FIG. 8 is a schematic structural diagram of a snapshot mode determining module in another distributed block storage system according to an embodiment of the present disclosure.
  • FIG. 9 is a schematic structural diagram of another distributed block storage system according to an embodiment of the present invention.
  • a distributed block storage system usually uses a copy-on-write (COW) and a redirect on-write (ROW) snapshot method.
  • COW has better read performance
  • ROW has better write performance and rollback performance.
  • COW performs write and rollback operations
  • data copying is required first, so write performance and rollback performance are poor.
  • ROW is read, the read performance is poor due to the need to relocate the data.
  • an embodiment of the present invention provides a schematic flowchart of a method for processing a snapshot.
  • the method is applied to a distributed block storage system, and includes the following steps:
  • Step S301 Acquire a number of read operations and a number of read and write operations of the data block in the distributed block storage system.
  • the distributed storage system can obtain the number of read operations and the number of read and write operations of each data block through the disk operation monitoring tool. For example, input the iostat command in the distributed storage system, and store the data of the data block in real time. Load data such as the number of times the server reads and writes (I/O) operations (Input/Output Operations Per Second, IOPS), so as to obtain the number of read operations and the number of read and write operations of the data block.
  • I/O Input/Output Operations Per Second
  • Step S302 Determine a target snapshot mode based on the number of read operations and the number of read and write operations, where the target snapshot mode includes copy-on-write or write-time redirection.
  • the read and write of the data block can be determined based on the number of read operations of the data block and the number of read and write operations. If it is determined that the data block is mainly read, the COW can be used as a snapshot mode for performing snapshot operations on the data block; if it is determined that the data block is write-oriented, ROW can be used as a snapshot mode for performing snapshot operations on the data block.
  • the target snapshot mode may be determined by the ratio of the number of read operations of the data block to the number of read and write operations or the difference between the read and write operations, and the ratio or difference is larger, indicating The data block is dominated by reading; the smaller the ratio or difference, the more the data block is written. For example, if the ratio of the number of read operations to the number of read and write operations is 10, the data block is read-based, and the target snapshot mode can be determined as COW; if the ratio of the number of read operations to the number of read and write operations is 0.1, the data block is Write-based, you can determine the target operation mode is ROW.
  • Step S303 Perform a snapshot operation on the data block in the distributed block storage system based on the target snapshot mode.
  • the distributed storage system may perform a snapshot operation on the data block by using the target snapshot mode.
  • the snapshot operation may include any one of a snapshot operation, a read operation, a write operation, a rollback operation, and a delete snapshot operation.
  • the number of read operations and the number of read and write operations of the data block in the distributed block storage system are obtained; and the target snapshot mode is determined based on the number of read operations and the number of read and write operations, wherein the target snapshot mode includes copy-on-write or Write-time redirection; snapshot operations on data blocks in a distributed block storage system based on the target snapshot mode.
  • the snapshot of the distributed storage system when the data block is mainly read, it is determined that the data block is snapshotted by copy-on-write; when the data block is write-based, it is determined that the data block is redirected by writing. Snapshot operations, so that write performance, rollback performance, and read performance are both good.
  • the embodiment of the present invention provides a schematic flowchart of another method for processing a snapshot, including the following steps:
  • Step S401 Acquire a number of read operations and a number of read and write operations of the data block in the distributed block storage system.
  • the distributed block storage system when the client initiates the snapshot operation request, responds to the snapshot operation request, and searches for the data block corresponding to the snapshot operation request in the data server by using the snapshot metadata stored in the metadata server.
  • the position in the middle, and the number of read operations and the number of read and write operations of each data block are obtained through the disk operation monitoring tool.
  • the IOPS command is used to count the load data such as the IOPS of the data server storing the data block, and the data block is read. Number of operations and number of read and write operations.
  • Step S402 Determine whether the preset parameter value of the number of read operations and the number of read/write operations is greater than or equal to a first preset threshold, wherein the preset parameter value is a ratio or a difference between the number of read operations and the number of read and write operations.
  • the ratio or the difference between the number of read operations of the data block and the number of read and write operations is used as a preset parameter value, which can intuitively reflect the read and write status of the data block, the ratio of the number of read operations to the number of read and write operations or When the difference is larger, it indicates that the data block has more read operations; when the ratio of the number of read operations to the number of read/write operations or the difference is larger, it indicates that the data block write operation is more.
  • the first preset threshold may be set to 1.5. If the preset parameter value is greater than or equal to 1.5, the data block may be determined to be read.
  • the first preset threshold may be set to 50. If the preset parameter value is greater than or equal to 50, the data block may be determined to be read. .
  • Step S403 If the preset parameter value of the number of read operations and the number of read/write operations is greater than or equal to the first preset threshold, determine that the target snapshot mode is copy-on-write.
  • the data block when the step S403 determines that the preset parameter value of the number of read operations and the number of read and write operations is greater than or equal to the first preset threshold, the data block may be determined to be read, so that the target snapshot mode may be determined as Write-on-write with better read performance, resulting in better read performance for distributed storage systems.
  • Step S404 If the preset parameter value of the number of read operations and the number of read/write operations is less than or equal to the first preset threshold, determine that the target snapshot mode is write-time redirection.
  • step S403 when the step S403 determines that the preset parameter value of the number of read operations and the number of read and write operations is less than the first preset threshold, it may be determined that the data block is not mainly read, and the data block may be read/write close or Write-based, you can determine the target snapshot mode for write-time redirection with better write performance and rollback performance, so that distributed storage systems have better write performance and rollback performance.
