CN103428288B - Based on the copies synchronized method of subregion state table and coordinator node - Google Patents
Based on the copies synchronized method of subregion state table and coordinator node Download PDFInfo
- Publication number
- CN103428288B CN103428288B CN201310350158.7A CN201310350158A CN103428288B CN 103428288 B CN103428288 B CN 103428288B CN 201310350158 A CN201310350158 A CN 201310350158A CN 103428288 B CN103428288 B CN 103428288B
- Authority
- CN
- China
- Prior art keywords
- node
- subregion
- partition holding
- state
- subregion state
- 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.)
- Active
Links
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of copies synchronized method based on subregion state table and coordinator node, the method safeguards copy consistency by the copies synchronized mechanism of this mandate, the memory node at the copy place upgraded is needed to send synchronization request to coordinator node, according to subregion state summary table, coordinator node judges whether this memory node meets the condition performing copies synchronized operation, license to qualified memory node and perform copies synchronized operation.Copies synchronized method based on subregion state table and coordinator node of the present invention greatly reduces the size of the duplicate detection rate of copy consistency, synchronization bandwidth occupancy and inconsistent window, avoid the version conflict problem of multiple copy, and increase the ageing of copies synchronized.Make distributed file storage system more rapid to the detection of the response of external request, copy consistency, the service ability of elevator system and the efficiency of final consistency.
Description
Technical field
The present invention relates to technical field of memory, be specifically related to a kind of copies synchronized method based on subregion state table and coordinator node.
Background technology
A data store on one node by traditional data storage technology, need the user of these data all must read it to same node.In recent years, along with the demand of the user in the Internet to resource is increasing, network easily occurs congested, and the limited node of disposal ability also can because access number is too large machine of delaying.And traditional data storage technology is in extensibility, the aspects such as high availability cannot meet current data storage requirement far away, the eighties in 20th century, and the concept that distributed document stores is suggested.
Distributed document storage refers to and utilizes distributed computing technology, be connected with node by computer network, memory device on node different in network is got up by Distributed Application software assembly, by storage system unified management, collaborative work, the common data that externally provide store and business, provide access interface that is unified, objectification, shield the direct control to physical equipment and resource management to user.Distributed file storage system is by distributed framework, create multiple copies, and by their reasonable layouts on multiple server node, for user's access, utilize multiple stage storage server to share and store load and process access request, can effectively reduce node failure rate, reduce subscriber response time.
Stored by distributed document, create multiple copies, utilize location server to locate storage information, not only increase the reliability of system, availability and access efficiency, all kinds of memory device can be combined, good extendible stores service is provided.Compare traditional memory device, it is in autgmentability, fail safe, and the aspects such as reliability are all greatly improved.
But owing to there is multiple copy, there will be the nonsynchronous problem of copy, i.e. multiple copy inconsistency, the result causing different user to access is inconsistent, distributed file storage system availability and reliability is caused to reduce.In order to keep the consistency of copy in distributed file storage system, propose duplicate of the document strategy.Be the consistency safeguarding copy on the one hand, improve availability and the reliability of distributed file storage system.The overall performance of elevator system on the other hand.Such as by increasing copy in storage server, allowing more storage server provide service, strengthening load balancing and the extended capability of system.
Propose the file stored copies Managed Solution under a lot of distributed structure/architecture in recent years.Different according to the framework of distributed file system, mainly can be divided into the replication strategy of centralization framework storage system and the replication strategy of symmetrical architecture storage system.
(1) replication strategy of centralization framework storage system
Mainly in the storage system of centralization framework adopt analog storage state method, consistency detection that major-minor this law realizes copy.Analog storage state method is the list of maintenance documentation block in each memory node, carries out block message exchange by heartbeat and namenode, controls the synchronous of copy and copy by namenode.Major-minor this law picks out a reference copies in multiple copies of same file, and all write operations all carry out primary copy, and controls the synchronous of other copies by primary copy and copy.
