CN100359476C - Snapshot backup method - Google Patents
Snapshot backup method Download PDFInfo
- Publication number
- CN100359476C CN100359476C CNB200410037139XA CN200410037139A CN100359476C CN 100359476 C CN100359476 C CN 100359476C CN B200410037139X A CNB200410037139X A CN B200410037139XA CN 200410037139 A CN200410037139 A CN 200410037139A CN 100359476 C CN100359476 C CN 100359476C
- Authority
- CN
- China
- Prior art keywords
- snapshot
- write
- read
- volume
- source data
- Prior art date
- Legal status (The legal status 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 status listed.)
- Expired - Fee Related
Links
Images
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention relates to the technical field of data backup, particularly to a method for realizing the data backup of continuous read-write access by a host machine by a snapshot technique. The method comprises the steps: a snapshot initializing step, namely obtaining the information of a source film and a target film, establishing a structure variable, setting a snapshot originating mark, establishing a bitmap and carrying out initialization operation; a read-write processing function registering step in which a background copying progress is created; a read-write request processing step, namely receiving read-write requests and correspondingly processing the requests according to the types of the read-write requests; a background copying step, namely scanning the bitmap, copying a corresponding area according to priority until all the data areas of the source film are copied to the target film, setting the end mark of the snapshot and updating the bitmap mark. The method achieves the goal of no block in the read-write requests of the host machine during the snapshot, effectively reduces the influences of the snapshot on read-write performance, and can effectively enhance the response speed of the host machine during the snapshot.
Description
Technical field
The present invention relates to the data backup technical field, relate to a kind of data back up method of realizing the uninterrupted read and write access of main frame by snapping technique specifically.
Background technology
Along with the develop rapidly of science and technology and the widespread usage of computer technology, the increasing sharply and Development of E-business of internet information, realizes that the storage backup of data occupies important position in enterprise application system at protected data safety.Yet backup procedure requires not ISR business, requires to reduce as far as possible the influence to the data access performance.Therefore, the storage array snapping technique arises at the historic moment as a kind of new technology, has solved this problem to a great extent.
Snapping technique is to scan fast by the data of software to the disk subsystem that will back up, thereby sets up instant data mapping.It is existing realizes that by snapshot mode of backup has two kinds: before writing and to tear mirror image open right.Wherein, writing previous mode just is meant in the snapshot processes, upgrade the write request of source data volume for main frame, the snapshot program is not arrived auxilliary volume as yet if find the counterpart of this write request on the source data volume, then hang host requests earlier, the source data volume is gone up corresponding legacy data after the auxiliary volume of snapshot is gone up, just respond.It is smaller to mode that this mode is torn mirror image open to the influence of system performance.
At present, at writing previous mode relevant patent is arranged, it mainly is in snapshot processes elder generation each magnetic track of scan source book successively, the source data volume magnetic track that was upgraded by main frame is not rolled up to auxilliary, then successively the legacy data on the auxiliary volume of snapshot (be positioned at same data-storage system with master file, store the more legacy data of new portion of main frame during the snapshot temporarily) to auxilliary volume.
During this patent snapshot, main frame hangs host requests earlier during for the renewal write request of source data volume, and legacy data is just responded after the auxiliary volume of snapshot is gone up.Obviously,, then still to hang host requests earlier if there is not legacy data not rolled up as yet to snapshot is auxiliary, like this can be bigger to the write performance influence of main frame.
Another relevant patent, it mainly is during the host snapshot during read data, directly reads from the auxiliary volume of snapshot, if read less than then reading from the source data volume again.Obviously, this patent is also bigger to the performance impact of main frame.
Summary of the invention
The present invention proposes a kind of method of snapshot, all the readwrite performance of main frame is had the problem of considerable influence to solve existing snapping technique.