  • the foregoing determining the target snapshot mode may include:
  • the preset parameter value of the number of read operations and the number of read and write operations is greater than the second preset threshold, determine whether the front end load value is greater than a preset load threshold, where the front end load value is a load value of the data server storing the data block;
  • the front-end load value is greater than the preset load threshold or the number of read operations and the preset parameter value of the number of read/write operations are less than or equal to the second preset threshold, performing the step of determining that the target snapshot mode is write-time redirection.
  • the data block when the number of read operations of the data block and the preset parameter value of the number of read/write operations are smaller than the second preset threshold, the data block may be determined to be write-based, so that the target snapshot mode may be determined to be better.
  • Write-time redirection of write performance and rollback performance when the preset number of read operations of the data block and the number of read/write operations are smaller than the first preset threshold and greater than the second preset threshold, the data block at this time The read and write is close.
  • the front end load value of the data server storing the data block can be obtained. Comparing the front-end load value with the preset load threshold, if the front-end load value is greater than the preset load threshold, indicating that the data server has a high load, the target snapshot mode may be determined to be ROW, thereby reducing the write request overhead of the data block; The front-end load value is less than or equal to the preset load threshold, indicating that the data server load is low, and the target snapshot mode can be determined as COW, thereby improving the performance of the read request of the data block.
  • the first preset threshold may be set to 1.5
  • the second preset threshold is 0.7
  • the preset load threshold is 10000, if the read operation is performed.
  • the ratio of the number of times to the number of read and write operations is less than or equal to 0.7, indicating that the data block is write-based, and the target snapshot mode is determined to be ROW; if the ratio of the number of read operations to the number of read and write operations is greater than 0.7 and less than 1.5, indicating that the data block is read.
  • Write close if the front end load value of the corresponding data server is 15000, it is determined that the target snapshot mode is ROW.
  • first preset threshold, the second preset threshold, and the preset load threshold may be set according to actual needs, and are not limited herein.
  • Step S405 Perform a snapshot operation on the data block in the distributed block storage system based on the target snapshot mode.
  • the distributed storage system may perform a snapshot operation on the data block by using the target snapshot mode.
  • the method further includes: determining whether the current snapshot mode matches the target snapshot mode;
  • the current snapshot mode is switched to the target snapshot mode.
  • the distributed storage system may match the target snapshot mode with the current snapshot mode of the data block. If the target snapshot mode does not match the current snapshot mode, the current snapshot mode may be switched. The way to target snapshots. For example, if the current snapshot mode is COW and the target snapshot mode is ROW, you can switch the current snapshot mode of the data block from COW to ROW. If the current snapshot mode is ROW and the target snapshot mode is COW, you can The current snapshot mode of the data block is switched from ROW to COW.
  • the foregoing snapshot operations include creating a snapshot operation, a read operation, a write operation, a rollback operation, and a delete snapshot operation, so that the distributed storage system satisfies the write performance simultaneously under any one of the snapshot operations. , rollback performance and read performance are better.
  • the distributed storage system includes a plurality of data metadata servers and a plurality of data servers, each data server including a plurality of data blocks, and the data blocks are minimum storage units in the distributed storage system.
  • the snapshot metadata of the distributed storage system is stored on the metadata server, and the snapshot metadata information of the data block included in the distributed storage system is stored on the data server where the data block is located.
  • the current data and the snapshot data of the data block are stored in the same directory of the data server in the form of a file.
  • the data file of the snapshot file, the snapshot mode is COW
  • the data file when the snapshot mode is ROW have different naming formats.
  • the naming format of the snapshot file is the data block id+"_"+snap_id
  • the naming format of the data file when the snapshot mode is COW is the data block id+"_"+”head”
  • the naming of the data file when the snapshot mode is ROW The format is id+"_"+"head_id".
  • the data file can be directly generated, and the file name of the data file can be directly generated without searching or calculating the position of the data block;
  • the snapshot mode is ROW, the data file is operated, and the head_id is read from the snapshot metadata information, thereby generating The file name of the data file for IO operations.
  • the distributed block storage system sends a snapshot request to the data server where the corresponding data block is located, and determines the data according to the number of read operations of the data block, the number of read and write operations, and the front-end load value.
  • the target snapshot mode of the block set the current snapshot mode to the target snapshot mode, create a snapshot in the current snapshot mode, and update the snapshot metadata information of the data block.
  • the distributed block storage system sends the read request to the data server where the corresponding data block is located, and determines according to the number of read operations of the data block, the number of read and write operations, and the front-end load value.
  • the target snapshot mode of the data block set the current snapshot mode to the target snapshot mode, generate the file name of the data file corresponding to the target snapshot mode, and use the attribute of the snapshot sequence number of the snapshot metadata information to the corresponding relationship of the snapshot file, from the data block. Get the data file under the data directory.
  • the distributed block storage system sends a write request to the data server where the corresponding data block is located, and determines according to the number of read operations of the data block, the number of read and write operations, and the front-end load value.
  • the target snapshot mode of the data block set the current snapshot mode to the target snapshot mode, perform copy write or redirect write in the current snapshot mode, and update the snapshot metadata information of the data block.
  • the distributed block storage system sends a rollback request to the data server where the corresponding data block is located, and determines the data according to the number of read operations of the data block, the number of read and write operations, and the front-end load value.
  • the target snapshot mode of the block is set to the target snapshot mode.
  • the current snapshot mode is used to copy and roll back or directly modify the pointer, and update the snapshot metadata information of the data block.