(2) replication strategy of symmetrical architecture storage system
Because centralization framework exists obvious namenode bottleneck problem, therefore the better symmetrical architecture of autgmentability is suggested and is tending towards main flow.The main duplicate detection mechanism adopting " point-to-point " between memory node in symmetrical architecture storage system.As its name suggests, in the duplicate detection mechanism of " point-to-point ", memory node is by the copy check code on the distant-end node of its equity of request and version number, compare with local replica, if local replica and far-end copy is inconsistent and far-end copy version is older, then local replica by local replica data-pushing to peer node, can realize the copies synchronized of " point-to-point ".
The replication strategy of traditional symmetrical storage framework has the following disadvantages:
1) duplicate detection amount is large.Each subregion is had to the symmetrical architecture storage system of 3 copies, the consistency of 3 copies be ensured, need between 3 memory nodes, to send 6 message and detect.Although this mode is easy to realize, takies massive band width, reduce the overall performance of system.
2) inconsistent window amplifies.For single memory node, each takes turns the conforming detection of copy all can consume the regular hour, this interval time can increase along with the increasing of number of partitions on memory node, thus cause copy inconsistent to be in a large number delayed by could detecting comparatively for a long time, increase the inconsistent window of system, make the efficiency of final consistency lower, user very easily has access to inconsistent data.
Summary of the invention
For the deficiencies in the prior art, the invention provides the copies synchronized method based on subregion state table and coordinator node, avoid duplicate detection, and inconsistent window is little.
A kind of copies synchronized method based on subregion state table and coordinator node, described copies synchronized method is applied to the distributed file storage system of symmetrical architecture, described distributed file storage system comprises multiple memory node and a coordinator node, and described copies synchronized method comprises:
(1) initialization
1-1) each memory node sets up subregion state table, the partition number of the partition holding of described subregion state table real time record respective stored node, subregion state and version number, after initialization, subregion state and version number are empty;
Described subregion state comprises: " Complete Synchronization ", " up-to-date ", " being synchronized to far-end " and " being updated to this locality " four kinds of states;
All subregion state tables gather and set up subregion state summary table by the coordinator node 1-2);
(2) synchronization request is sent
If the data of a certain partition holding are modified in certain memory node, and the conflict of this partition holding stateless, the subregion state then upgrading this partition holding is " up-to-date ", simultaneously with this memory node for source node to send the synchronization request of current partition holding to described coordinator node;
(3) return instruction is asked
Synchronization request described in coordinator node receives, and according to subregion state summary table, judge whether current partition holding meets the executive condition of copies synchronized operation, and send corresponding return instruction, be specially:
If 3-1) do not meet described executive condition, then to maintain the subregion state of current partition holding constant for coordinator node, and return " applying for unsuccessfully " to source node;
If 3-2) meet described executive condition, then coordinator node upgrades the subregion state of front partition holding is " being synchronized to far-end ", and returns " applying for successfully " to source node;
(4) simultaneous operation is performed
Return instruction described in source node receives, and make following operation:
If 4-1) return instruction is " applying for unsuccessfully ", then source node keeps the subregion state of current partition holding constant, and waits for copies synchronized;
If 4-2) return instruction is " applying for successfully ", then source node performs copies synchronized operation to the memory node at all copy places of current partition holding, makes all copies of current partition holding consistent.
In copies synchronized method of the present invention, each memory node safeguards a subregion state summary table respectively, have recorded the partition number of all partition holdings in respective stored node, subregion state and version number; Coordinator node safeguards subregion state summary table, have recorded all partition informations of all nodes in this system, comprises partition number, subregion state and version number.Coordinator node receives the synchronization request of memory node, judges whether this memory node meets the condition performing copies synchronized operation, licenses to qualified memory node according to subregion state summary table, allows this node perform copies synchronized operation; Ineligible memory node, then show have other replica node to have sent copies synchronized application prior to source node, source node hold mode is constant, and waits for copies synchronized.The copies synchronized mechanism of this mandate effectively can reduce the useless generation copied, reduce synchronization bandwidth to take, can well manage copy version simultaneously, avoid the version conflict problem of multiple copy, and increase the ageing of copies synchronized, reduce inconsistent window.And copies synchronized operation is carried out between memory node, not through coordinator node, coordinator node is only responsible for authorizing according to the simultaneous operation of request authoritative copy, effectively can alleviate the pressure of coordinator node.