For this reason, the invention provides following technical scheme:
A kind of method of snapshot, wherein, described method comprises following steps:
Buffer memory receives snapshot request, beginning snapshot initialization thread;
Buffer memory involves in capable write-back mark to the snapshot source data;
Finish described write-back mark, buffer memory begins write-back, finishes behind the described write-back notice snapshot program updating and finishes;
Obtain the information of source data volume and target volume, according to the source data volume structure, set up the snapshot bitmap of source data volume logic region, the initial snapshot bitmap starts the background copy process between source data volume and the target volume;
Buffer memory receives the also read-write requests of process source book;
Receive the snapshot program updating finish notice back according to priority with the copying data of source data volume to target volume.
The present invention relates to a kind of method of snapshot, it mainly is when providing online snapshot for storage system, as much as possible to improve the response to the main frame read-write requests.The present invention optimizes duplication technology before writing, and cache write mechanism is combined with snapping technique.Buffer memory receives the main frame write request at every turn, and just the relevant position with write request corresponding data zone in the snapshot bitmap is labeled as 01.Snapshot process at first scans the snapshot bitmap, preferentially the snapshot bitmap is labeled as 01 data area and makees the snapshot copy, be not labeled as under 01 the situation, and duplicate copy is labeled as 00 data area.Before this snapshot place volume of the each write-back of buffer memory, search the snapshot bitmap earlier, only will be labeled as 10 data area write-back.Under normal conditions, this technology has realized the clog-free of main frame read-write requests during the snapshot, effectively reduces the influence of snapshot to readwrite performance, and then can effectively improve the response of host speed during the snapshot.
Description of drawings
Fig. 1 is the process flow diagram of the method for the invention;
Fig. 2 is the snapshot module overall construction drawing;
Fig. 3 is the snapshot initialization flowchart;
Fig. 4 is that snapshot is write processing flow chart;
Fig. 5 is a snapshot background copy process flow diagram.
Specific implementation
For clear more method of the present invention, and the overall operation mode, conjunction with figs. and specific embodiment are described in detail as follows now:
Snapshot module is between CACHE module (cache memory) and RAID module (raid-array; RAID:Redundant Arrays of Inexpensive Disk) between, it includes three sub-function module: i.e. initialization module, read-write processing module and background copy module.Wherein, the source data volume that initialization module receives and initialization CACHE transmits and the information of target volume, and the processing function is write in registration; The read-write processing module is used to handle the read-write requests to source data volume and snapped volume; The background copy module is used for copying data to target volume from the source data volume.
As shown in Figure 2, snapshot module reaches the read-write requests information to source data volume and snapped volume from CACHE module reception sources book and snapped volume information.The read-write processing module of snapshot module and background copy thread read the data of source data volume from the RAID module, write then in the target volume, finish data are rolled onto target volume from source data copy.Snapshot module also needs the snapped volume attribute is provided with, and information such as snapshot copy cell ZONE (data area, snapshot copy cell) size are inquired about and be provided with to query logic unit number (LUN:Logic Unit Number) operating position, LUN size.
Snapshot module is safeguarded a snapshot bitmap to each snapshot LUN, and each snapshot bitmap adopts 2bit (position) to identify, in order to write down the state of current snapshot copy cell.Snapshot background copy thread preferentially with CACHE might write-back data area (the snapshot bitmap is designated 01) copy the snapshot target volume to.Before the RAID group at this snapshot place of the each write-back of CACHE, search the snapshot bitmap earlier, the snapshot bitmap is masked as 10 data area write-back.
Set sign for bitmap specifies as shown in table 1:
Table 1
| Sign | Explanation |
| 00 | Be used to identify the source data volume data area that copies, the priority of its snapshot background copy is low; |
| 01 | Being used to identify CACHE receives after the main frame write request position, snapshot bitmap respective data area position, this data area does not still have the requirement that is written back to disk, the priority height of its snapshot background copy; |
| 10 | Be used to identify and finished the source data volume data area that copies target volume to. |
Describe the function of each submodule of snapshot module in detail below in conjunction with Fig. 2.