  • the distributed block storage system sends the delete snapshot request to the data server where the corresponding data block is located, and determines the data according to the number of read operations of the data block, the number of read and write operations, and the front-end load value.
  • the target snapshot mode of the block set the current snapshot mode to the target snapshot mode, and determine whether the actual corresponding snapshot file needs to be deleted. If yes, delete the snapshot file and update the snapshot metadata information of the data block.
  • the present invention by determining the relationship between the ratio of the number of read operations of the data block and the number of read/write operations or the difference between the first preset threshold and the second preset threshold, if the ratio or difference is greater than or equal to the first pre-predetermined Set the threshold to use the COW as the snapshot mode for the snapshot operation of the data block. If the ratio or difference is less than or equal to the second preset threshold, use ROW as the snapshot mode for the snapshot operation of the data block; if the ratio or difference is greater than the first The preset threshold is smaller than the first preset threshold, and it is determined whether the front end load value of the data server storing the data block is greater than a preset load threshold.
  • the ROW is used as a snapshot operation on the data block. Snapshot mode; if the front-end load value is less than or equal to the preset load threshold, the COW is used as a snapshot mode for snapshot operations on the data block. In this way, in the snapshot of the distributed storage system, the write performance, the rollback performance, and the read performance can be satisfied at the same time.
  • the embodiment of the present invention provides a schematic structural diagram of a distributed block storage system, where the distributed block storage system 60 includes:
  • the data obtaining module 61 is configured to acquire the number of read operations and the number of read and write operations of the data block in the distributed block storage system.
  • the snapshot mode determining module 62 is configured to determine a target snapshot mode based on the number of read operations and the number of read and write operations, wherein the target snapshot mode includes copy-on-write or write-time redirection;
  • the processing module 63 is configured to perform a snapshot operation on the data block in the distributed block storage system based on the target snapshot mode.
  • the snapshot mode determining module 62 may include:
  • the first determining unit 621 is configured to determine whether the preset parameter value of the number of read operations and the number of read and write operations is greater than a first preset threshold, where the preset parameter value is the number of read operations and The ratio or difference of the number of read and write operations;
  • the first determining unit 622 is configured to determine that the target snapshot mode is copy-on-write if the number of read operations and the preset parameter value of the number of read and write operations are greater than a first preset threshold;
  • the second determining unit 623 is configured to determine that the target snapshot mode is a write-time redirection if the number of the read operation and the preset parameter value of the number of read/write operations are less than or equal to the first preset threshold.
  • the snapshot mode determining module 62 may further include:
  • the second determining unit 624 is configured to determine whether the preset parameter value of the number of read operations and the number of read and write operations is greater than a second preset threshold, wherein the second preset threshold is smaller than the first preset Threshold value
  • the third determining unit 625 is configured to determine whether the front end load value is greater than a preset load threshold if the preset number of read operations and the preset parameter value of the number of read and write operations are greater than a second preset threshold, wherein the front end load a value is a load value of a data server storing the data block;
  • the third determining unit 626 is configured to determine that the target snapshot mode is copy-on-write if the front-end load value is less than or equal to a preset load threshold;
  • the executing unit 627 is configured to perform the determining if the front-end load value is greater than a preset load threshold or the preset number of read operations and the preset parameter value of the number of read/write operations are less than or equal to a second preset threshold.
  • the target snapshot mode is the step of redirecting when writing.
  • the distributed block storage system 60 may further include:
  • the determining module 64 is configured to determine whether the current snapshot mode matches the target snapshot mode
  • the switching module 65 is configured to switch the current snapshot mode to the target snapshot mode if the current snapshot mode does not match the target snapshot mode.
  • the snapshot operation may include creating any one of a snapshot operation, a read operation, a write operation, a rollback operation, and a delete snapshot operation.
  • the distributed block storage system in this embodiment may be the distributed block storage system in the embodiment shown in FIG. 1 to FIG. 5, and the distributed block storage system introduced in the embodiment shown in FIG. 1 to FIG. Any of the embodiments may be implemented by the above distributed block storage system, and details are not described herein again.
  • the step of determining a target snapshot mode based on the number of read operations and the number of read and write operations includes:
  • the method before the step of determining that the target snapshot mode is a write-time redirection, the method further includes:
  • the preset parameter value of the number of read operations and the number of read and write operations is greater than a second preset threshold, determine whether the front-end load value is greater than a preset load threshold, where the front-end load value stores the data block.
  • the load value of the data server
  • the front end load value is greater than the preset load threshold or the read operation number and the preset parameter value of the number of read and write operations are less than or equal to the second preset threshold, performing the determining that the target snapshot mode is writing The steps to redirect.
  • the method before the step of performing a snapshot operation on the data block in the distributed block storage system, the method further includes:
  • the current snapshot mode is switched to the target snapshot mode.
  • the snapshot operation includes any one of a snapshot operation, a read operation, a write operation, a rollback operation, and a delete snapshot operation.
  • computer storage medium includes volatile and nonvolatile, implemented in any method or technology for storing information, such as computer readable instructions, data structures, program modules, or other data. , removable and non-removable media.
  • Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridge, magnetic tape, magnetic disk storage or other magnetic storage device, or may Any other medium used to store the desired information and that can be accessed by the computer.
  • communication media typically embodies computer readable instructions, data structures, program modules or other data in a modulated data signal, such as a carrier wave or other transport mechanism, and can include any information delivery media.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Quality & Reliability (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Retry When Errors Occur (AREA)