Four kinds of subregion states in described step (1) are changed according to finite state machine principle.
Realize four kinds of subregion state conversions by finite state machine principle, the order that four kinds of subregion states are changed mutually can be limited.
Synchronization request in described step (2) carries the partition number of current partition holding, subregion state and version number.
The information of carrying in synchronization request is convenient to coordinator node and is processed.
Executive condition in described step (3) is the conflict of current partition holding stateless and version conflict.
Described state conflict is determined by the transfer principle of state, if current partition holding does not meet the switch condition of finite state machine, then thinks current partition holding state conflict.
Described version conflict judges according to following methods: if up-to-date in the not all copy of the copy version of current partition holding, then think current partition holding version conflict.
The judgement of state conflict prior to version conflict, when existence conflict just can not judge whether to there is version conflict further.State conflict both can judge according to the Condition Areas table on memory node also can according to the total list deciding of subregion state on coordinator node.Version conflict judges on coordinator node, by more all memory nodes deposit the version number of copy, judging that whether the version attempting to upgrade is up-to-date in all copy version, is then version conflict if not up-to-date.State conflict is generally due to the simultaneous operation and this simultaneous operation do not perform current memory node causes prior to current memory node application of other memory nodes.The time difference of these two requests is inconsistent window.By the Synchronization Design framework of coordinator node, and the state of state summary table and Version Control, can effectively reduce inconsistent window.
As preferably, described copies synchronized operating basis NWR strategy carries out.
The consistency (Consistency) of distributed system, availability (Availability), subregion tolerable (ToleranceofnetworkPartition) cannot meet simultaneously, adopt classical NWR strategy, in above three kinds of performances, make balance.N represents the N number of data backup of needs, and W represents that data write N number of node and are just counted as merit, and R represents that the data guarantee needing to read R node wherein has portion to be up-to-date.Such as need backup number to be 3 parts, write 2 parts and namely represent that data successfully write, then must read 2 numbers wherein has portion to be up-to-date according to guarantee.
Described step 3-2) in:
Before copies synchronized operation, the state updating of partition holding current in subregion state table is " being synchronized to far-end " by source node;
In copies synchronized operating process, the subregion state of respective stored subregion is updated to " being updated to this locality " by the memory node at each copy place;
After copies synchronized has operated, the subregion state of corresponding partition holding is all updated to " Complete Synchronization " by source memory node and copy place memory node and coordinator node.
Operate in forward and backward and simultaneous operation process in execution copies synchronized, associated storage node is according to the subregion state of the corresponding partition holding of operation real-time update.Because the realization of data read-write operation and copies synchronized is quite independent, adopt and carry out data manipulation strategy based on subregion state, for the different read-write strategy of the subregion state setting that setting is different, efficiently solve the subregion read/write conflict problem occurred in state conversion process, prevent the problem of multi version confusion, concrete grammar is as follows:
1) subregion state of copy place partition holding is " being updated to this locality ":
If read operation, then operate from the copies synchronized that this memory node is corresponding the data reading respective partition corresponding distant-end node (memory node at each copy place),
If retouching operation, then this memory node no longer accept to upload, the retouching operation such as deletion, and described retouching operation is transferred to distant-end node corresponding to copies synchronized operation corresponding to this memory node, respective stored subregion in distant-end node is modified operation;
2) subregion state of copy place partition holding is " being synchronized to far-end ":
It is if read operation, then direct from current memory node reading data,
If retouching operation, then wait for after copies synchronized has operated, then will this retouching operation be carried out;
3) subregion state of copy place partition holding is " up-to-date " or " Complete Synchronization " state:
It is if read-write operation, then direct from current memory node reading data,
If retouching operation, then perform this read-write operation immediately, and after this read-write operation completes, be " up-to-date " by corresponding partition holding state updating, and send the synchronization request of this partition holding to coordinator node.
Described distributed file storage system also comprises a hot standby node, the synchronization request continuous several times time-out that current memory node sends, then think that coordinator node is delayed machine, hot standby node is automatically taken over coordinator node and is received synchronization request and send return instruction, after coordinator node recovers, then synchronization request is transferred to coordinator node process.
Hot standby node, as the secondary node of coordinator node, when coordinator node delays machine, is automatically enabled, is improve the reliability of system.
Copies synchronized method based on subregion state table and coordinator node provided by the invention, the copy consistency in distributed file storage system is safeguarded by the copies synchronized operation mechanism of authorizing, greatly reduce the duplicate detection rate of copy consistency, synchronization bandwidth takies and inconsistent window, thus reduce the pressure of copies synchronized operation to distributed file storage system, make distributed file storage system more rapid to the detection of the response of external request, copy consistency, the service ability of elevator system and the efficiency of final consistency.
Accompanying drawing explanation
Fig. 1 is the Organization Chart with the distributed file storage system of the symmetrical architecture of coordinator node of the present embodiment;
To be the present embodiment be Fig. 2 based on the flow chart of the copies synchronized method of subregion state table and coordinator node;
Fig. 3 is the finite state machine schematic diagram of subregion state conversion.
Embodiment
Below in conjunction with the drawings and specific embodiments, the present invention will be further described.
The applied environment of the copies synchronized method based on subregion state table and coordinator node in the present embodiment is the distributed file storage system of symmetrical architecture.As shown in Figure 1, the distributed file storage system of this symmetrical architecture comprises the memory node of multiple equity, a coordinator node and a hot standby node.Originally execute the communication of memory node and coordinator node in example, and the communication between memory node all adopts the synchronous communications protocol based on http protocol.
The copies synchronized method based on subregion state table and coordinator node of the present embodiment, specific works flow process as shown in Figure 2, comprising:
(1) initialization
1-1) each memory node is set up subregion state table (in the present embodiment for Hash table), the partition number of the partition holding of real time record respective stored node, subregion state and version number, after initialization, subregion state and version number be sky;
All subregion state tables gather and set up subregion state summary table by the coordinator node 1-2);
Subregion state has following four kinds:
Complete Synchronization: show that all copies keep the state of Complete Synchronization;
Up-to-date: to show that copy is up-to-date, and do not produce synchronous with other copies;
Be synchronized to far-end: show up-to-date copy, and this copy is being synchronized to far-end copy;
Be updated to this locality: show old copy, and distally receive up-to-date copy.
Four kinds of subregion states are changed according to finite state machine principle, concrete transfer process as shown in Figure 3:
A the subregion state of () current partition holding is " up-to-date ":
If this partition holding is modified, then it is still " up-to-date " state;
If this subregion sends synchronization request to coordinator node, and when the return information received is " applying for successfully ", be then transformed to " being synchronized to far-end " state;
If receive the copies synchronized operation sent source node, be then transformed to " being updated to this locality " state;
B the subregion state of () current partition holding is " being synchronized to far-end ":
After copies synchronized operation performs, be then transformed to " Complete Synchronization " state;
C () is in " being updated to this locality " state when copy:
Copies synchronized operation is finished, be then transformed to " Complete Synchronization " state;
D the subregion state of () current partition holding is " Complete Synchronization ":
If the data of current partition holding are modified, be then transformed to " up-to-date " state,
If current bay receives the copies synchronized operation sent source node, be then transformed to " being updated to this locality ".
(2) synchronization request is sent
If the data of a certain partition holding are modified in certain memory node, and the conflict of this partition holding stateless, namely this partition holding can meet state machine principle from current bay State Transferring to " up-to-date ", the subregion state then upgrading the respective partition in the subregion state table of this node is " up-to-date ", simultaneously with this memory node for source node to send the synchronization request of current bay to coordinator node; This synchronization request carries the partition number of current partition holding, subregion state and version number.
(3) request returns
Coordinator node receives synchronization request, and according to subregion state summary table, whether what judge current partition holding meets copy executive condition, sends return instruction to source node:
3-1) first judge whether state conflict, if there is state conflict, returned " applying for unsuccessfully " to source node;
If 3-2) stateless conflict, has determined whether version conflict further, if without version conflict, then return " applying for successfully " to source node.
The subregion state of current partition holding is obtained in subregion state summary table from the subregion state table or coordinator node of memory node, if the subregion state of current partition holding cannot be " being synchronized to far-end " from current State Transferring according to finite state machine principle, then think current partition holding state conflict;
From the subregion state summary table coordinator node, obtain the version number of all copies of current partition holding, if in the not all copy of the copy version of current partition holding " up-to-date ", then think current partition holding version conflict;
(4) simultaneous operation is performed
Return instruction described in source node receives, and make following operation:
If 4-1) return instruction is " applying for unsuccessfully ", then source node keeps the subregion state of current partition holding constant, and waits for copies synchronized;
If 4-2) return instruction is for apply for successfully, then the state updating of partition holding current in subregion state table is " being synchronized to far-end " by source node; And operate to the copies synchronized of the memory node execution foundation NWR strategy at all copy places of current partition holding, make all copies of current partition holding consistent; The subregion state of respective stored subregion changes to " being updated to this locality " after receiving the copies synchronized operation that source node sends by the memory node at each copy place; After copies synchronized has operated, the subregion state of respective partition is updated to " Complete Synchronization " by source memory node and copy place memory node and coordinator node.
For preventing because user writable operates the chaotic problem of the version caused, user carries out data manipulation according to the subregion state of accessed copy place partition holding:
1) subregion state of copy place partition holding is " being updated to this locality ":
If read operation, then save from the far-end that the copies synchronized operation that this memory node is corresponding is corresponding the data reading respective partition (source node sending copies synchronized operation) point,
If retouching operation, then this node no longer accept to upload, the retouching operation such as deletion, and described retouching operation is transferred to distant-end node corresponding to copies synchronized operation corresponding to this node, is modified operation in respective partition in distant-end node;
2) subregion state of copy place subregion is " being synchronized to far-end ":
It is if read operation, then direct from present node reading data,
If retouching operation, then wait for after copies synchronized has operated, then will this retouching operation be carried out;
3) subregion state of copy place subregion is " up-to-date " or " Complete Synchronization " state:
If read-write operation, then direct from present node reading data;
If retouching operation, then perform this read-write operation immediately, and after this read-write operation completes, corresponding subregion state is updated to " up-to-date ", and send the synchronization request of this subregion to coordinator node.
Described distributed file storage system also comprises a hot standby node, the synchronization request continuous several times time-out that current memory node sends, be 3 times in the present embodiment, then think that coordinator node is delayed machine, hot standby node is automatically taken over coordinator node and is received synchronization request and send return instruction, after coordinator node recovers, then synchronization request is transferred to coordinator node process.The machine if coordinator node and hot standby node are delayed simultaneously, then memory node can carry out the retry operation of 3 times, congested in order to avoid asking, and the stand-by period of 3 retry operations increases progressively gradually, is respectively 10s, 20s, 30s.If 3 times retry is all failed, then memory node reactivates the copies synchronized strategy of " point-to-point ", and finally starts " point-to-point " copies synchronized strategy of whole storage cluster by the mode of Gossip message propagation.
The above; be only the specific embodiment of the present invention, but protection scope of the present invention is not limited thereto, is anyly familiar with those skilled in the art in the technical scope that the present invention discloses; the change that can expect easily or replacement, all should be encompassed within protection scope of the present invention.
