CN106325768A - Dual storage system and method - Google Patents
Dual storage system and method Download PDFInfo
- Publication number
- CN106325768A CN106325768A CN201610692191.1A CN201610692191A CN106325768A CN 106325768 A CN106325768 A CN 106325768A CN 201610692191 A CN201610692191 A CN 201610692191A CN 106325768 A CN106325768 A CN 106325768A
- Authority
- CN
- China
- Prior art keywords
- management node
- version
- data block
- node
- storage system
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0668—Interfaces specially adapted for storage systems adopting a particular infrastructure
- G06F3/067—Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0602—Interfaces specially adapted for storage systems specifically adapted to achieve a particular effect
- G06F3/0604—Improving or facilitating administration, e.g. storage management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
- G06F3/0646—Horizontal data movement in storage systems, i.e. moving data in between storage devices or systems
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a dual storage system and method. Management nodes of the system consist of a first management node and a second management node, the first management node and the second management node are mutually mirror images and are used for storing metadata information and data block version information, monitoring a client request and reading and providing data to a client according to the client request. The method comprises a reading operation, a writing operation and a synchronizing operation. As compared with the existing more than three management nodes, the dual storage system and method disclosed by the invention have the advantages that the framework structure is simple and the data consistency is maintained without depending on additional devices. The two management nodes in the dual system are completely independent. When one management node fails, the other management node can continuously provide limited reading and writing service, the problem of single-point failure of the management nodes can be avoided, the reliability of the storage system is guaranteed and the availability of the system is improved.
Description
Technical field
The invention belongs to technical field of memory, more particularly, to a kind of two-shipper storage system and method.
Background technology
Along with developing rapidly of information technology, data also present volatile growth trend.The most a lot of applications pair
The fault-tolerance aspect of storage system has the highest requirement, such as finance, military affairs, industrial quarters, aviation etc..At these applied environments
In, the fault-tolerance of storage system can be brought threat greatly by several factors.Once cause the even overall mistake of memory node part
Effect, will increase the risk servicing unavailable even loss of data.Therefore, improve storage system survivability be one very
Crucial problem.
Common storage system, such as, in distributed memory system, its node mistake can be divided into two classes, a class to be management joints
Point failure, two is that back end lost efficacy.And the access to storage system will add management joint inevitably by way of management node
There is Single Point of Faliure problem in point, once manages node failure, and whole storage system paralyses most probably, it is impossible to externally provide clothes
Business.To this, existing researcher proposes high-availability arrangement and ensures to manage the fault-tolerant ability of node, the reliability in reduction system
With available risk.
Following three kinds of the fault-toleranr technique of existing management node:
The first, metadata backup method.The main thought of the method is that metadata information is saved in multiple by management node
Catalogue is to reach fault-tolerant purpose.Common practice, selects a local directory, a remote directory, and the two catalogue is passed through
NFS shares.When managing nodes break down, start the management node of a standby machine, load in remote directory
Metadata information, it is provided that service.
The second, the second management nodal method.The purpose of the method is exactly to examine the metadata in file system internal memory
Look into, when in fixing a period of time or having reached, when the size of journal file, the size set, just by journal file record
Content update is in metadata image file, thus avoids the unconfined increase of journal file.
The third, hot standby management nodal method.This method provide a hot standby management node, its backup main management node
All data, can replace master node functionality at any time.Main management node includes that customer in response end is asked, Backup Data node data
Etc. function.The metadata of the two management node can keep consistent, and when main management node breaks down and delays machine, hot standby management saves
The time of some switching state and switching is the shortest, this guarantees system reliability.
Above-mentioned the first and second method are the modes restarted based on management node and recover service, simply reduce startup
Time, do not accomplish hot standby.Although the third above-mentioned method breaches restarts restriction, but it depends on independent NFS clothes
Business device ensures the metadata consistency of two management nodes, and the switching of hot standby management node state needs manpower intervention.
Summary of the invention
It is an object of the invention to provide a kind of two-shipper storage system and method, ensureing the reliability of storage system and carrying
While high system availability, maintain the data consistency of dual systems.The method is foundation two-shipper data version during read-write
The different conditions of this number is distinguished the one of data and is made peace inconsistent situation.
For achieving the above object, according to one aspect of the present invention, it is provided that a kind of two-shipper storage system, described double-click is deposited
The management node of storage system is made up of the first and second management nodes, described first and second management node mirror images each other, is used for
Storage metadata information and data block version information, and monitor client request, provide according to client requests data reading
To client.
Preferably, described two-shipper storage system, its first management node, it is additionally operable to store the data block of the second management node
Version information.