1, initialization module: judge command type,, then read control information data and registration read and create the I/O thread from the system backup device if restart behind the accident recovery; If normally begin, then set up control information structure, registration read and create the background copy thread according to Data transmission.
2, read-write processing module: have only after initialization module has been registered read-write processing function, this module begins to receive read-write requests.Specifically:
Module for reading and writing/read is handled: the read request that issues, directly call and read the Processing Interface function, by handle function according to the data-moving situation with goal orientation to correct target volume: read the source data volume, be redirected to source data and roll up; Read snapped volume, then judge earlier the snapshot bitmap,, otherwise be redirected to the source data volume, and the result that will be redirected returns to caller as the output of interface function if being designated 10 is redirected to snapped volume.
Module for reading and writing/write processing: when write request is arranged, call and write the Processing Interface function.If there is not the read-write processing threads in the system, then create the read-write processing threads by interface function; If there is the read-write processing threads in the system, then will read and write processing threads and be made as active state, write request is put into write message queue.The read-write processing threads is read a write order at every turn from write message queue, judge whether to carry out the preceding copy work of writing according to the snapshot bitmap.
3, background copy module: after initialization module created the background copy thread, this module was just started working.The corresponding background copy thread of each snapshot, this thread is created by initialization module.The background copy thread needs source data is rolled up copying data to snapped volume, the priority level according to snapshot bitmap sign during copy copies, and promptly copy snapshot bitmap is 01 earlier, and then to copy the snapshot bitmap be 00 data area.
Describe the detailed process of snapshot flow process in detail below in conjunction with Fig. 1 and 3:
1, the processing procedure of CACHE during the snapshot initialization:
(1) CACHE receives snapshot request, forbids the processing host write request, and starts the snapshot initialization thread;
(2) the CACHE volume that begins whole needs are done snapshot is done the write-back mark, allows the processing host write request when write-back mark is finished;
(3) CACHE begins write-back, and after write-back was finished, CACHE notice snapshot program updating was finished.
2, normal snapshot initialization thread:
(1) obtain the information of source data volume and target volume, carry out validity checking, it is a read states that target volume is set;
(2) information with source data volume and target volume writes snapshot control information structure variable, and the opening flag of snapshot is set;
(3), set up the snapshot bitmap of source data volume logic region, the initial snapshot bitmap according to the source data volume structure;
(4) receive CACHE refresh finish notice after, to CACHE module location registration process function, and start background copy process between source data volume and the target volume.
3, read-write requests processing threads:
After the processing function is finished registration, accept the read-write requests that the CACHE module transmits, do respective handling according to different read-write requests types, specific as follows:
(1) read operation:
The judgement read operation is read the source data volume or is read the snapshot dish, and its operation is as follows respectively:
Read the operation of source data volume: read the data on the source data volume;
Read snapshot operation: search the snapshot bitmap of data area correspondence in the read operation earlier, if 00 or 01 data that directly read on the source data volume; If 10 data that read on the target volume.
(2) write operation:
For write operation, adopt and write preceding copy method.At first search the snapshot bitmap of all snapshots of this data area correspondence, if corresponding snapshot bitmap is masked as 00 and 01, copying data in then earlier source data being rolled up in the data area of correspondence is in target volume, and will to be masked as 00 and 01 snapshot bitmap set be 10, and the data that will write write back in the source data volume then.If all the set of snapshot bitmap is masked as 10, then data are directly write in the source data volume.
4, background copy thread:
(1) process at first scans the snapshot bitmap, and the corresponding bitmap of preferential copy is masked as 01 data area, be not labeled as under 01 the situation, and the copy marker bit is 00 data area.Copy finishes and revises snapshot bitmap mark.When the snapshot bitmap all is labeled as 10, the end mark of this snapshot is set, and the resource of notifying CACHE to discharge this snapshot, this time snapshot is finished.Then, judge whether to also have ongoing snapshot, if all snapshot is all finished, notice CACHE nullifies snapshot and handles function;
(2) copy function is carried out in this data area, the copying data during source data is rolled up is in one or more target volume.For the situation that a source data volume is copied to a plurality of target volume, at first duplication source data is rolled onto a target volume, and from then on target volume copies other target volume to again.
In addition, need abandon the snapshot flow process for abnormal case, situation is as follows:
(1) abandons all snapshots
When the busy needs of storage system are abandoned all snapshots, receive the order of abandoning snapshot after, at first nullify snapshot read-write and handle function, check then whether the snapshot read-write requests of carrying out or hanging is arranged, if have then after waiting for that it finishes dealing with, stop the snapshot thread, discharge resource.
(2) abandon single snapshot
When the busy needs of storage system are abandoned certain snapshot, after receiving the order of abandoning snapshot, at first the status indication of this snapshot is set to abandon, no longer handle read request to this snapshot, also no longer inquire about the snapshot bitmap of this snapshot for the write request of source data volume, check then whether this snapshot has the snapshot read-write requests of carrying out or hanging, if having then after waiting for that it finishes dealing with, stop this snapshot thread, discharge resource.
The method of a kind of snapshot of the present invention, it mainly is when providing online snapshot for storage system, as much as possible to improve the response to the main frame read-write requests.The present invention optimizes copy duplication technology before writing, and CACIIE is write mechanism combine with snapping technique.CACHE receives the main frame write request at every turn, and just the relevant position with write request corresponding data zone in the snapshot bitmap is 01.Snapshot process at first scans the snapshot bitmap, preferentially the snapshot bitmap is labeled as 01 data area and makees the snapshot copy, be not labeled as under 01 the situation, and the duplicate copy marker bit is 00 data area.Before this snapshot place volume of the each write-back of CACHE, search the snapshot bitmap earlier, only will be labeled as 10 data area write-back.Under normal conditions, this technology has realized the clog-free of main frame read-write requests during the snapshot, effectively reduces the influence of snapshot to readwrite performance, and then can effectively improve the response of host speed during the snapshot.
Claims (8)
1, a kind of method of snapshot is characterized in that, described method comprises following steps:
Buffer memory receives snapshot request, beginning snapshot initialization thread;
Buffer memory involves in capable write-back mark to the snapshot source data;
Finish described write-back tag cache and begin write-back, finish behind the described write-back notice snapshot program updating and finish;
Obtain the information of source data volume and target volume, according to the source data volume structure, set up the snapshot bitmap of source data volume logic region, the initial snapshot bitmap starts the background copy process between source data volume and the target volume;
Buffer memory receives the also read-write requests of process source book;
Receive the snapshot program updating finish notice back according to priority with the copying data of source data volume to target volume.
2, the method for a kind of snapshot as claimed in claim 1, it is characterized in that, the described information step of obtaining source data volume and target volume further comprises: after obtaining the information of source data volume and target volume, carry out validity checking, and target volume is set is a read states.
3. the method for a kind of snapshot as claimed in claim 1 is characterized in that, described processing read request is specially:
If read the source data volume, directly read the data on the source data volume;
If read snapped volume, scanning snapshot bitmap judges whether to copy to target volume earlier, if copy on the target volume, then reads the data on the target volume, otherwise directly reads the data on the source data volume.
4. the method for a kind of snapshot as claimed in claim 1 is characterized in that, described processing write requests is specially:
Scanning snapshot bitmap has all copied target volume to if the snapshot bitmap is designated its source data volume data area, then data is directly write in the source data volume; Otherwise the copying data in now source data being rolled up in the corresponding region is in target volume, and the correspondence sign of renewal snapshot bitmap, and the data that will write write back in the source data volume then.
5. the method for a kind of snapshot as claimed in claim 1, it is characterized in that, described buffer memory receives and the read-write requests of process source book further is: after the write request of process source book, be the data area of preferential copy targeting volume with this zone marker of writing.
