CN109981741A - A kind of maintaining method of distributed memory system - Google Patents
A kind of maintaining method of distributed memory system Download PDFInfo
- Publication number
- CN109981741A CN109981741A CN201910140854.2A CN201910140854A CN109981741A CN 109981741 A CN109981741 A CN 109981741A CN 201910140854 A CN201910140854 A CN 201910140854A CN 109981741 A CN109981741 A CN 109981741A
- Authority
- CN
- China
- Prior art keywords
- node
- service mode
- offline
- write
- memory 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.)
- Pending
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/06—Management of faults, events, alarms or notifications
- H04L41/0654—Management of faults, events, alarms or notifications using network fault recovery
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1095—Replication or mirroring of data, e.g. scheduling or transport for data synchronisation between network nodes
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1097—Protocols in which an application is distributed across nodes in the network for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/56—Provisioning of proxy services
- H04L67/59—Providing operational support to end devices by off-loading in the network or by emulation, e.g. when they are unavailable
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of maintaining methods of distributed memory system, comprising the following steps: under step 1, service mode, node off-line will not trigger recovery process;When step 2, the data object for causing copy to lose for offline node carry out write operation, handled by the way of effective copy+marking, i.e., degradation WriteMode is handled;When step 3, offline node are again online, only this online node can trigger recovery process again, and only synchronize its it is offline during be labeled with the data object of write-in label.The present invention promotes the maintainability of storage cluster, realizes the operations such as disk replacement and the program upgrading under presence, guaranty period read-write operation can be normal by introducing service mode.
Description
Technical field
The present invention relates to field of computer technology more particularly to a kind of maintaining methods of distributed memory system.
Background technique
Distributed memory system will start recovery process in topology change, carry out automaticdata rebalancing.Usually
In the case of, recovery process be fully automated and it is not interruptable.
During Recovery, in order to guarantee the consistency of data, part operation can not be executed, and influence normally making for user
With.When power system capacity is gradually increased, it is meant that need the data volume of recovery to increase, the time of cost is consequently increased, and is
The availability of system reduces.Best engineering practice shows how to reduce the frequency of recovery generation to the greatest extent, is distributed storage system
The important content of system O&M.
However, during practical O&M, if to carry out disk replacement or edition upgrading, corresponding number can only be first allowed
It is offline according to node, an online new node again is replaced after disk or upgrade procedure, this, which means that, must carry out 2 times
Recovery, it is time-consuming and laborious.
Summary of the invention
In view of the above drawbacks of the prior art, technical problem to be solved by the invention is to provide a kind of distributed storages
The maintaining method of system, so as to solve the deficiencies in the prior art.
To achieve the above object, the present invention provides a kind of maintaining methods of distributed memory system, comprising the following steps:
Under step 1, service mode, node off-line will not trigger recovery process;
When step 2, the data object for causing copy to lose for offline node carry out write operation, using effective copy+beat
The mode of label is handled, i.e., degradation WriteMode is handled;
When step 3, offline node are again online, only this online node can trigger recovery process again, and
And the data object of write-in label is labeled with during only synchronizing it offline.
The degradation WriteMode of a kind of maintaining method of above-mentioned distributed memory system, the step 2 handles specific steps
Are as follows:
1, gateway receives client's write request;
2, Gateway determines the back end for needing to forward according to present topology;
3, back end receives the write request of forwarding;
4, judge whether that in service mode, be writes process into degrading, and obtains service mode topology, otherwise normally writes
Process;
5, pass through service mode topology equivalence locations of copies;
6, compare present topology and service mode topology;
7, judge whether there is node off-line under present topology, marked if it is write-in, otherwise write copy;Sentence after write-in label
Whether disconnected present node is write copy, is otherwise terminated under service mode topology if being.
A kind of maintaining method of above-mentioned distributed memory system, under service mode, can only offline unlatching service mode when
Already existing node in cluster.
A kind of maintaining method of above-mentioned distributed memory system, under service mode, can only by offline node again on
Line, cannot online new node.
A kind of maintaining method of above-mentioned distributed memory system under service mode, to offline node, cannot be removed
Or change its working directory.
A kind of maintaining method of above-mentioned distributed memory system, under service mode, when node is again online, start-up parameter
It must be offline completely the same before with its.