Preferably, described two-shipper storage system, its second management node, it is additionally operable to store the data block of the first management node
Version information.
Preferably, described two-shipper storage system, it also includes in the first and the of local file system storage data
Two back end;Described first back end is for providing data to the first management node;Described second back end for
Second management node provides data.
Preferably, described two-shipper storage system, its version information includes local data version number and strange land versions of data number,
Described local data block version number is the reference numeral that on the first or second management node, this data block this locality preserves version;Described
Strange land data block version number be on the first or second management node this data block relative external management node that is second or the
The reference numeral of version is preserved on one management node.
Preferably, described two-shipper storage system, its local version probable value is " 0 " and "-1 ", wherein " 0 " expression " this
Ground data block write operation success ", "-1 " represents " failure of local data block write operation ";" 0 ", "-1 " and "?", wherein " 0 " represents
" receive external management node writes successful information ", "-1 " represents " receive external management node writes failure information ", "?”
Represent " being not received by the feedback report of external management node ".
It is another aspect of this invention to provide that provide a kind of two-shipper storage method, comprise the following steps:
Read operation: local management node receives the read request from client, verification both sides manage the release status of node,
Judge whether to need data syn-chronization.After processing read request, complete to ask to report to client feedback according to read request.
Write operation: local management node receives the write request from client, verification both sides manage the version of node to shape
State, it may be judged whether need data syn-chronization.After processing write requests, report to client and external management node feeding back, and according to writing
Request performance and outside feedback report complete the renewal of local version pair.
Synchronize: all can check, before the read-write operation of data block, the state that both sides manage the version pair of node every time, find
When needing to synchronize, need to wait one hyposynchronous complete after could perform read-write operation.
In general, by the contemplated above technical scheme of the present invention compared with prior art, there is following useful effect
Really:
1, a kind of two-shipper storage system and method that the present invention proposes, manages node with existing more than three relatively, this
Invention frame structure is simple, is no longer rely on extra equipment to maintain data consistency.Two management nodes in dual systems
It is completely self-contained.When a management node failure, another one management node can continue externally to provide limited read-write
Service, can avoid managing the Single Point of Faliure problem of node, it is ensured that the reliability of storage system, improves the availability of system.
2, the management method of a kind of dual systems that the present invention proposes, every time by comparing the shape of two-shipper versions of data number
State, distinguishes the one of data and makes peace inconsistent situation, thus maintain the data consistency of dual systems.
Accompanying drawing explanation
Fig. 1 is the frame diagram of the two-shipper storage system that the embodiment of the present invention provides;
Fig. 2 is the flow chart of the write request of the two-shipper storage method that the embodiment of the present invention provides;Wherein, Fig. 2 a, Fig. 2 b and
Fig. 2 c is respectively the flow process of situation write operation in 3;
Fig. 3 is the operating procedure of local management node under the two-shipper storage method read request that the embodiment of the present invention provides.
Detailed description of the invention
In order to make the purpose of the present invention, technical scheme and advantage clearer, below in conjunction with drawings and Examples, right
The present invention is further elaborated.Should be appreciated that specific embodiment described herein only in order to explain the present invention, and
It is not used in the restriction present invention.If additionally, technical characteristic involved in each embodiment of invention described below
The conflict of not constituting each other just can be mutually combined.
The two-shipper storage system that the present invention provides, including:
First and second management nodes, described first and second management node mirror images each other, it is used for storing metadata information
And data block version information, and monitor client request, it is supplied to client according to client requests data reading;
Described first management node, is additionally operable to store the data block version information of the second management node;
Described second management node, is additionally operable to store the data block version information of the first management node;
Described system, also includes for the first and second back end in local file system storage data;Described
One back end is for providing data to the first management node;Described second back end is for providing number to the second management node
According to.
Described data block version information is version number, including local data block version number and strange land data block version.Described
Local data block version number is the reference numeral that on the first or second management node, this data block this locality preserves version;Described strange land
Data block version number is that on the first or second management node, this data block is managed relative external management node that is second or first
The reference numeral of version is preserved on reason node.The most such as, can be in correspondence while data block is saved on back end A/B
A logarithm value version number, shape such as (vera, verb) is record on management node A/B.Wherein vera is local data block version, table
Showing the version information of local data block, verb is strange land data block version, represents the external management node that same data block is corresponding
Version information.Vera is likely to occur two kinds of values: " 0 " and "-1 ", wherein " 0 " expression " success of local data block write operation ", "-
1 " " failure of local data block write operation " is represented;Verb is likely to occur three kinds of values: " 0 ", "-1 " and "?", wherein " 0 " expression " connects
Receive external management node writes successful information ", "-1 " represents " receive external management node writes failure information ", "?" table
Show " feedback report being not received by external management node ".First management node, in node A record form be (Avera,
Averb), the second management node, the form that records in node B is (Bvera, Bverb).The initial shape of the version number of each data block
State is (0,0).