6, the method for a kind of snapshot as claimed in claim 5 is characterized in that: according to priority source data is rolled up all copying datas and be specially to target volume:
Sweep bitmap determines that according to bitmap source data rolls up the priority of each data area;
According to described priority the copying data of source data being rolled up each data area is arrived corresponding target volume;
Updated space figure mark discharges Snapshot Resources.
7, the method for a kind of snapshot as claimed in claim 1 is characterized in that, described method also further comprises:
The order of all snapshots is abandoned in reception;
Nullify the snapshot read-write and handle function;
Whether have the snapshot read-write requests of carrying out or hanging, if having then after waiting for that it finishes dealing with, otherwise stop the snapshot thread, the release resource if checking.
8, the method for a kind of snapshot as claimed in claim 1 is characterized in that, described method also further comprises:
The order of certain snapshot is abandoned in reception, and the status indication of this snapshot is set to abandon, and stops to handle the read request to this snapshot, also stops to inquire about the snapshot bitmap of this snapshot for the write request of source data volume;
Check whether this snapshot has the snapshot read-write requests of carrying out or hanging,, discharge resource if having then after waiting for that it finishes dealing with, stop this snapshot thread.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB200410037139XA CN100359476C (en) | 2004-06-03 | 2004-06-03 | Snapshot backup method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB200410037139XA CN100359476C (en) | 2004-06-03 | 2004-06-03 | Snapshot backup method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1704903A CN1704903A (en) | 2005-12-07 |
| CN100359476C true CN100359476C (en) | 2008-01-02 |
Family
ID=35577205
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB200410037139XA Expired - Fee Related CN100359476C (en) | 2004-06-03 | 2004-06-03 | Snapshot backup method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100359476C (en) |
Families Citing this family (28)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100464307C (en) * | 2006-05-26 | 2009-02-25 | 任永坚 | Method and system for accomplishing data backup and recovery |
| CN100462977C (en) * | 2007-03-26 | 2009-02-18 | 腾讯科技（深圳）有限公司 | Capture server, distribution server, method and system for generating webpage capture |
| CN100456255C (en) * | 2007-04-29 | 2009-01-28 | 华为技术有限公司 | Device and method for fetching snapshot data |
| CN101458655B (en) * | 2007-12-12 | 2011-01-05 | 成都市华为赛门铁克科技有限公司 | Internal memory management method, system and internal memory control device |
| CN101777018B (en) * | 2010-02-08 | 2012-04-25 | 北京同有飞骥科技股份有限公司 | Copying and snapshot combined Windows system protection method |
| US9047233B2 (en) | 2012-06-25 | 2015-06-02 | International Business Machines Corporation | Source cleaning cascaded volumes using write and background copy indicators |
| US9135121B2 (en) * | 2012-10-16 | 2015-09-15 | International Business Machines Corporation | Managing updates and copying data in a point-in-time copy relationship expressed as source logical addresses and target logical addresses |
| US9116852B2 (en) | 2012-10-16 | 2015-08-25 | International Business Machines Corporation | Processing a copy command directed to a first storage architecture for data that is stored in a second storage architecture |
| US9405704B2 (en) | 2012-10-16 | 2016-08-02 | International Business Machines Corporation | Establishing a point-in-time copy relationship between source logical addresses and target logical addresses |
| WO2015010327A1 (en) | 2013-07-26 | 2015-01-29 | 华为技术有限公司 | Data sending method, data receiving method and storage device |