The beneficial effects of the present invention are:
The present invention promotes the maintainability of storage cluster, realizes that the disk under presence replaces by introducing service mode
Change with program upgrading etc. operation, guaranty period read-write operation can be normal.
It is described further below with reference to technical effect of the attached drawing to design of the invention, specific structure and generation, with
It is fully understood from the purpose of the present invention, feature and effect.
Detailed description of the invention
Fig. 1 is the degradation write-in flow chart under service mode of the invention.
Specific embodiment
A kind of maintaining method of distributed memory system, comprising the following steps:
Under step 1, service mode, node off-line will not trigger recovery process;
When step 2, the data object for causing copy to lose for offline node carry out write operation, using effective copy+beat
The mode of label is handled, i.e., degradation WriteMode is handled;
When step 3, offline node are again online, only this online node can trigger recovery process again, and
And the data object of write-in label is labeled with during only synchronizing it offline.
Firstly, it is necessary to service mode topology and present topology be recorded, for calculating locations of copies and the write-in of data object
Mark position.The premise write that degrades is that recovery does not occur, else if the data object is carrying out recovery, writes behaviour
Make to enter and wait.Degradation write-in process under service mode is as shown in Figure 1, degradation WriteMode handles specific steps are as follows:
1, gateway receives client's write request;
2, Gateway determines the back end for needing to forward according to present topology;
3, back end receives the write request of forwarding;
4, judge whether that in service mode, be writes process into degrading, and obtains service mode topology, otherwise normally writes
Process;
5, pass through service mode topology equivalence locations of copies;
6, compare present topology and service mode topology;
7, judge whether there is node off-line under present topology, marked if it is write-in, otherwise write copy;Sentence after write-in label
Whether disconnected present node is write copy, is otherwise terminated under service mode topology if being.
In addition, under service mode, can only offline unlatching service mode when cluster in already existing node.Service mode
Under, it can only be again online by offline node, it cannot online new node.Under service mode, to offline node, it cannot remove
Or change its working directory.Under service mode, when node is again online, start-up parameter must with its it is offline before complete one
It causes.
One, inventive principle:
Distributed memory system uses strong consistency strategy to the write-in of copy, thus in copy missing and not yet
When recovery is finished, write operation can not be executed.But if without recovery during node off-line, it is secondary to missing
When this object carries out write operation, and it is not written into the copy (also not writing certainly, because of node off-line) of missing, but is directed to
Missing copy stamps write-in label, guarantees that effective copy+write-in label sum reaches maximum number of copies, then the strongly consistent of write operation
Property is not destroyed.It when offline node is again online, then will be distributed in the data copy on offline node, stamp write-in
The part of label synchronizes, that is, can guarantee the strong consistency of data.
Specific embodiment given below illustrates operation of the present invention Method And Principle:
Using service mode, the functions such as online disk replacement and the upgrading of online gray scale, implementation Datong District may be implemented
Small difference illustrates the application method of service mode here by taking the replacement of online disk as an example:
Firstly, making in the normal situation of cluster state (without offline, the recovery not completed on node)
Starting service mode is ordered with " dog cluster mmode on ";
The corresponding data object service processes of disk to be replaced are killed, and waits topology to update and completes (that is, passing through
" dog cluster info " order, it can be seen that topology caused by the node is offline updates);
The whole disk of disk to be replaced is copied into new disk;
Back end corresponding to original disk (it is required that start-up parameter with before consistent) is restarted, is waited
Recovery (synchronizing the data object of write-in label) is completed;
Close service mode.
The method for realizing gray scale upgrading is similar with the above process, the step of Replace Disk and Press Anykey To Reboot is changed into upgrade installation package, just
It is the process for upgrading a node;Then, according to the repetition of this process one node, one node, until all nodes all rise
Grade at latest edition program.
The preferred embodiment of the present invention has been described in detail above.It should be appreciated that those skilled in the art without
It needs creative work according to the present invention can conceive and makes many modifications and variations.Therefore, all technologies in the art
Personnel are available by logical analysis, reasoning, or a limited experiment on the basis of existing technology under this invention's idea
Technical solution, all should be within the scope of protection determined by the claims.