The two-shipper storage method that the present invention provides, comprises the following steps:
Read operation: local management node receives the read request from client, verification both sides manage the release status of node,
Judge whether to need data syn-chronization.After processing read request, complete to ask to report to client feedback according to read request.
Write operation: local management node receives the write request from client, verification both sides manage the version of node to shape
State, it may be judged whether need data syn-chronization.After processing write requests, report to client and external management node feeding back, and according to writing
Request performance and outside feedback report complete the renewal of local version pair.
Synchronize: all can check, before the read-write operation of data block, the state that both sides manage the version pair of node every time, find
When needing to synchronize, need to wait one hyposynchronous complete after could perform read-write operation.
It is below embodiment:
As it is shown in figure 1, be the frame diagram of two-shipper provided by the present invention storage system.First management node A and the second pipe
The status of reason node B is the most reciprocity, is responsible for metadata information and data block version information is managed and safeguards, monitor
With the request of process client, management node receives the feedback of version information each other and processes.First management node A and the second pipe
Reason node B is each responsible for the back end that management is corresponding.Back end A and back end B is responsible for being stored in data block
In local file system.
Two management nodes receive simultaneously and process the request of user, so can be effectively prevented from Single Point of Faliure problem.
In the case, can produce, because of many nodes, the problem that data are inconsistent.The invention provides a kind of two-shipper storage method, bag
Include:
(1) version number's initialization step: data block can be at corresponding management node while being saved on back end A/B
A logarithm value version number is record on A/B, shape such as (vera, verb), wherein vera represents the version information of local data block,
Verb represents the version information of external management node corresponding to same data block.Management node A in record form be (Avera,
Averb), recording form in management node B is (Bvera, Bverb).The original state of the version number of each data block be (0,
0)。
(2) version number modification step: perform the write operation requests of data block each time, local management node can monitor outside
The write operation feedback report of management node, processes write operation requests simultaneously, can be anti-to client and external management node after completing
Feedback report.With the version of management node A to (Avera, Averb) as a example by, after the success of local data block write operation, Avera can quilt
It is set to " 0 ", is otherwise set to "-1;Averb is likely to occur three kinds of values: " 0 ", "-1 " and "?", wherein " 0 " expression is " outside receiving
Portion management node write successful information ", "-1 " represent " receive external management node writes failure information ", "?" expression " do not has
Receive the feedback report of external management node ".
(3) data syn-chronization step: all can check both sides before the read-write operation of data block every time and manage the version pair of node
State, when finding to need to synchronize, need to wait one hyposynchronous complete after could perform read-write operation.
(4) operating procedure of local management node under write request: local management node receives the write request from client,
Verification both sides manage the version of node to state, it may be judged whether need data syn-chronization.After processing write requests, to client and outside
Management node feeding back report, and the renewal of local version pair is completed according to write request performance and outside feedback report.
(5) operating procedure of local management node under read request: local management node receives the read request from client,
Verification both sides manage the version of node to state, it may be judged whether need data syn-chronization.After processing read request, complete according to read request
Situation is reported to client feedback.
Wherein, step (1) is run when receiving the write request of new data block.Step (2) writes behaviour in each data block
Run when making.When data are inconsistent need synchronize time, operating procedure (3).
Further, as in figure 2 it is shown, described step (4) specifically includes:
(4.1) receiving the write request from client, verification both sides manage the version of node to state, according to same data
State is judged by two versions of block, if need not synchronize, enters step (4.5);If needing local synchronization, enter
Step (4.2);If needing external sync, enter step (4.3);Otherwise, step (4.4) is entered;
(4.2) this locality needs to synchronize, and synchronizes, and synchronizes to be successfully entered step (4.6);Otherwise enter step (4.7);
(4.3) wait that external sync completes, enter step (4.5);
(4.4) confirm that version number finds that this data block is invalid, do not perform write operation, report to client feedback;
(4.5) perform write operation, receive outside report, and complete according to write request performance and outside feedback report
The renewal of local version pair.
(4.6) local version is to changing to (0,0), performs write operation, receives outside report, and completes feelings according to write request
Condition and outside feedback report complete the renewal of local version pair;
(4.7) local version is to changing to (-1,0), does not performs write operation but receives the feedback report of external management node,
The renewal of local version pair is completed according to outside feedback report.