| CN104699574B (en) * | 2013-12-09 | 2018-04-20 | 华为技术有限公司 | A kind of method, apparatus and system for establishing processor Cache checkpoints |
| CN106537353B (en) | 2014-07-15 | 2019-10-18 | 华为技术有限公司 | A kind of data copy method, device and system |
| CN104298609B (en) * | 2014-08-14 | 2017-11-10 | 浪潮(北京)电子信息产业有限公司 | Data copying method and device |
| CN105469001B (en) * | 2014-08-15 | 2020-02-11 | 中兴通讯股份有限公司 | Disk data protection method and device |
| CN105487810B (en) * | 2014-09-18 | 2018-04-03 | 先智云端数据股份有限公司 | For reaching the system of non interference data reconstruction |
| CN104462290B (en) * | 2014-11-27 | 2017-10-10 | 华为技术有限公司 | File system clone method and device |
| CN105045542B (en) * | 2015-09-11 | 2018-08-03 | 浪潮(北京)电子信息产业有限公司 | A kind of method and device for realizing snapshot management |
| US9632724B1 (en) * | 2015-10-12 | 2017-04-25 | International Business Machines Corporation | Point-in-time copy with chain cloning |
| CN105160045B (en) * | 2015-10-29 | 2019-06-18 | 小米科技有限责任公司 | Data copying method and device |
| CN106055431B (en) * | 2016-05-30 | 2018-09-21 | 河海大学 | The method and device of backup and recovery based on the online rollback of snapshot |
| CN106790378A (en) * | 2016-11-11 | 2017-05-31 | 北京奇虎科技有限公司 | The full synchronous method of data of equipment room, apparatus and system |
| US10261719B2 (en) * | 2017-01-31 | 2019-04-16 | Hewlett Packard Enterprise Development Lp | Volume and snapshot replication |
| CN107391312B (en) * | 2017-07-31 | 2020-11-24 | 苏州浪潮智能科技有限公司 | Data cloud backup method and system |
| CN109542674A (en) * | 2017-09-21 | 2019-03-29 | 阿里巴巴集团控股有限公司 | Snapshot creation method, device, equipment and the medium of distributed system |
| CN108762988B (en) * | 2018-05-31 | 2022-04-22 | 郑州云海信息技术有限公司 | Data processing method and related equipment |
| US10705927B2 (en) * | 2018-07-19 | 2020-07-07 | Hewlett Packard Enterprise Development Lp | Freeze a volume of a replication set and unfreeze the volume based on at least one of a snapshot permit message, a snapshot abort message, and expiration of a timeout |
| CN110442476A (en) * | 2019-06-25 | 2019-11-12 | 平安科技（深圳）有限公司 | Data snapshot method, device, equipment and storage medium |
| CN113868023B (en) * | 2021-09-24 | 2023-11-14 | 苏州浪潮智能科技有限公司 | Snapshot method and device of storage system, electronic equipment and readable storage medium |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6473775B1 (en) * | 2000-02-16 | 2002-10-29 | Microsoft Corporation | System and method for growing differential file on a base volume of a snapshot |
| US20030195903A1 (en) * | 2002-03-19 | 2003-10-16 | Manley Stephen L. | System and method for asynchronous mirroring of snapshots at a destination using a purgatory directory and inode mapping |
| CN1452737A (en) * | 2000-04-24 | 2003-10-29 | 微软公司 | Method and apparatus for providing volume snapshot dependencies in computer system |
| CN1452738A (en) * | 2000-04-24 | 2003-10-29 | 微软公司 | Method and system for providing common coordination and administration of multiple snapshot providers |
| US6647473B1 (en) * | 2000-02-16 | 2003-11-11 | Microsoft Corporation | Kernel-based crash-consistency coordinator |
| US20030212752A1 (en) * | 2002-05-08 | 2003-11-13 | Thunquest Gary Lee | Mehthod and apparatus for supporting snapshots with direct I/O in a storage area network |
| US6651075B1 (en) * | 2000-02-16 | 2003-11-18 | Microsoft Corporation | Support for multiple temporal snapshots of same volume |
| CN1481534A (en) * | 2000-12-21 | 2004-03-10 | ���ܿ���ϵͳ����˾ | Restoration of data between primary and backup systems |
-
2004
- 2004-06-03 CN CNB200410037139XA patent/CN100359476C/en not_active Expired - Fee Related
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6473775B1 (en) * | 2000-02-16 | 2002-10-29 | Microsoft Corporation | System and method for growing differential file on a base volume of a snapshot |