Claims (6)
1. a kind of maintaining method of distributed memory system, which comprises the following steps:
Under step 1, service mode, node off-line will not trigger recovery process;
When step 2, the data object for causing copy to lose for offline node carry out write operation, using effective copy+marking
Mode handle, i.e., degradation WriteMode handle;
When step 3, offline node are again online, only this online node can trigger recovery process again, and only
The data object of write-in label is labeled with during only synchronizing it offline.
2. a kind of maintaining method of distributed memory system as described in claim 1, it is characterised in that: the drop of the step 2
Grade WriteMode handles specific steps are as follows:
1, gateway receives client's write request;
2, Gateway determines the back end for needing to forward according to present topology;
3, back end receives the write request of forwarding;
4, judge whether that in service mode, be writes process into degrading, and obtains service mode topology, otherwise normally writes stream
Journey;
5, pass through service mode topology equivalence locations of copies;
6, compare present topology and service mode topology;
7, judge whether there is node off-line under present topology, marked if it is write-in, otherwise write copy;Judge to work as after write-in label
Whether front nodal point is write copy, is otherwise terminated under service mode topology if being.
3. a kind of maintaining method of distributed memory system as described in claim 1, it is characterised in that: under service mode, only
Already existing node in cluster when energy offline unlatching service mode.
4. a kind of maintaining method of distributed memory system as described in claim 1, it is characterised in that: under service mode, only
Can be again online by offline node, it cannot online new node.
5. a kind of maintaining method of distributed memory system as described in claim 1, it is characterised in that: right under service mode
Its working directory cannot be removed or be changed to offline node.
6. a kind of maintaining method of distributed memory system as described in claim 1, it is characterised in that: under service mode, section
When point is again online, start-up parameter must be offline completely the same before with its.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910140854.2A CN109981741A (en) | 2019-02-26 | 2019-02-26 | A kind of maintaining method of distributed memory system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910140854.2A CN109981741A (en) | 2019-02-26 | 2019-02-26 | A kind of maintaining method of distributed memory system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109981741A true CN109981741A (en) | 2019-07-05 |
Family
ID=67077337
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910140854.2A Pending CN109981741A (en) | 2019-02-26 | 2019-02-26 | A kind of maintaining method of distributed memory system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109981741A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110990339A (en) * | 2019-10-15 | 2020-04-10 | 平安科技(深圳)有限公司 | Distributed storage file reading and writing method, device and platform and readable storage medium |
Citations (23)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101599032A (en) * | 2009-05-31 | 2009-12-09 | 成都市华为赛门铁克科技有限公司 | Storage node management method, control subsystem and storage system |
WO2011023134A1 (en) * | 2009-08-28 | 2011-03-03 | Beijing Innovation Works Technology Company Limited | Method and system for managing distributed storage system through virtual file system |
CN102202044A (en) * | 2011-02-25 | 2011-09-28 | 北京兴宇中科科技开发股份有限公司 | Portable cloud storage method and device |
US20120166390A1 (en) * | 2010-12-23 | 2012-06-28 | Dwight Merriman | Method and apparatus for maintaining replica sets |
CN103546579A (en) * | 2013-11-07 | 2014-01-29 | 陈靓 | Method for improving availability of distributed storage system through data logs |
US20140089912A1 (en) * | 2012-09-21 | 2014-03-27 | Silver Spring Networks, Inc. | System and method for efficiently updating firmware for nodes in a mesh network |
EP2755161A1 (en) * | 2013-01-14 | 2014-07-16 | Accenture Global Services Limited | Secure online distributed data storage services |
CN104216719A (en) * | 2013-05-30 | 2014-12-17 | 深圳创维无线技术有限公司 | Method and device for updating android system |
CN104615606A (en) * | 2013-11-05 | 2015-05-13 | 阿里巴巴集团控股有限公司 | Hadoop distributed file system and management method thereof |
CN104618487A (en) * | 2015-02-06 | 2015-05-13 | 杭州华三通信技术有限公司 | Internet protocol storage on-line upgrading method and device |
CN105094913A (en) * | 2015-07-31 | 2015-11-25 | 广东欧珀移动通信有限公司 | System, base band fastener and system application upgrading method and device |
CN105659213A (en) * | 2013-10-18 | 2016-06-08 | 日立数据系统工程英国有限公司 | Target-driven independent data integrity and redundancy recovery in a shared-nothing distributed storage system |
CN106406758A (en) * | 2016-09-05 | 2017-02-15 | 华为技术有限公司 | Data processing method based on distributed storage system, and storage equipment |
US20170116302A1 (en) * | 2015-10-22 | 2017-04-27 | Maxta, Inc. | Replica Checkpointing Without Quiescing |
CN106776142A (en) * | 2016-12-23 | 2017-05-31 | 深圳市深信服电子科技有限公司 | A kind of date storage method and data storage device |
CN107526536A (en) * | 2016-06-22 | 2017-12-29 | 伊姆西公司 | For managing the method and system of storage system |
CN107943510A (en) * | 2017-11-23 | 2018-04-20 | 郑州云海信息技术有限公司 | Distributed memory system upgrade method, system, device and readable storage medium storing program for executing |
US9983823B1 (en) * | 2016-12-09 | 2018-05-29 | Amazon Technologies, Inc. | Pre-forking replicas for efficient scaling of a distribued data storage system |
CN108319618A (en) * | 2017-01-17 | 2018-07-24 | 阿里巴巴集团控股有限公司 | A kind of data distribution control method, system and the device of distributed memory system |
CN108427537A (en) * | 2018-01-12 | 2018-08-21 | 上海凯翔信息科技有限公司 | Distributed memory system and its file write-in optimization method, client process method |
US10069914B1 (en) * | 2014-04-21 | 2018-09-04 | David Lane Smith | Distributed storage system for long term data storage |
CN108780460A (en) * | 2016-03-25 | 2018-11-09 | 英特尔公司 | For the distribution index in distributed memory system and the method and apparatus of repositioning object segmentation |
US20180349071A1 (en) * | 2017-05-30 | 2018-12-06 | Kyocera Document Solutions Inc. | Image forming apparatus management system including plural image forming apparatuses and management server for remotely managing plural image forming apparatuses via network, and image forming apparatus management method |
-
2019
- 2019-02-26 CN CN201910140854.2A patent/CN109981741A/en active Pending
Patent Citations (23)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101599032A (en) * | 2009-05-31 | 2009-12-09 | 成都市华为赛门铁克科技有限公司 | Storage node management method, control subsystem and storage system |
WO2011023134A1 (en) * | 2009-08-28 | 2011-03-03 | Beijing Innovation Works Technology Company Limited | Method and system for managing distributed storage system through virtual file system |
US20120166390A1 (en) * | 2010-12-23 | 2012-06-28 | Dwight Merriman | Method and apparatus for maintaining replica sets |
CN102202044A (en) * | 2011-02-25 | 2011-09-28 | 北京兴宇中科科技开发股份有限公司 | Portable cloud storage method and device |
US20140089912A1 (en) * | 2012-09-21 | 2014-03-27 | Silver Spring Networks, Inc. | System and method for efficiently updating firmware for nodes in a mesh network |
EP2755161A1 (en) * | 2013-01-14 | 2014-07-16 | Accenture Global Services Limited | Secure online distributed data storage services |
CN104216719A (en) * | 2013-05-30 | 2014-12-17 | 深圳创维无线技术有限公司 | Method and device for updating android system |
CN105659213A (en) * | 2013-10-18 | 2016-06-08 | 日立数据系统工程英国有限公司 | Target-driven independent data integrity and redundancy recovery in a shared-nothing distributed storage system |
CN104615606A (en) * | 2013-11-05 | 2015-05-13 | 阿里巴巴集团控股有限公司 | Hadoop distributed file system and management method thereof |
CN103546579A (en) * | 2013-11-07 | 2014-01-29 | 陈靓 | Method for improving availability of distributed storage system through data logs |
US10069914B1 (en) * | 2014-04-21 | 2018-09-04 | David Lane Smith | Distributed storage system for long term data storage |
CN104618487A (en) * | 2015-02-06 | 2015-05-13 | 杭州华三通信技术有限公司 | Internet protocol storage on-line upgrading method and device |
CN105094913A (en) * | 2015-07-31 | 2015-11-25 | 广东欧珀移动通信有限公司 | System, base band fastener and system application upgrading method and device |
US20170116302A1 (en) * | 2015-10-22 | 2017-04-27 | Maxta, Inc. | Replica Checkpointing Without Quiescing |
CN108780460A (en) * | 2016-03-25 | 2018-11-09 | 英特尔公司 | For the distribution index in distributed memory system and the method and apparatus of repositioning object segmentation |
CN107526536A (en) * | 2016-06-22 | 2017-12-29 | 伊姆西公司 | For managing the method and system of storage system |
CN106406758A (en) * | 2016-09-05 | 2017-02-15 | 华为技术有限公司 | Data processing method based on distributed storage system, and storage equipment |
US9983823B1 (en) * | 2016-12-09 | 2018-05-29 | Amazon Technologies, Inc. | Pre-forking replicas for efficient scaling of a distribued data storage system |
CN106776142A (en) * | 2016-12-23 | 2017-05-31 | 深圳市深信服电子科技有限公司 | A kind of date storage method and data storage device |
CN108319618A (en) * | 2017-01-17 | 2018-07-24 | 阿里巴巴集团控股有限公司 | A kind of data distribution control method, system and the device of distributed memory system |
US20180349071A1 (en) * | 2017-05-30 | 2018-12-06 | Kyocera Document Solutions Inc. | Image forming apparatus management system including plural image forming apparatuses and management server for remotely managing plural image forming apparatuses via network, and image forming apparatus management method |
CN107943510A (en) * | 2017-11-23 | 2018-04-20 | 郑州云海信息技术有限公司 | Distributed memory system upgrade method, system, device and readable storage medium storing program for executing |
CN108427537A (en) * | 2018-01-12 | 2018-08-21 | 上海凯翔信息科技有限公司 | Distributed memory system and its file write-in optimization method, client process method |
Non-Patent Citations (1)
Title |
---|
施超: "基于Android平台OTA增量升级系统研究与设计", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110990339A (en) * | 2019-10-15 | 2020-04-10 | 平安科技(深圳)有限公司 | Distributed storage file reading and writing method, device and platform and readable storage medium |
CN110990339B (en) * | 2019-10-15 | 2023-09-19 | 平安科技(深圳)有限公司 | File read-write method, device and platform for distributed storage and readable storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102402596B (en) | A kind of reading/writing method of master slave separation database and system | |
CN101964820B (en) | Method and system for keeping data consistency | |
CN109241185A (en) | A kind of method and data synchronization unit that data are synchronous | |
CN106021016A (en) | Virtual point in time access between snapshots | |
CN102981931A (en) | Backup method and device for virtual machine | |
CN103294675A (en) | Method and device for updating data in distributed storage system | |
CN109918229B (en) | Database cluster copy construction method and device in non-log mode | |
US10747776B2 (en) | Replication control using eventually consistent meta-data | |
CN102937909B (en) | A kind of method of disposing and upgrading linux system | |
CN102750322B (en) | Method and system for guaranteeing distributed metadata consistency for cluster file system | |
CN103473287A (en) | Method and system for automatically distributing, running and updating executable programs | |
CN102279857B (en) | Method and system for realizing data reproduction | |
CN105630571A (en) | Virtual machine creating method and device | |
JP5521595B2 (en) | Storage system and storage control method | |
CN114942965B (en) | Method and system for accelerating synchronous operation of main database and standby database | |
CN102495739A (en) | Data compatible method and system as well as inter-plate message method and system | |
CN110007941A (en) | A kind of the MCU firmware and upgrade method of Intelligent refuse classification recovery system | |
CN102193841A (en) | Backup method and device of Subversion configuration database | |
CN102708166B (en) | Data replication method, data recovery method and data recovery device | |
CN102833273A (en) | Data restoring method when meeting temporary fault and distributed caching system | |
CN105205178A (en) | Multi-process access memory database system | |
CN109981741A (en) | A kind of maintaining method of distributed memory system | |
CN107861838A (en) | Method and device of the automated back-up MySQL database from storehouse | |
CN113296804B (en) | Method and device for upgrading database | |
CN104407932B (en) | A kind of data back up method and device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
AD01 | Patent right deemed abandoned |
Effective date of abandoning: 20220830 |
|
AD01 | Patent right deemed abandoned |