Further, as it is shown on figure 3, described step (5) specifically includes:
(5.1) receiving the read request from client, verification both sides manage the version of node to state, according to same data
State is judged by two versions of block, if need not synchronize, enters step (5.5);If needing local synchronization, enter
Step (5.2);If needing external sync, enter step (5.3);Otherwise, step (5.4) is entered;
(5.2) this locality needs to synchronize, and synchronizes, and synchronizes to be successfully entered step (5.6);Otherwise enter step (5.7);
(5.3) wait that external sync completes, enter step (5.5);
(5.4) confirm that version number finds that this data block is invalid, do not perform read operation, report to client feedback;
(5.5) read operation is performed;
(5.6) local version is to changing to (0,0), performs read operation;
(5.7) local version is to changing to (-1,0), does not perform read operation, reports to client feedback.
As it will be easily appreciated by one skilled in the art that and the foregoing is only presently preferred embodiments of the present invention, not in order to
Limit the present invention, all any amendment, equivalent and improvement etc. made within the spirit and principles in the present invention, all should comprise
Within protection scope of the present invention.
Claims (7)
1. a two-shipper storage system, it is characterised in that the management node of described double-click storage system is by the first and second management
Node forms, described first and second management node mirror images each other, is used for storing metadata information and data block version information,
And monitor client request, it is supplied to client according to client requests data reading.
2. two-shipper storage system as claimed in claim 1, it is characterised in that described first management node, is additionally operable to store the
The data block version information of two management nodes.
3. two-shipper storage system as claimed in claim 1, it is characterised in that described second management node, is additionally operable to store the
The data block version information of one management node.
4. two-shipper storage system as claimed in claim 1, it is characterised in that also include for storing number at local file system
According to the first and second back end;Described first back end is for providing data to the first management node;Described second number
According to node for providing data to the second management node.
5. two-shipper storage system as claimed in claim 1, it is characterised in that described version information includes local data version number
With strange land versions of data number, described local data block version number is that on the first or second management node, this data block this locality preserves version
This reference numeral;Described strange land data block version number be first or second management node on this data block in relative exterior tube
The reference numeral of version is preserved on reason node that is second or the first management node.
6. two-shipper storage system as claimed in claim 5, it is characterised in that described local version probable value be " 0 " and "-
1 ", wherein " 0 " expression " success of local data block write operation ", "-1 " represents " failure of local data block write operation ";“0”、“-1”
"?", wherein " 0 " expression " receive external management node writes successful information ", "-1 " expression " receives external management node
Write failure information ", "?" expression " is not received by the feedback report of external management node ".
7. a two-shipper storage method, it is characterised in that comprise the following steps:
Read operation: local management node receives the read request from client, verification both sides manage the release status of node, it is judged that
The need of data syn-chronization;After processing read request, complete to ask to report to client feedback according to read request;
Write operation: local management node receives from the write request of client, verification both sides manage the version of node to state, sentence
Disconnected the need of data syn-chronization;After processing write requests, report to client and external management node feeding back, and complete according to write request
One-tenth situation and outside feedback report complete the renewal of local version pair.
Synchronize: all can check, before the read-write operation of data block, the state that both sides manage the version pair of node every time, find to need
During synchronization, need to wait one hyposynchronous complete after could perform read-write operation.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610692191.1A CN106325768B (en) | 2016-08-19 | 2016-08-19 | A kind of two-shipper storage system and method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610692191.1A CN106325768B (en) | 2016-08-19 | 2016-08-19 | A kind of two-shipper storage system and method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106325768A true CN106325768A (en) | 2017-01-11 |
CN106325768B CN106325768B (en) | 2019-08-13 |
Family
ID=57744635
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610692191.1A Active CN106325768B (en) | 2016-08-19 | 2016-08-19 | A kind of two-shipper storage system and method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106325768B (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108234641A (en) * | 2017-12-29 | 2018-06-29 | 北京奇虎科技有限公司 | Data read-write method and device based on distributed consensus protocol realization |