Claims (6)
1. the copies synchronized method based on subregion state table and coordinator node, it is characterized in that, described copies synchronized method is applied to the distributed file storage system of symmetrical architecture, described distributed file storage system comprises multiple memory node and a coordinator node, and described copies synchronized method comprises:
(1) initialization
1-1) each memory node sets up subregion state table, the partition number of the partition holding of described subregion state table real time record respective stored node, subregion state and version number, after initialization, subregion state and version number are empty;
Described subregion state comprises: " Complete Synchronization ", " up-to-date ", " being synchronized to far-end " and " being updated to this locality " four kinds of states;
All subregion state tables gather and set up subregion state summary table by the coordinator node 1-2);
(2) synchronization request is sent
If the data of a certain partition holding are modified in certain memory node, and the conflict of this partition holding stateless, the subregion state then upgrading this partition holding is " up-to-date ", simultaneously with this memory node for source node to send the synchronization request of current partition holding to described coordinator node;
(3) return instruction is asked
Synchronization request described in coordinator node receives, and according to subregion state summary table, judge whether current partition holding meets the executive condition of copies synchronized operation, and send corresponding return instruction, be specially:
If 3-1) do not meet described executive condition, then to maintain the subregion state of current partition holding constant for coordinator node, and return " applying for unsuccessfully " to source node;
If 3-2) meet described executive condition, then coordinator node upgrades the subregion state of front partition holding is " being synchronized to far-end ", and returns " applying for successfully " to source node;
(4) simultaneous operation is performed
Return instruction described in source node receives, and make following operation:
If 4-1) return instruction is " applying for unsuccessfully ", then source node keeps the subregion state of current partition holding constant, and waits for copies synchronized;
If 4-2) return instruction is " applying for successfully ", then source node performs copies synchronized operation to the memory node at all copy places of current partition holding, makes all copies of current partition holding consistent;
Executive condition in described step (3) is the conflict of current partition holding stateless and version conflict;
Described state conflict is determined by the transfer principle of state, if current partition holding does not meet the switch condition of finite state machine, then thinks current partition holding state conflict;
Described version conflict judges according to following methods: if up-to-date in the not all copy of the copy version of current partition holding, then think current partition holding version conflict.
2., as claimed in claim 1 based on the copies synchronized method of subregion state table and coordinator node, it is characterized in that, four kinds of subregion states in described step (1) are changed according to finite state machine principle.
3. as claimed in claim 2 based on the copies synchronized method of subregion state table and coordinator node, it is characterized in that, the synchronization request in described step (2) carries the partition number of current partition holding, subregion state and version number.
4., as claimed in claim 3 based on the copies synchronized method of subregion state table and coordinator node, it is characterized in that, described copies synchronized operating basis NWR strategy carries out.
5., as claimed in claim 4 based on the copies synchronized method of subregion state table and coordinator node, it is characterized in that, described step 3-2) in:
Before copies synchronized operation, the state updating of partition holding current in subregion state table is " being synchronized to far-end " by source node;
In copies synchronized operating process, the subregion state of respective stored subregion is updated to " being updated to this locality " by the memory node at each copy place;
After copies synchronized has operated, the subregion state of corresponding partition holding is all updated to " Complete Synchronization " by source memory node and copy place memory node and coordinator node.
6. as claimed in claim 5 based on the copies synchronized method of subregion state table and coordinator node, it is characterized in that, described distributed file storage system also comprises a hot standby node, the synchronization request continuous several times time-out that current memory node sends, then think that coordinator node is delayed machine, hot standby node is automatically taken over coordinator node and is received synchronization request and send return instruction, after coordinator node recovers, then synchronization request is transferred to coordinator node process.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310350158.7A CN103428288B (en) | 2013-08-13 | 2013-08-13 | Based on the copies synchronized method of subregion state table and coordinator node |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310350158.7A CN103428288B (en) | 2013-08-13 | 2013-08-13 | Based on the copies synchronized method of subregion state table and coordinator node |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103428288A CN103428288A (en) | 2013-12-04 |
| CN103428288B true CN103428288B (en) | 2016-03-09 |
Family
ID=49652452
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310350158.7A Active CN103428288B (en) | 2013-08-13 | 2013-08-13 | Based on the copies synchronized method of subregion state table and coordinator node |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103428288B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105516230B (en) * | 2014-10-15 | 2019-06-25 | 深圳市腾讯计算机系统有限公司 | A kind of data processing method and device |
| CN105610879B (en) * | 2014-10-31 | 2019-05-17 | 华为技术有限公司 | Data processing method and device |
| CN105550306A (en) * | 2015-12-14 | 2016-05-04 | 北京奇虎科技有限公司 | Multi-copy data reading/writing method and system |
| CN107219997B (en) * | 2016-03-21 | 2020-08-18 | 阿里巴巴集团控股有限公司 | Method and device for verifying data consistency |
| CN110968641B (en) * | 2018-09-30 | 2023-05-02 | 阿里巴巴集团控股有限公司 | Data writing control method and device and electronic equipment |
| CN115328880B (en) * | 2022-10-13 | 2023-03-24 | 浙江智臾科技有限公司 | Distributed file online recovery method, system, computer equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101188544A (en) * | 2007-12-04 | 2008-05-28 | 浙江大学 | File transfer method for distributed file server based on buffer |
| CN101187931A (en) * | 2007-12-12 | 2008-05-28 | 浙江大学 | Distribution type file system multi-file copy management method |
| CN102368267A (en) * | 2011-10-25 | 2012-03-07 | 曙光信息产业(北京)有限公司 | Method for keeping consistency of copies in distributed system |
| CN103139302A (en) * | 2013-02-07 | 2013-06-05 | 浙江大学 | Real-time copy scheduling method considering load balancing |
| CN103150347A (en) * | 2013-02-07 | 2013-06-12 | 浙江大学 | Dynamic replica management method based on file heat |
-
2013
- 2013-08-13 CN CN201310350158.7A patent/CN103428288B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101188544A (en) * | 2007-12-04 | 2008-05-28 | 浙江大学 | File transfer method for distributed file server based on buffer |
| CN101187931A (en) * | 2007-12-12 | 2008-05-28 | 浙江大学 | Distribution type file system multi-file copy management method |
| CN102368267A (en) * | 2011-10-25 | 2012-03-07 | 曙光信息产业(北京)有限公司 | Method for keeping consistency of copies in distributed system |
| CN103139302A (en) * | 2013-02-07 | 2013-06-05 | 浙江大学 | Real-time copy scheduling method considering load balancing |
| CN103150347A (en) * | 2013-02-07 | 2013-06-12 | 浙江大学 | Dynamic replica management method based on file heat |
Non-Patent Citations (1)
| Title |
|---|
| P2P 存储系统副本一致性维护策略;王禹 等;《计算机工程》;20100105;第36卷(第1期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103428288A (en) | 2013-12-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103428288B (en) | Based on the copies synchronized method of subregion state table and coordinator node | |
| CN107707393B (en) | Multi-active system based on Openstack O version characteristics | |
| US9280430B2 (en) | Deferred replication of recovery information at site switchover | |
| US9031910B2 (en) | System and method for maintaining a cluster setup | |
| US8676760B2 (en) | Maintaining data integrity in data servers across data centers | |
| US20140244578A1 (en) | Highly available main memory database system, operating method and uses thereof | |
| CN105607954A (en) | Stateful container online migration method and apparatus | |
| JP2014123218A (en) | Program, data management method, and information processing device | |
| CN103488546A (en) | Backup and recovery method capable of supporting online concurrency of multi-level data and database | |
| CN103780638A (en) | Data synchronization method and system | |
| CN110912991A (en) | Super-fusion-based high-availability implementation method for double nodes | |
| JP2013114684A (en) | System and method for managing virtual machine | |
| CN102571991A (en) | Multistage-mapping-based large-scale multi-copy distributed storage system and application method thereof | |
| CN105426427A (en) | MPP database cluster replica realization method based on RAID 0 storage | |
| WO2018010501A1 (en) | Global transaction identifier (gtid) synchronization method, apparatus and system, and storage medium | |
| EP4213038A1 (en) | Data processing method and apparatus based on distributed storage, device, and medium | |
| WO2016177231A1 (en) | Dual-control-based active-backup switching method and device | |
| CN113987064A (en) | Data processing method, system and equipment | |
| CN105069152A (en) | Data processing method and apparatus | |
| CN110807039A (en) | Data consistency maintenance system and method in cloud computing environment | |
| CN101727496A (en) | Method for realizing load balancing cluster of MICROSOFT SQL SERVER database | |
| US20160100006A1 (en) | Transferring Data Between Sites | |
| WO2014177085A1 (en) | Distributed multicopy data storage method and device | |
| CN114363154A (en) | Node election method and device, electronic equipment and storage medium | |
| US11288237B2 (en) | Distributed file system with thin arbiter node |
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 | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20131204 Assignee: Hundsun Technologies Inc. Assignor: Zhejiang University Contract record no.: 2018330000035 Denomination of invention: Method for synchronizing copies on basis of partition state tables and coordinator nodes Granted publication date: 20160309 License type: Common License Record date: 20180417 |
|
| EE01 | Entry into force of recordation of patent licensing contract |