| US6647473B1 (en) * | 2000-02-16 | 2003-11-11 | Microsoft Corporation | Kernel-based crash-consistency coordinator |
| US6651075B1 (en) * | 2000-02-16 | 2003-11-18 | Microsoft Corporation | Support for multiple temporal snapshots of same volume |
| CN1452737A (en) * | 2000-04-24 | 2003-10-29 | 微软公司 | Method and apparatus for providing volume snapshot dependencies in computer system |
| CN1452738A (en) * | 2000-04-24 | 2003-10-29 | 微软公司 | Method and system for providing common coordination and administration of multiple snapshot providers |
| CN1481534A (en) * | 2000-12-21 | 2004-03-10 | ���ܿ���ϵͳ����˾ | Restoration of data between primary and backup systems |
| US20030195903A1 (en) * | 2002-03-19 | 2003-10-16 | Manley Stephen L. | System and method for asynchronous mirroring of snapshots at a destination using a purgatory directory and inode mapping |
| US20030212752A1 (en) * | 2002-05-08 | 2003-11-13 | Thunquest Gary Lee | Mehthod and apparatus for supporting snapshots with direct I/O in a storage area network |
Non-Patent Citations (6)
| Title |
|---|
| 一个有效的重复快照算法. 贺乐天，孙永强.计算机学报，第20卷第8期. 1997 * |
| 一种在存储子系统中实现连续时间点快照的技术. 李中，王刚，刘璟.计算机工程与应用，第40卷第9期. 2004 * |
| 分布式数据库复制方法探讨. 舒后，齐红心.计算机应用，第23卷. 2003 * |
| 基于Snapshot加锁机制的事务模型的研究. 钟远明，奚建清.计算机工程，第27卷第4期. 2001 * |
| 基于快照机制的事务处理及其在RTDBS中的应用. 钟远明，奚建清.计算机工程与应用，第7期. 2002 * |
| 附网存储中快照技术的设计. 周敬利，汤文晖，余胜生，杨立辉.计算机工程与应用，第27期. 2003 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1704903A (en) | 2005-12-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN100359476C (en) | Snapshot backup method | |
| JP2557172B2 (en) | Method and system for secondary file status polling in a time zero backup copy process | |
| US6857057B2 (en) | Virtual storage systems and virtual storage system operational methods | |
| US7991749B2 (en) | Database recovery method applying update journal and database log | |
| US6366987B1 (en) | Computer data storage physical backup and logical restore | |
| EP1814033B1 (en) | Storage system and data restoration method thereof | |
| JP3409859B2 (en) | Control method of control device | |
| US6618794B1 (en) | System for generating a point-in-time copy of data in a data storage system | |
| US7861047B2 (en) | Storage device and information management system | |
| US7127557B2 (en) | RAID apparatus and logical device expansion method thereof | |
| US7587562B2 (en) | Data duplication system, data duplication method and program | |
| US7181581B2 (en) | Method and apparatus for mirroring data stored in a mass storage system | |
| US5600816A (en) | System and method for managing data in a cache system for a disk array | |
| US20040123068A1 (en) | Computer systems, disk systems, and method for controlling disk cache | |
| US6996582B2 (en) | Virtual storage systems and virtual storage system operational methods | |
| US20060224639A1 (en) | Backup system, program and backup method | |
| KR20180025128A (en) | Stream identifier based storage system for managing array of ssds | |
| US6658541B2 (en) | Computer system and a database access method thereof | |
| CN102272733B (en) | Determining modified data in cache for use during a recovery operation | |
| US20050086432A1 (en) | Disk array device having snapshot simulation function | |
| CN101226457A (en) | On-line capacity-enlarging system and method for magnetic disc array | |
| JP7318367B2 (en) | Storage control device and storage control program | |
| US7698500B2 (en) | Disk array system, host interface unit, control method for disk array system, and computer program product for disk array system | |
| US20050223180A1 (en) | Accelerating the execution of I/O operations in a storage system | |
| JPH0573221A (en) | Magnetic tape controller |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20080102 Termination date: 20180603 |