Abstract

一种快照的处理方法及分布式块存储系统，该方法可包括：获取所述分布式块存储系统中数据块的读操作次数和读写操作次数（S301）；基于所述读操作次数和读写操作次数，确定目标快照方式，其中，所述目标快照方式包括写时复制或写时重定向（S302）；基于所述目标快照方式对所述分布式块存储系统中数据块进行快照操作（S303）。

Description

一种快照的处理方法及分布式块存储系统 技术领域
本文涉及但不限于通信技术领域，特别涉及一种快照的处理方法及分布式块存储系统。
背景技术
分布式块存储系统具有可靠、高效、可扩展、接口灵活等特性，被广泛用于云存储、高性能计算、设备虚拟化等场景。典型的分布式存储系统通常由客户端、元数据服务器和数据服务器三个部分构成。快照技术为存储系统提供了在线备份、数据恢复、数据重用等能力，是分布式块存储系统不可缺少的一部分。
发明内容
以下是对本文详细描述的主题的概述。本概述并非是为了限制权利要求的保护范围。
本发明实施例提供一种快照的处理方法及分布式块存储系统。
本发明实施例提供一种快照的处理方法，包括：
获取所述分布式块存储系统中数据块的读操作次数和读写操作次数；
基于所述读操作次数和读写操作次数，确定目标快照方式，其中，所述目标快照方式包括写时复制或写时重定向；
基于所述目标快照方式对所述分布式块存储系统中数据块进行快照操作。
本发明实施例还提供一种分布式块存储系统，包括：
数据获取模块，设置为获取所述分布式块存储系统中数据块的读操作次数和读写操作次数；
快照方式确定模块，设置为基于所述读操作次数和读写操作次数，确定 目标快照方式，其中，所述目标快照方式包括写时复制或写时重定向；
处理模块，设置为基于所述目标快照方式对所述分布式块存储系统中数据块进行快照操作。
本发明实施例还提供一种计算机存储介质，所述计算机存储介质中存储有计算机可执行的一个或多个程序，所述一个或多个程序被所述计算机执行时使所述计算机执行如上述提供的一种快照的处理方法。
本发明实施例具有如下优点或有益效果：
获取所述分布式块存储系统中数据块的读操作次数和读写操作次数；基于所述读操作次数和读写操作次数，确定目标快照方式，其中，所述目标快照方式包括写时复制或写时重定向；基于所述目标快照方式对所述分布式块存储系统中数据块进行快照操作。这样在分布式存储系统的快照中，当数据块以读为主时，确定以写时复制对数据块进行快照操作；当数据块以写为主时，确定以写时重定向对数据块进行快照操作，从而可以同时满足写入性能、回滚性能以及读性能都较佳。
在阅读并理解了附图和详细描述后，可以明白其他方面。
附图概述
图1为写时复制快照示意图；
图2为写时重定快照示意图；
图3为本发明实施例提供的一种快照的处理方法的流程示意图；
图4为本发明实施例提供的另一种快照的处理方法的流程示意图；
图5为本发明实施例提供的分布式块存储系统的部分架构示意图；
图6为本发明实施例提供的一种分布式块存储系统的结构示意图；
图7为本发明实施例提供的一种分布式块存储系统中快照方式确定模块的结构示意图；
图8为本发明实施例提供的另一种分布式块存储系统中快照方式确定模块的结构示意图；
图9为本发明实施例提供的另一种分布式块存储系统的结构示意图。
本发明的实施方式
下面将结合附图及具体实施例进行详细描述。
如图1和图2所示，分布式块存储系统通常使用写时复制(Copy on Write，COW)和写时重定向(Redirect on Write，ROW)两种快照方式。其中，COW具有更好的读性能，而ROW具有更好的写性能和回滚性能。但是，由于COW进行写入和回滚操作时，需要先进行数据复制，因此写性能和回滚性能不佳；ROW在读操作时，由于需要对数据进行重定位导致读性能不佳。
如图3所示，本发明实施例提供一种快照的处理方法的流程示意图，该方法应用于分布式块存储系统，包括以下步骤：
步骤S301、获取分布式块存储系统中数据块的读操作次数和读写操作次数。
本发明实施例中，分布式存储系统可以通过磁盘操作监控工具获取每个数据块的读操作次数和读写操作次数，例如：在分布式存储系统输入iostat命令，实时统计存储该数据块的数据服务器的每秒进行读写(I/O)操作的次数(Input/Output Operations Per Second，IOPS)等负载数据，从而获取该数据块的读操作次数和读写操作次数。
步骤S302、基于读操作次数和读写操作次数，确定目标快照方式，其中，目标快照方式包括写时复制或写时重定向。
本发明实施例中，由于COW具有更好的读性能，而ROW具有更好的写性能和回滚性能，可以基于数据块的读操作次数和读写操作次数确定数据块的读写情况。若确定该数据块以读为主，可以将COW作为对数据块进行快照操作的快照方式；若确定该数据块以写为主，可以将ROW作为对数据块进行快照操作的快照方式。
其中，上述基于读操作次数和读写操作次数，确定目标快照方式，可以通过数据块的读操作次数与读写操作次数的比值或差值大小确定目标快照方式，比值或差值越大，表明该数据块以读为主；比值或差值越小，表明该数据块以写为主。例如：若读操作次数与读写操作次数的比值为10，该数 据块以读为主，可以确定目标快照方式为COW；若读操作次数与读写操作次数的比值为0.1，该数据块以写为主，可以确定目标操作方式为ROW。
步骤S303、基于目标快照方式对分布式块存储系统中数据块进行快照操作。
本发明实施例中，上述步骤S302确定目标快照方式时，分布式存储系统可以采用该目标快照方式对数据块进行快照操作。其中，上述快照操作可以包括创建快照操作、读操作、写操作、回滚操作以及删除快照操作等中的任意一种。
本发明实施例中，获取分布式块存储系统中数据块的读操作次数和读写操作次数；基于读操作次数和读写操作次数，确定目标快照方式，其中，目标快照方式包括写时复制或写时重定向；基于目标快照方式对分布式块存储系统中数据块进行快照操作。这样在分布式存储系统的快照中，当数据块以读为主时，确定以写时复制对数据块进行快照操作；当数据块以写为主时，确定以写时重定向对数据块进行快照操作，从而可以同时满足写入性能、回滚性能以及读性能都较佳。
如图4所示，本发明实施例提供另一种快照的处理方法的流程示意图，包括以下步骤：
步骤S401、获取分布式块存储系统中数据块的读操作次数和读写操作次数。
本发明实施例中，当客户端发起快照操作请求时，分布式块存储系统会响应该快照操作请求，通过元数据服务器中存储的快照元数据，查找该快照操作请求对应的数据块在数据服务器中的位置，并通过磁盘操作监控工具获取每个数据块的读操作次数和读写操作次数，例如：通过iostat命令统计存储该数据块的数据服务器的IOPS等负载数据，获取该数据块的读操作次数和读写操作次数。
步骤S402、判断读操作次数与读写操作次数的预设参数值是否大于或等于第一预设阈值，其中，预设参数值为读操作次数与读写操作次数的比值或差值。
本发明实施例中，将数据块的读操作次数与读写操作次数的比值或差值 作为预设参数值，可以直观反映数据块的读写情况，读操作次数与读写操作次数的比值或差值越大时，表明该数据块读操作较多；读操作次数与读写操作次数的比值或差值越大时，表明该数据块写操作较多。
其中，若读操作次数与读写操作次数的比值或差值大于或等于第一预设阈值，则可以确定该数据块以读为主。例如：当预设参数值为读操作次数与读写操作次数的比值时，可以设置第一预设阈值为1.5，若预设参数值大于或等于1.5，则可以确定该数据块以读为主；当预设参数值为读操作次数与读写操作次数的差值时，可以设置第一预设阈值为50，若预设参数值大于或等于50，则可以确定该数据块以读为主。
步骤S403、若读操作次数与读写操作次数的预设参数值大于或等于第一预设阈值，确定目标快照方式为写时复制。
本发明实施例中，当上述步骤S403确定读操作次数与读写操作次数的预设参数值大于或等于第一预设阈值，可以确定该数据块以读为主，从而可以确定目标快照方式为具有更好的读性能的写时复制，从而使分布式存储系统具有较佳的读性能。
步骤S404、若读操作次数与读写操作次数的预设参数值小于或等于第一预设阈值，确定目标快照方式为写时重定向。
本发明实施例中，当上述步骤S403确定读操作次数与读写操作次数的预设参数值小于第一预设阈值，可以确定该数据块不以读为主，该数据块可能读写接近或者以写为主，可以确定目标快照方式为具有更好的写性能和回滚性能的写时重定向，从而使分布式存储系统具有较佳的写性能和回滚性能。
可选的，上述确定目标快照方式为写时重定向之前还可以包括：
判断读操作次数与读写操作次数的预设参数值是否大于第二预设阈值，其中，第二预设阈值小于第一预设阈值；
若读操作次数与读写操作次数的预设参数值大于第二预设阈值，判断前端负载值是否大于预设负载阈值，其中，前端负载值为存储数据块的数据服务器的负载值；
若前端负载值小于或等于预设负载阈值，确定目标快照方式为写时复 制；
若前端负载值大于预设负载阈值或者读操作次数与读写操作次数的预设参数值小于或等于第二预设阈值，执行确定目标快照方式为写时重定向的步骤。
本发明实施例中，当数据块的读操作次数与读写操作次数的预设参数值小于第二预设阈值时，可以确定数据块以写为主，从而可以确定目标快照方式为具有较佳的写性能和回滚性能的写时重定向；而当数据块的读操作次数与读写操作次数的预设参数值小于第一预设阈值且大于第二预设阈值时，此时数据块的读写接近。
其中，为了保证分布式块存储系统在数据块的读写接近情况下仍同时满足写入性能、回滚性能以及读性能都较佳，可以通过获取存储该数据块的数据服务器的前端负载值，并将该前端负载值与预设负载阈值进行比较，若前端负载值大于预设负载阈值，表明该数据服务器负载高，可以确定目标快照方式为ROW，从而降低数据块的写请求的开销；若前端负载值小于或等于预设负载阈值，表明数据服务器负载低，可以确定目标快照方式为COW，从而提升数据块的读请求的性能。
例如：当预设参数值为读操作次数与读写操作次数的比值时，可以设定第一预设阈值为1.5，第二预设阈值为0.7，而预设负载阈值为10000，若读操作次数与读写操作次数的比值小于或等于0.7，表明数据块以写为主，确定目标快照方式为ROW；若读操作次数与读写操作次数的比值大于0.7且小于1.5，表明数据块的读写接近，此时若对应数据服务器的前端负载值为15000，则确定目标快照方式为ROW。
需要说明的是，上述第一预设阈值、第二预设阈值以及预设负载阈值可以根据实际需要进行设定，在此并不进行限定。
步骤S405、基于目标快照方式对分布式块存储系统中数据块进行快照操作。
本发明实施例中，当确定目标快照方式时，分布式存储系统可以采用该目标快照方式对数据块进行快照操作。
可选的，上述步骤S405之前，还可以包括：判断当前快照方式是否与 目标快照方式匹配；
若当前快照方式与目标快照方式不匹配，切换当前快照方式为目标快照方式。
本发明实施例中，当上述目标快照方式确定时，分布式存储系统可以将目标快照方式与数据块的当前快照方式进行匹配，若目标快照方式与当前快照方式不匹配，可以将当前快照方式切换为目标快照方式。例如：若当前的快照方式为COW，而目标快照方式为ROW，可以将该数据块的当前快照方式由COW切换为ROW；若当前的快照方式为ROW，而目标快照方式为COW，可以将该数据块的当前快照方式由ROW切换为COW。
可选的，上述快照操作包括创建快照操作、读操作、写操作、回滚操作以及删除快照操作中的任意一种，从而使分布式存储系统在任意一种快照操作下都同时满足写入性能、回滚性能以及读性能都较佳。
本发明实施例中，以上述快照的处理方法在读操作、写操作、回滚操作以及删除快照操作中的实际应用为例进行说明。其中，如图5所示，分布式存储系统包括多个数据元数据服务器以及多个数据服务器，每一数据服务器包括多个数据块，数据块为分布式存储系统中的最小存储单元。分布式存储系统的快照元数据存储于元数据服务器上，分布式存储系统所包含的数据块的快照元数据信息存储于该数据块所在的数据服务器上。
其中，数据块的当前数据和快照数据以文件的形式存放于数据服务器的同一目录下，快照文件、快照方式为COW时的数据文件以及快照方式为ROW时的数据文件均具有不同的命名格式，例如：快照文件的命名格式为数据块id+“_”+snap_id；快照方式为COW时的数据文件的命名格式为数据块id+“_”+“head”；快照方式为ROW时的数据文件的命名格式为id+“_”+“head_id”。并且快照方式为COW时操作数据文件，可以直接生成数据文件的文件名，无需经过查找或计算数据块的位置；快照方式为ROW时操作数据文件，从快照元数据信息中读取head_id，从而生成数据文件的文件名，进行IO操作。
在快照操作为创建快照操作时，分布式块存储系统将创建快照请求发送到对应的数据块所在的数据服务器上，根据数据块的读操作次数、读写操作 次数以及前端负载值，确定该数据块的目标快照方式，设置当前快照方式为目标快照方式，以当前快照方式创建快照，并更新数据块的快照元数据信息。
在快照操作为读取数据的读操作时，分布式块存储系统将读请求发送到对应的数据块所在的数据服务器上，根据数据块的读操作次数、读写操作次数以及前端负载值，确定该数据块的目标快照方式，设置当前快照方式为目标快照方式，生成目标快照方式对应的数据文件的文件名，利用快照元数据信息的快照序号到快照文件的对应关系的属性，从该数据块的数据目录下获取数据文件。
在快照操作为写入数据的写操作时，分布式块存储系统将写请求发送到对应的数据块所在的数据服务器上，根据数据块的读操作次数、读写操作次数以及前端负载值，确定该数据块的目标快照方式，设置当前快照方式为目标快照方式，以当前快照方式进行复制写或者重定向写，并更新数据块的快照元数据信息。
在快照操作为回滚操作时，分布式块存储系统将回滚请求发送到对应的数据块所在的数据服务器上，根据数据块的读操作次数、读写操作次数以及前端负载值，确定该数据块的目标快照方式，设置当前快照方式为目标快照方式，以当前快照方式进行复制回滚或者直接修改指针，并更新数据块的快照元数据信息。
在快照操作为删除快照操作时，分布式块存储系统将删除快照请求发送到对应的数据块所在的数据服务器上，根据数据块的读操作次数、读写操作次数以及前端负载值，确定该数据块的目标快照方式，设置当前快照方式为目标快照方式，判断是否需要删除实际对应的快照文件，若为是，删除快照文件，并更新数据块的快照元数据信息。
本发明实施例中，通过判断数据块的读操作次数和读写操作次数的比值或差值与第一预设阈值和第二预设阈值的关系，若比值或差值大于或等于第一预设阈值，将COW作为对数据块进行快照操作的快照方式；若比值或差值小于或等于第二预设阈值，将ROW作为对数据块进行快照操作的快照方式；若比值或差值大于第二预设阈值且小于第一预设阈值，判断存储数据块的数据服务器的前端负载值是否大于预设负载阈值，若前端负载值大于预设 负载阈值，将ROW作为对数据块进行快照操作的快照方式；若前端负载值小于或等于预设负载阈值，将COW作为对数据块进行快照操作的快照方式。这样，在分布式存储系统的快照中，可以同时满足写入性能、回滚性能以及读性能都较佳。
如图6所示，本发明实施例提供一种分布式块存储系统的结构示意图，所述分布式块存储系统60包括：
数据获取模块61，设置为获取所述分布式块存储系统中数据块的读操作次数和读写操作次数；
快照方式确定模块62，设置为基于所述读操作次数和读写操作次数，确定目标快照方式，其中，所述目标快照方式包括写时复制或写时重定向；
处理模块63，设置为基于所述目标快照方式对所述分布式块存储系统中数据块进行快照操作。
可选的，如图7所示，所述快照方式确定模块62可以包括：
第一判断单元621，设置为判断所述读操作次数与所述读写操作次数的预设参数值是否大于第一预设阈值，其中，所述预设参数值为所述读操作次数与所述读写操作次数的比值或差值；
第一确定单元622，设置为若所述读操作次数与所述读写操作次数的预设参数值大于第一预设阈值，确定所述目标快照方式为写时复制；
第二确定单元623，设置为若所述读操作次数与所述读写操作次数的预设参数值小于或等于第一预设阈值，确定所述目标快照方式为写时重定向。
可选的，如图8所示，所述快照方式确定模块62还可以包括：
第二判断单元624，设置为判断所述读操作次数与所述读写操作次数的预设参数值是否大于第二预设阈值，其中，所述第二预设阈值小于所述第一预设阈值；
第三判断单元625，设置为若所述读操作次数与所述读写操作次数的预设参数值大于第二预设阈值，判断前端负载值是否大于预设负载阈值，其中，所述前端负载值为存储所述数据块的数据服务器的负载值；
第三确定单元626，设置为若所述前端负载值小于或等于预设负载阈值，确定所述目标快照方式为写时复制；
执行单元627，设置为若所述前端负载值大于预设负载阈值或者所述读操作次数与所述读写操作次数的预设参数值小于或等于第二预设阈值，执行所述确定所述目标快照方式为写时重定向的步骤。
可选的，如图9所示，所述分布式块存储系统60还可以包括：
判断模块64，设置为判断当前快照方式是否与所述目标快照方式匹配；
切换模块65，设置为若所述当前快照方式与所述目标快照方式不匹配，切换所述当前快照方式为所述目标快照方式。
可选的，所述快照操作可以包括创建快照操作、读操作、写操作、回滚操作以及删除快照操作中的任意一种。
需要说明的是，本实施例的分布式块存储系统可以是图1至图5所示的实施例中的分布式块存储系统，图1至图5所示实施例介绍的分布式块存储系统的任意实施方式都可以被上述分布式块存储系统实现，此处不再赘述。
本领域普通技术人员可以理解实现上述实施例方法的全部或者部分步骤是可以通过程序指令相关的硬件来完成，所述的程序可以存储于一计算机可读取介质中，该程序在执行时，包括以下步骤：
获取所述分布式块存储系统中数据块的读操作次数和读写操作次数；
基于所述读操作次数和读写操作次数，确定目标快照方式，其中，所述目标快照方式包括写时复制或写时重定向；
基于所述目标快照方式对所述分布式块存储系统中数据块进行快照操作。
可选的，所述基于所述读操作次数和读写操作次数，确定目标快照方式的步骤，包括：
判断所述读操作次数与所述读写操作次数的预设参数值是否大于或等于第一预设阈值，其中，所述预设参数值为所述读操作次数与所述读写操作次数的比值或差值；
若所述读操作次数与所述读写操作次数的预设参数值大于或等于第一预设阈值，确定所述目标快照方式为写时复制；
若所述读操作次数与所述读写操作次数的预设参数值小于第一预设阈值，确定所述目标快照方式为写时重定向。
可选的，所述确定所述目标快照方式为写时重定向的步骤之前，还包括：
判断所述读操作次数与所述读写操作次数的预设参数值是否大于第二预设阈值，其中，所述第二预设阈值小于所述第一预设阈值；
若所述读操作次数与所述读写操作次数的预设参数值大于第二预设阈值，判断前端负载值是否大于预设负载阈值，其中，所述前端负载值为存储所述数据块的数据服务器的负载值；
若所述前端负载值小于或等于预设负载阈值，确定所述目标快照方式为写时复制；
若所述前端负载值大于预设负载阈值或者所述读操作次数与所述读写操作次数的预设参数值小于或等于第二预设阈值，执行所述确定所述目标快照方式为写时重定向的步骤。
可选的，所述基于所述目标快照方式对所述分布式块存储系统中数据块进行快照操作步骤之前，还包括：
判断当前快照方式是否与所述目标快照方式匹配；
若所述当前快照方式与所述目标快照方式不匹配，切换所述当前快照方式为所述目标快照方式。
可选的，所述快照操作包括创建快照操作、读操作、写操作、回滚操作以及删除快照操作中的任意一种。
本领域普通技术人员可以理解，上文中所公开方法中的全部或某些步骤、系统、装置中的功能模块/单元可以被实施为软件、固件、硬件及其适当的组合。在硬件实施方式中，在以上描述中提及的功能模块/单元之间的划分不一定对应于物理单元的划分；例如，一个物理组件可以具有多个功能，或者一个功能或步骤可以由若干物理组件合作执行。某些组件或所有组件可以被实施为由处理器，如数字信号处理器或微处理器执行的软件，或者被实施为硬件，或者被实施为集成电路，如专用集成电路。这样的软件可以分布在计算机可读介质上，计算机可读介质可以包括计算机存储介质(或非暂时性介质)和通信介质(或暂时性介质)。如本领域普通技术人员公知的，术 语计算机存储介质包括用于存储信息(诸如计算机可读指令、数据结构、程序模块或其他数据)的任何方法或技术中实施的易失性和非易失性、可移除和不可移除介质。计算机存储介质包括但不限于RAM、ROM、EEPROM、闪存或其他存储器技术、CD-ROM、数字多功能盘(DVD)或其他光盘存储、磁盒、磁带、磁盘存储或其他磁存储装置、或者可以用于存储期望的信息并且可以被计算机访问的任何其他的介质。此外，本领域技术人员公知的是，通信介质通常包含计算机可读指令、数据结构、程序模块或者诸如载波或其他传输机制之类的调制数据信号中的其他数据，并且可包括任何信息递送介质。
以上所述是本发明的可选实施方式，应当指出，对于本技术领域的普通技术人员来说，在不脱离本发明所述原理的前提下，还可以作出若干改进和润饰，这些改进和润饰也应视为本发明的保护范围。
工业实用性
本发明实施例可实现在分布式存储系统的快照中，当数据块以读为主时，确定以写时复制对数据块进行快照操作；当数据块以写为主时，确定以写时重定向对数据块进行快照操作，从而可以同时满足写入性能、回滚性能以及读性能都较佳。

Claims (11)

  1. 一种快照的处理方法，应用于分布式块存储系统，包括：
    获取所述分布式块存储系统中数据块的读操作次数和读写操作次数(S301)；
    基于所述读操作次数和读写操作次数，确定目标快照方式，其中，所述目标快照方式包括写时复制或写时重定向(S302)；
    基于所述目标快照方式对所述分布式块存储系统中数据块进行快照操作(S303)。
  2. 如权利要求1所述的方法，其中，所述基于所述读操作次数和读写操作次数，确定目标快照方式(S302)的步骤，包括：
    判断所述读操作次数与所述读写操作次数的预设参数值是否大于或等于第一预设阈值，其中，所述预设参数值为所述读操作次数与所述读写操作次数的比值或差值(S401、S402)；
    若所述读操作次数与所述读写操作次数的预设参数值大于或等于第一预设阈值，确定所述目标快照方式为写时复制(S403)；
    若所述读操作次数与所述读写操作次数的预设参数值小于第一预设阈值，确定所述目标快照方式为写时重定向(S404)。
  3. 如权利要求2所述的方法，还包括：
    所述确定所述目标快照方式为写时重定向(S404)的步骤之前，判断所述读操作次数与所述读写操作次数的预设参数值是否大于第二预设阈值，其中，所述第二预设阈值小于所述第一预设阈值；
    若所述读操作次数与所述读写操作次数的预设参数值大于第二预设阈值，判断前端负载值是否大于预设负载阈值，其中，所述前端负载值为存储所述数据块的数据服务器的负载值；
    若所述前端负载值小于或等于预设负载阈值，确定所述目标快照方式为写时复制；
    若所述前端负载值大于预设负载阈值或者所述读操作次数与所述读写操作次数的预设参数值小于或等于第二预设阈值，执行所述确定所述目标快 照方式为写时重定向的步骤。
  4. 如权利要求1至3中任意一项所述的方法，还包括：
    所述基于所述目标快照方式对所述分布式块存储系统中数据块进行快照操作(S303)步骤之前，判断当前快照方式是否与所述目标快照方式匹配；
    若所述当前快照方式与所述目标快照方式不匹配，切换所述当前快照方式为所述目标快照方式。
  5. 如权利要求1至3中任意一项所述的方法，其中，所述快照操作包括创建快照操作、读操作、写操作、回滚操作以及删除快照操作中的任意一种。
  6. 一种分布式块存储系统(60)，包括：
    数据获取模块(61)，设置为获取所述分布式块存储系统中数据块的读操作次数和读写操作次数；
    快照方式确定模块(62)，设置为基于所述读操作次数和读写操作次数，确定目标快照方式，其中，所述目标快照方式包括写时复制或写时重定向；
    处理模块(63)，设置为基于所述目标快照方式对所述分布式块存储系统中数据块进行快照操作。
  7. 如权利要求6所述的分布式块存储系统(60)，其中，所述快照方式确定模块(62)包括：
    第一判断单元(621)，设置为判断所述读操作次数与所述读写操作次数的预设参数值是否大于第一预设阈值，其中，所述预设参数值为所述读操作次数与所述读写操作次数的比值或差值；
    第一确定单元(622)，设置为若所述读操作次数与所述读写操作次数的预设参数值大于第一预设阈值，确定所述目标快照方式为写时复制；
    第二确定单元(623)，设置为若所述读操作次数与所述读写操作次数的预设参数值小于或等于第一预设阈值，确定所述目标快照方式为写时重定向。
  8. 如权利要求7所述的分布式块存储系统(60)，所述快照方式确定模块(62)还包括：
    第二判断单元(624)，设置为判断所述读操作次数与所述读写操作次 数的预设参数值是否大于第二预设阈值，其中，所述第二预设阈值小于所述第一预设阈值；
    第三判断单元(625)，设置为若所述读操作次数与所述读写操作次数的预设参数值大于第二预设阈值，判断前端负载值是否大于预设负载阈值，其中，所述前端负载值为存储所述数据块的数据服务器的负载值；
    第三确定单元(626)，设置为若所述前端负载值小于或等于预设负载阈值，确定所述目标快照方式为写时复制；
    执行单元(627)，设置为若所述前端负载值大于预设负载阈值或者所述读操作次数与所述读写操作次数的预设参数值小于或等于第二预设阈值，执行所述确定所述目标快照方式为写时重定向的步骤。
  9. 如权利要求6至8中任意一项所述的分布式块存储系统(60)，所述分布式块存储系统(60)还包括：
    判断模块(64)，设置为判断当前快照方式是否与所述目标快照方式匹配；
    切换模块(65)，设置为若所述当前快照方式与所述目标快照方式不匹配，切换所述当前快照方式为所述目标快照方式。
  10. 如权利要求6至8中任意一项所述的分布式块存储系统(60)，其中，所述快照操作包括创建快照操作、读操作、写操作、回滚操作以及删除快照操作中的任意一种。
  11. 一种计算机可读存储介质，存储有计算机可执行指令，所述计算机可执行指令被处理器执行时实现权利要求1至5中任一项所述的方法。
PCT/CN2018/075458 2017-02-21 2018-02-06 一种快照的处理方法及分布式块存储系统 WO2018153251A1 (zh)

Priority Applications (4)

Application Number Priority Date Filing Date Title
EP18757805.9A EP3588330A4 (en) 2017-02-21 2018-02-06 SNAPSHOT PROCESSING AND DISTRIBUTED BLOCK STORAGE SYSTEM
AU2018223401A AU2018223401A1 (en) 2017-02-21 2018-02-06 Method for processing snapshots and distributed block storage system
US16/487,820 US11455276B2 (en) 2017-02-21 2018-02-06 Method for processing snapshots and distributed block storage system
AU2021204579A AU2021204579A1 (en) 2017-02-21 2021-06-30 Method For Processing Snapshots And Distributed Block Storage System

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201710091836.0 2017-02-21
CN201710091836.0A CN108460045B (zh) 2017-02-21 2017-02-21 一种快照的处理方法及分布式块存储系统

Publications (1)

Publication Number Publication Date
WO2018153251A1 true WO2018153251A1 (zh) 2018-08-30

Family

ID=63229311

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/075458 WO2018153251A1 (zh) 2017-02-21 2018-02-06 一种快照的处理方法及分布式块存储系统

Country Status (5)

Country Link
US (1) US11455276B2 (zh)
EP (1) EP3588330A4 (zh)
CN (1) CN108460045B (zh)
AU (2) AU2018223401A1 (zh)
WO (1) WO2018153251A1 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111831472A (zh) * 2019-04-18 2020-10-27 阿里巴巴集团控股有限公司 快照创建方法和装置以及电子设备
CN115061856A (zh) * 2022-08-16 2022-09-16 四川大学 基于块的持久卷备份方法、系统、存储介质及其恢复方法
CN115865097A (zh) * 2023-02-17 2023-03-28 浪潮电子信息产业股份有限公司 一种数据压缩方法、系统、设备及计算机可读存储介质

Families Citing this family (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108460045B (zh) 2017-02-21 2022-11-18 中兴通讯股份有限公司 一种快照的处理方法及分布式块存储系统
CN110609807B (zh) * 2018-06-15 2023-06-23 伊姆西Ip控股有限责任公司 用于删除快照数据的方法、设备和计算机可读存储介质
CN110008175A (zh) * 2019-04-11 2019-07-12 苏州浪潮智能科技有限公司 一种分布式对象存储快照方法、系统、设备及计算机介质
CN110321321B (zh) * 2019-05-31 2023-06-23 平安科技（深圳）有限公司 网络块设备快照读写方法、装置、设备及存储介质
CN111158858B (zh) * 2019-12-26 2024-07-09 深信服科技股份有限公司 一种虚拟机的克隆方法和装置，及计算机可读存储介质
CN111966531B (zh) * 2020-07-21 2022-07-12 苏州浪潮智能科技有限公司 数据快照方法、装置、计算机设备及存储介质
CN113138722B (zh) * 2021-04-30 2024-01-12 北京百度网讯科技有限公司 用于分布式块存储系统的复制快照方法、系统和介质
CN115079952A (zh) * 2022-06-29 2022-09-20 阿里巴巴（中国）有限公司 实现写操作互斥的方法及装置
CN115470178A (zh) * 2022-10-11 2022-12-13 中电云数智科技有限公司 一种基于域对象的域快照回滚方法
CN116204124B (zh) * 2023-02-23 2024-03-22 安超云软件有限公司 一种基于冲突锁的数据处理方法、系统及电子设备
CN118069432B (zh) * 2024-04-19 2024-07-05 济南浪潮数据技术有限公司 分布式块存储卷快照回滚方法、装置、设备和存储介质

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060206677A1 (en) * 2003-07-03 2006-09-14 Electronics And Telecommunications Research Institute System and method of an efficient snapshot for shared large storage
CN101183383A (zh) * 2007-12-17 2008-05-21 中国科学院计算技术研究所 一种快照系统及其使用方法
CN101482838A (zh) * 2009-02-06 2009-07-15 杭州华三通信技术有限公司 基于写时拷贝的数据写入方法及设备
CN102971698A (zh) * 2012-06-29 2013-03-13 华为技术有限公司 快照数据处理方法及系统、存储系统、快照代理

Family Cites Families (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7676510B1 (en) * 2006-12-22 2010-03-09 Network Appliance, Inc. Space reservation monitoring in a fractionally reserved data storage system
US8650145B2 (en) * 2008-10-07 2014-02-11 Hewlett-Packard Development Company, L.P. Creating snapshots of data using a selected one of different snapshot algorithms
US8229894B2 (en) 2009-02-23 2012-07-24 Hitachi Ltd. Storage system and control method for the same
US8239584B1 (en) * 2010-12-16 2012-08-07 Emc Corporation Techniques for automated storage management
CN102073739A (zh) * 2011-01-25 2011-05-25 中国科学院计算技术研究所 带有快照功能的分布式文件系统中的数据读与数据写方法
EP2821925B1 (en) * 2012-08-09 2017-05-10 Huawei Technologies Co., Ltd. Distributed data processing method and apparatus
US9235535B1 (en) 2012-10-31 2016-01-12 Emc Corporation Method and apparatus for reducing overheads of primary storage by transferring modified data in an out-of-order manner
GB2509057A (en) 2012-12-18 2014-06-25 Ibm Predictive point-in-time copy for storage systems
CN108460045B (zh) 2017-02-21 2022-11-18 中兴通讯股份有限公司 一种快照的处理方法及分布式块存储系统

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060206677A1 (en) * 2003-07-03 2006-09-14 Electronics And Telecommunications Research Institute System and method of an efficient snapshot for shared large storage
CN101183383A (zh) * 2007-12-17 2008-05-21 中国科学院计算技术研究所 一种快照系统及其使用方法
CN101482838A (zh) * 2009-02-06 2009-07-15 杭州华三通信技术有限公司 基于写时拷贝的数据写入方法及设备
CN102971698A (zh) * 2012-06-29 2013-03-13 华为技术有限公司 快照数据处理方法及系统、存储系统、快照代理

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3588330A4 *

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111831472A (zh) * 2019-04-18 2020-10-27 阿里巴巴集团控股有限公司 快照创建方法和装置以及电子设备
CN111831472B (zh) * 2019-04-18 2023-12-22 阿里云计算有限公司 快照创建方法和装置以及电子设备
CN115061856A (zh) * 2022-08-16 2022-09-16 四川大学 基于块的持久卷备份方法、系统、存储介质及其恢复方法
CN115865097A (zh) * 2023-02-17 2023-03-28 浪潮电子信息产业股份有限公司 一种数据压缩方法、系统、设备及计算机可读存储介质
CN115865097B (zh) * 2023-02-17 2023-05-23 浪潮电子信息产业股份有限公司 一种数据压缩方法、系统、设备及计算机可读存储介质

Also Published As

Publication number Publication date
AU2021204579A1 (en) 2021-07-29
US20210133146A1 (en) 2021-05-06
AU2018223401A1 (en) 2019-10-10
EP3588330A4 (en) 2020-12-02
CN108460045A (zh) 2018-08-28
US11455276B2 (en) 2022-09-27
EP3588330A1 (en) 2020-01-01
CN108460045B (zh) 2022-11-18

Similar Documents

Publication Publication Date Title
WO2018153251A1 (zh) 一种快照的处理方法及分布式块存储系统
US9411821B1 (en) Block-based backups for sub-file modifications
US9348827B1 (en) File-based snapshots for block-based backups
US9703803B2 (en) Replica identification and collision avoidance in file system replication
US9996421B2 (en) Data storage method, data storage apparatus, and storage device
US20130325803A1 (en) User initiated replication in a synchronized object replication system
US11630741B2 (en) System and method for backing up data in a load-balanced clustered environment
US20120110287A1 (en) Snapshots in a hybrid storage device comprising a magnetic disk and a solid state disk
US10114551B2 (en) Space reclamation in asynchronously mirrored space-efficient secondary volumes
US10545838B2 (en) Data recovery in a multi-pipeline data forwarder
US20120016842A1 (en) Data processing apparatus, data processing method, data processing program, and storage apparatus
US20160062671A1 (en) Restoring data
US10146683B2 (en) Space reclamation in space-efficient secondary volumes
US11899542B2 (en) File data access method, apparatus, and computer-readable storage medium
US9395923B1 (en) Method and system for recovering from embedded errors from writing data to streaming media
US20150302021A1 (en) Storage system
JP2006268139A (ja) データ複製装置、方法及びプログラム並びに記憶システム
US9886211B2 (en) Selecting a virtual tape server in a storage system to provide data copy while minimizing system job load
US9575679B2 (en) Storage system in which connected data is divided
US10423494B2 (en) Trimming unused blocks from a versioned image backup of a source storage that is stored in a sparse storage
US11256716B2 (en) Verifying mirroring of source data units to target data units
US11188248B2 (en) System and method to achieve an uninterrupted file level backup using a pass-through snapshot engine
US11755425B1 (en) Methods and systems for synchronous distributed data backup and metadata aggregation
US10394483B2 (en) Target volume shadow copy
US20210064245A1 (en) Virtual tape backspace file performance enhancement

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 18757805

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2018757805

Country of ref document: EP

Effective date: 20190923

ENP Entry into the national phase

Ref document number: 2018223401

Country of ref document: AU

Date of ref document: 20180206

Kind code of ref document: A

点击 这是indexloc提供的php浏览器服务，不要输入任何密码和下载