CN109151071A (en) * | 2018-10-26 | 2019-01-04 | 福建深空信息技术有限公司 | A kind of web page files synchronous method and system |
WO2020140523A1 (en) * | 2018-12-30 | 2020-07-09 | 浙江宇视科技有限公司 | Data storage switching method, apparatus, and device, and storage medium |
CN113986148A (en) * | 2021-12-27 | 2022-01-28 | 苏州浪潮智能科技有限公司 | Data reading method and device and related equipment |
CN114594914A (en) * | 2022-03-17 | 2022-06-07 | 阿里巴巴(中国)有限公司 | Control method and system for distributed storage system |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050114538A1 (en) * | 2001-11-28 | 2005-05-26 | Interactive Content Engines, Llc | Synchronized data transfer system |
CN101958782A (en) * | 2010-06-21 | 2011-01-26 | 中兴通讯股份有限公司 | Method for realizing node standby and system |
CN105069152A (en) * | 2015-08-25 | 2015-11-18 | 航天恒星科技有限公司 | Data processing method and apparatus |
-
2016
- 2016-08-19 CN CN201610692191.1A patent/CN106325768B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050114538A1 (en) * | 2001-11-28 | 2005-05-26 | Interactive Content Engines, Llc | Synchronized data transfer system |
CN101958782A (en) * | 2010-06-21 | 2011-01-26 | 中兴通讯股份有限公司 | Method for realizing node standby and system |
CN105069152A (en) * | 2015-08-25 | 2015-11-18 | 航天恒星科技有限公司 | Data processing method and apparatus |
Non-Patent Citations (2)
Title |
---|
张伟伟: "面向I/O密集型应用的高性能计算机存储系统性能优化方法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
王皓: "BlueOcean海量存储系统客户端的设计与实现", 《中国优秀硕士学位论文全文数据库 信息科技辑 》 * |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108234641A (en) * | 2017-12-29 | 2018-06-29 | 北京奇虎科技有限公司 | Data read-write method and device based on distributed consensus protocol realization |
CN108234641B (en) * | 2017-12-29 | 2021-01-29 | 北京奇元科技有限公司 | Data reading and writing method and device based on distributed consistency protocol |
CN109151071A (en) * | 2018-10-26 | 2019-01-04 | 福建深空信息技术有限公司 | A kind of web page files synchronous method and system |
CN109151071B (en) * | 2018-10-26 | 2021-09-03 | 福建深空信息技术有限公司 | Webpage file synchronization method and system |
WO2020140523A1 (en) * | 2018-12-30 | 2020-07-09 | 浙江宇视科技有限公司 | Data storage switching method, apparatus, and device, and storage medium |
CN113986148A (en) * | 2021-12-27 | 2022-01-28 | 苏州浪潮智能科技有限公司 | Data reading method and device and related equipment |
CN114594914A (en) * | 2022-03-17 | 2022-06-07 | 阿里巴巴(中国)有限公司 | Control method and system for distributed storage system |
CN114594914B (en) * | 2022-03-17 | 2024-04-02 | 阿里巴巴(中国)有限公司 | Control method and system for distributed storage system |
Also Published As
Publication number | Publication date |
---|---|
CN106325768B (en) | 2019-08-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US6823474B2 (en) | Method and system for providing cluster replicated checkpoint services | |
CN106325768A (en) | Dual storage system and method | |
US6249879B1 (en) | Root filesystem failover in a single system image environment | |
US8935211B2 (en) | Metadata management for fixed content distributed data storage | |
US7657581B2 (en) | Metadata management for fixed content distributed data storage | |
US6247139B1 (en) | Filesystem failover in a single system image environment | |
US7779295B1 (en) | Method and apparatus for creating and using persistent images of distributed shared memory segments and in-memory checkpoints | |
US20150347250A1 (en) | Database management system for providing partial re-synchronization and partial re-synchronization method of using the same | |
US20050193248A1 (en) | Computer system for recovering data based on priority of the data | |
US10180812B2 (en) | Consensus protocol enhancements for supporting flexible durability options | |
US8010490B2 (en) | Apparatus for managing remote copying between storage systems | |
GB2484086A (en) | Reliability and performance modes in a distributed storage system | |
CN104994168A (en) | distributed storage method and distributed storage system | |
CN111125021B (en) | Method and system for efficiently restoring consistent views of file system images from an asynchronous remote system | |
US8190947B1 (en) | Method and system for automatically constructing a replica catalog for maintaining protection relationship information between primary and secondary storage objects in a network storage system | |
JP5292351B2 (en) | Message queue management system, lock server, message queue management method, and message queue management program | |
US20050278382A1 (en) | Method and apparatus for recovery of a current read-write unit of a file system | |
CN112384906A (en) | MVCC-based database system asynchronous cache consistency | |
JP6940645B2 (en) | Replica processing methods, nodes, storage systems, servers and readable storage media | |
CN113010549A (en) | Data processing method based on remote multi-active system, related equipment and storage medium | |
US20160210204A1 (en) | Distributed disaster recovery file sync server system | |
US7284105B2 (en) | Fast and economical establishment of remote copy | |
CN116389233B (en) | Container cloud management platform active-standby switching system, method and device and computer equipment | |
JP2009265973A (en) | Data synchronization system, failure recovery method, and program | |
US8516023B1 (en) | Context based file system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |