CN102521297A - Method for achieving system dynamic expansion in shared-nothing database cluster - Google Patents
Method for achieving system dynamic expansion in shared-nothing database cluster Download PDFInfo
- Publication number
- CN102521297A CN102521297A CN2011103909309A CN201110390930A CN102521297A CN 102521297 A CN102521297 A CN 102521297A CN 2011103909309 A CN2011103909309 A CN 2011103909309A CN 201110390930 A CN201110390930 A CN 201110390930A CN 102521297 A CN102521297 A CN 102521297A
- Authority
- CN
- China
- Prior art keywords
- data
- back end
- tables
- shared
- newly
- 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
Images
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Multi Processors (AREA)
Abstract
The invention discloses a method for achieving system dynamic expansion in a shared-nothing database cluster, which includes the following steps: adding and detecting newly added data nodes and verifying connectivity of the newly added data nodes; initializing the newly added data nodes and synchronizing metadata information into the newly added data nodes; redistributing related tuples of a data table and distributed on the newly added data nodes in data nodes and the newly added data nodes, and leading operation on the data table in the redistributing process to be synchronous in the data nodes and the newly added data nodes; and modifying cluster configuration files in a cluster manager. The method solves the problem of difficulty in dynamic expansion of the shared-nothing database cluster, and greatly reduces influence on and loss of user service logic during expansion of the cluster. By means of the method, on-line expansion can be achieved, so that high expandability of a system can be ensured, and high availability of the system can further be ensured.
Description
Technical field
The present invention relates to a kind of data-base cluster extended method, relate in particular to a kind ofly in do not have sharing (Share Nothing) data-base cluster, the system dynamics extended method that operation realizes based on tables of data belongs to the data base cluster system technical field.
Background technology
Extensibility is to investigate one of any data-base cluster quality whether key property, and it has determined whether data-base cluster can satisfy the needs of the database application of quick growth.Current, the mode of two kinds of extending database clusters of existence: extends perpendicular and horizontal extension.
1. extends perpendicular
Add resource with the increase capacity at same logic unit.Such as the CPU of upgrade server, such as on the RAID/SAN memory device, increasing hard disk.
2. horizontal extension
Increase a plurality of logical block resources and they done as a whole in work.The horizontal extension mode does not also require expensive server, uses common memory device and PC promptly to can be used as solution.
At present, in database field two kinds of data-base clusters are arranged: a kind of is not have (Share Nothing) data-base cluster of sharing, and another kind is shared disk (Share Disk) data-base cluster.Along with the rise of development of database and cloud computing technology, the shared-nothing database cluster obtains widespread use with advantages such as its concurrency are good in the cloud computing field.Under cloud computing environment, new demand has been proposed for the extensibility of shared-nothing database cluster: increase and reduce node as required.In other words; Increase and reduce the part that node is the service of shared-nothing database cluster routine; Dynamically enlarge and to dwindle the scale of shared-nothing database cluster more frequent, and require in carrying out operating process, serve and to stop, can not having influence on the normal operation of business.
For the shared disk data-base cluster, realize that extensibility is fairly simple, need not move or Update Information, and for the shared-nothing database cluster, the implementation procedure relative complex of extensibility.
In the system extension process of shared-nothing database cluster, need move in a large number or Update Information.The each shared-nothing database cluster is expanded all also might lock whole tables of data or even entire database in migration or when Updating Information, cause user's service disconnection, so need be paid very big cost.For example in the shared-nothing database cluster, data are normally stored by certain algorithm burst.Slicing algorithm commonly used comprises: hash algorithm, random algorithm, round robin algorithm, range algorithm.Owing to increase and the minimizing node, cause slicing algorithm that variation has taken place.For data can correctly be stored under new algorithm, make the performance element on upper strata can correctly locate corresponding data, just need data in the shared-nothing database cluster, to redistribute.Along with the increase of data volume, the operation that this is normally very consuming time.
In addition, in the shared-nothing database cluster, add new node, need restart whole shared-nothing database cluster service.During this period, the shared-nothing database cluster can not provide service to application program.If in the traffic handling time section, this can the serious availability that reduces whole shared-nothing database group system.
Summary of the invention
Technical matters to be solved by this invention is to provide a kind of method that in the shared-nothing database cluster, realizes the system dynamics expansion.
For realizing above-mentioned goal of the invention, the present invention adopts following technical scheme:
Realize the method for system dynamics expansion in a kind of shared-nothing database cluster, said shared-nothing database cluster comprises a plurality of back end by cluster manager dual system control, stores tables of data in the said back end respectively, it is characterized in that may further comprise the steps,
Step 1: increase and the newly-increased back end of detection, verify the connectedness of said newly-increased back end;
Step 2: the said newly-increased back end of initialization is synchronized to metadata information in the said newly-increased back end;
Step 3: will with the relevant tuple of tables of data that distributes on the said newly-increased back end; In said back end and said newly-increased back end, redistribute; And in said redistribution process,, need in said back end and said newly-increased back end, carry out synchronously the operation of said tables of data;
Step 4: revise the cluster configuration file in the said cluster manager dual system.
Wherein more excellently, in the step 3, in said redistribution process, when a tuple was being duplicated, to any operation of said tuple of duplicating, needs just can carry out after waiting for and duplicating end.
Wherein more excellently, when redistributing in the step 3, in said back end, set up temporary table, said temporary table is the copy data table of the tables of data of storing in the said back end,
In said redistribution process,, need the tables of data of storing in the said tables of data of storing in the said back end of synchronous operation, said temporary table and the said newly-increased back end to the operation of said tables of data.
,,, each tuple is copied to respectively in the temporary table of specific data node wherein more excellently, perhaps copy in the tables of data of said newly-increased back end according to the Hash distribution mode for the tables of data of storing in the said back end.
Wherein more excellently,, change distribution mode, copy to respectively in the tables of data of said newly-increased back end according to wheel with the unnecessary data tuple of said back end.
Wherein more excellently, cluster manager dual system described in the step 4 locks the tables of data of each said back end after redistribution is accomplished; Replace with the said temporary table of corresponding data node again, and the lock on the said tables of data is discharged.
Wherein more excellently, redistribution described in the step 3 may further comprise the steps when duplicating distribution mode:
Select one with the redistribution the relevant load minimum data node of tables of data, the tables of data of the said minimum data node of sequential scanning copies to its tuple in the tables of data of said newly-increased back end.
Wherein more excellently, the relevant operation of tables of data of any and said redistribution between replicative phase all will be handled the tables of data of said newly-increased back end and the tables of data of said minimum data node.
Wherein more excellently, in the step 3, said back end is notified said cluster manager dual system after said redistribution is accomplished, and in the step 4, said cluster manager dual system is after receiving said notice, and the redistribution of flag data is accomplished.
The present invention utilizes the tables of data operation in the database to solve the difficult problem of shared-nothing database cluster dynamic extensibility cleverly, greatly reduces cluster influence and loss to the customer service logic in carrying out expansion process.Utilize this method to can be implemented in the line dilatation, both guaranteed the enhanced scalability of system, guaranteed the high availability of system again.
Description of drawings
Below in conjunction with accompanying drawing and embodiment the present invention is done further detailed description.
Fig. 1 is the configuration diagram of a typical shared-nothing database group system;
Fig. 2 is when increasing back end newly, the synoptic diagram of data redistribution in the tables of data;
Fig. 3 is a tables of data when duplicating distribution mode, the synoptic diagram of data redistribution in the tables of data;
Fig. 4 is a tables of data when being other distribution modes, the synoptic diagram of data redistribution in the tables of data.
Embodiment
Fig. 1 has shown the basic framework of a typical shared-nothing database group system.This shared-nothing database group system comprises telegon, cluster manager dual system and a plurality of back end.Wherein, telegon is responsible for introducing the SQL statement that database application is sent, generated query plan; Distribution Strategy according to tables of data; Find the back end relevant, SQL statement is rewritten as a plurality of correlatives and sends on each back end, the statement after going to carry out these and rewrite by back end with this SQL statement; Execution and return results are given telegon then, and telegon returns to database application with the result at last again.
In the shared-nothing database cluster, realize the key of system dynamics expansion is how not realize tables of data is redistributed according to new data distribution algorithms under the traffic affecting prerequisite.For this reason, the present invention has adopted a kind of technical solution based on the tables of data operation.Specifically, when tables of data in when redistribution, earlier data of database is redistributed in the mirror image data table, any operation to changing in data migration process is carried out in this tables of data and mirror image data table simultaneously.After data migtation is accomplished, lock this tables of data, with mirror image data table RNTO by migration table.Launch detailed explanation below.
Usually, in the shared-nothing database cluster, adopt MPP (MPP) framework, can just can accomplish the redistribution of data at short notice, and then realize system extension fast.In MPP (MPP) framework, increase memory capacity and the processing power that node just can effectively improve data base cluster system.
The expansion of shared-nothing database cluster comprises the expansion of data node and the expansion of telegon.Wherein, the expansion of back end is made up of following four-stage:
1. increase and detect and increase back end newly, the checking connectedness
In the time of in the middle of the keeper adds newly-increased back end to the shared-nothing database cluster; Cluster manager dual system can be connected with newly-increased back end according to the profile information of newly-increased back end, verifies whether this back end can carry out interconnected with existing shared-nothing database cluster.
2. initialization data node, synchronizing metadata information
After the checking of cluster manager dual system through a last step, confirmed that newly-increased back end can add in the shared-nothing database cluster.Cluster manager dual system can be created data directory for it, and the metamessage in the cluster is synchronized to newly-increased back end.At this moment, newly-increased back end just formally adds in the shared-nothing database cluster.
3. the redistribution of table data
After newly-increased back end is increased to the shared-nothing database cluster because the adding of this back end, in the former shared-nothing database cluster with the back end that increases newly on the relevant tuple of tables of data that distributes will redistribute based on the distribution mode of tables of data.
Referring to shown in Figure 2, the shared-nothing database cluster is that the tables of data of different distributions mode provides different redistribution modes, specifies as follows:
Hash distributes: the shared-nothing database cluster redistributes on all back end (comprising newly-increased back end) the tuple in the legacy data table based on new hash function;
Scope distributes: newly-increased back end can not exert an influence to the tables of data that scope distributes, and only when existing distribution mode being made amendment or tables of data increased a partition information, the redistribution of the tuple in the tables of data can take place;
Wheel changes distribution: the shared-nothing database cluster can be evenly distributed on all back end (comprising newly-increased back end) the tuple in the tables of data;
Duplicate distribution: the shared-nothing database cluster can copy to the tuple in the tables of data in the mirror image data table on the newly-increased back end.
4. modification cluster configuration file
After above-mentioned three phases was accomplished, cluster manager dual system can upgrade cluster configuration file, and newly-increased back end is accomplished at this point.
Below, be that example realizes that to utilizing the present invention the process of data redistribution describes with single tables of data.The situation of multidata table is similar with it, does not repeat them here.
When the distribution mode of tables of data distributed for duplicating, the process of data redistribution was as shown in Figure 3, comprises following step:
1) duplicate distribution: select a load minimum data nodes X (legacy data node) relevant with this tables of data (Table), this tables of data of sequential scanning copies to tuple in the mirror image data table on the newly-increased back end;
2) all to carry out handled in any relevant issued transaction that takes place between replicative phase to mirror image data table on the newly-increased back end and the tables of data on the back end X with this tables of data;
3) in order to guarantee the correctness of reproduction process; Tuple need guarantee the atomicity of reproduction process when newly-increased back end duplicates from back end X; Promptly when other affairs will be operated this tuple, could operate this tuple after must duplicating end by the time.So just guaranteed that tuple does not have other affairs to operate on it in reproduction process; In other words; To guarantee that exactly each transaction operation to the tables of data on the data nodes X all will be applied to the tables of data on the newly-increased back end simultaneously in reproduction process, thereby guarantee the consistance of affairs;
4) after scanning is accomplished to the tables of data on the data nodes X (when promptly can not find the tuple of not duplicating), send EOF (end of file) instruction, show that the data redistribution of tables of data is accomplished this moment to newly-increased back end; Back end X can notify the cluster manager dual system task to accomplish simultaneously.
For other distribution mode, the process of data redistribution is as shown in Figure 4, comprises following step:
1) in order before increasing the back end completion, to continue to make the shared-nothing database cluster service to be provided to application program; The shared-nothing database cluster is to original node data node 1 ... When the tuple of the tables of data on the back end n redistributes, need go up a temporary table (Temp Table) of copy data table at the related data node (legacy data node) of this tables of data;
2) tuple of sequential scanning local data table on each legacy data node is duplicated the 3rd of distribution and is gone on foot above reproduction process is similar, but different and different according to the distribution mode of tables of data:
For Hash distributed, each back end copied to new Hash distribution mode local data according to minute Boulez in the tables of data of relevant temporary table or new data node of specified node with tuple;
For wheel changeed distribution, according to the mean allocation principle, the data tuple that each node is unnecessary copied in the tables of data of new data node.
The tuple of duplicating there is any operation, need just can carries out after the reproduction process end of this tuple of wait.
3), then to operate the tables of data of new data node, the tables of data and the temporary table of legacy data node simultaneously this moment if application program has operation to this tables of data in the tuple transition process of tables of data;
4) when to data node 1 ... After tables of data scanning on the back end n is accomplished (when promptly can not find the tuple of not duplicating), show that the tuple distribution of tables of data is accomplished this moment; Each node can notify the redistribution of cluster manager dual system data to accomplish;
5) after cluster manager dual system is received the signal of redistribution completion, can the former tables of data on each node be locked, this moment, all can only be waited in line the affairs of data table handling; Former tables of data will be replaced by temporary table simultaneously, and this temporary table becomes new tables of data and the lock on it is discharged.This process has atomicity equally.
6) after cluster manager dual system was received and accomplished signal, the redistribution of flag data was accomplished.
Through above-mentioned step, can effective online dilatation means be provided for the shared-nothing database cluster, both guaranteed the enhanced scalability of shared-nothing database group system, guaranteed the high availability of shared-nothing database group system again.
More than the method that realizes the system dynamics expansion in the shared-nothing database cluster provided by the present invention has been carried out detailed explanation.To those skilled in the art, any conspicuous change of under the prerequisite that does not deviate from connotation of the present invention, it being done all will constitute to infringement of patent right of the present invention, with corresponding legal responsibilities.
Claims (9)
1. realize the method that system dynamics is expanded in a shared-nothing database cluster, said shared-nothing database cluster comprises a plurality of back end by cluster manager dual system control, stores tables of data in the said back end respectively, it is characterized in that may further comprise the steps:
Step 1: increase and the newly-increased back end of detection, verify the connectedness of said newly-increased back end;
Step 2: the said newly-increased back end of initialization is synchronized to metadata information in the said newly-increased back end;
Step 3: will with the relevant tuple of tables of data that distributes on the said newly-increased back end; In said back end and said newly-increased back end, redistribute; And in said redistribution process,, need in said back end and said newly-increased back end, carry out synchronously the operation of said tables of data;
Step 4: revise the cluster configuration file in the said cluster manager dual system.
2. realize the method for system dynamics expansion in the shared-nothing database cluster as claimed in claim 1, it is characterized in that:
In the said step 3, in said redistribution process, when a tuple was being duplicated, to any operation of said tuple of duplicating, needs just can carry out after waiting for and duplicating end.
3. realize the method for system dynamics expansion in the shared-nothing database cluster as claimed in claim 1, it is characterized in that:
When redistributing in the said step 3, in said back end, set up temporary table, said temporary table is the copy data table of the tables of data of storing in the said back end,
In said redistribution process,, need the tables of data of storing in the said tables of data of storing in the said back end of synchronous operation, said temporary table and the said newly-increased back end to the operation of said tables of data.
4. realize the method for system dynamics expansion in the shared-nothing database cluster as claimed in claim 3, it is characterized in that:
For the tables of data of storing in the said back end, according to the Hash distribution mode, each tuple is copied to respectively in the temporary table of specific data node, perhaps copy in the tables of data of said newly-increased back end.
5. realize the method for system dynamics expansion in the shared-nothing database cluster as claimed in claim 3, it is characterized in that:
With the unnecessary data tuple of said back end, change distribution mode according to wheel, copy to respectively in the tables of data of said newly-increased back end.
6. realize the method for system dynamics expansion in the shared-nothing database cluster as claimed in claim 5, it is characterized in that:
Cluster manager dual system described in the said step 4 locks the tables of data of each said back end after redistribution is accomplished; Replace with the said temporary table of corresponding data node again, and the lock on the said tables of data is discharged.
7. realize the method for system dynamics expansion in the shared-nothing database cluster as claimed in claim 1, it is characterized in that:
Redistribution may further comprise the steps when duplicating distribution mode in the said step 3:
Select one with the redistribution the relevant load minimum data node of tables of data, the tables of data of the said minimum data node of sequential scanning copies to its tuple in the tables of data of said newly-increased back end.
8. realize the method for system dynamics expansion in the shared-nothing database cluster as claimed in claim 7, it is characterized in that:
The relevant operation of tables of data any and said redistribution between replicative phase all will be handled the tables of data of said newly-increased back end and the tables of data of said minimum data node.
9. realize the method for system dynamics expansion in the shared-nothing database cluster as claimed in claim 1, it is characterized in that:
In the said step 3, said back end is notified said cluster manager dual system after said redistribution is accomplished, and in the said step 4, said cluster manager dual system is after receiving said notice, and the redistribution of flag data is accomplished.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110390930.9A CN102521297B (en) | 2011-11-30 | 2011-11-30 | The method of achieving system dynamic expansion in shared-nothing database cluster |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110390930.9A CN102521297B (en) | 2011-11-30 | 2011-11-30 | The method of achieving system dynamic expansion in shared-nothing database cluster |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102521297A true CN102521297A (en) | 2012-06-27 |
| CN102521297B CN102521297B (en) | 2015-09-09 |
Family
ID=46292218
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110390930.9A Active CN102521297B (en) | 2011-11-30 | 2011-11-30 | The method of achieving system dynamic expansion in shared-nothing database cluster |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102521297B (en) |
Cited By (25)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102968503A (en) * | 2012-12-10 | 2013-03-13 | 曙光信息产业(北京)有限公司 | Data processing method for database system, and database system |
| CN103106286A (en) * | 2013-03-04 | 2013-05-15 | 曙光信息产业(北京)有限公司 | Method and device for managing metadata |
| CN103944964A (en) * | 2014-03-27 | 2014-07-23 | 上海云数信息科技有限公司 | Distributed system and method carrying out expansion step by step through same |
| CN104035944A (en) * | 2013-03-08 | 2014-09-10 | 中兴通讯股份有限公司 | File system attribute synchronous control method, device and system |
| CN104378427A (en) * | 2014-11-14 | 2015-02-25 | 浪潮电子信息产业股份有限公司 | Method for keeping consistency of cluster key data |
| CN104484469A (en) * | 2014-12-31 | 2015-04-01 | 天津南大通用数据技术股份有限公司 | Expansion method for supporting non-halt of multi-hash map database cluster system |
| CN104615657A (en) * | 2014-12-31 | 2015-05-13 | 天津南大通用数据技术股份有限公司 | Expanding and shrinking method for distributed cluster with nodes supporting multiple data fragments |
| WO2015087237A1 (en) * | 2013-12-13 | 2015-06-18 | International Business Machines Corporation | Incremental and collocated redistribution for expansion of online shared nothing database |
| CN105069109A (en) * | 2015-08-07 | 2015-11-18 | 北京思特奇信息技术股份有限公司 | Dilatation method and system of distributed database |
| CN105095417A (en) * | 2015-07-13 | 2015-11-25 | 小米科技有限责任公司 | Method and apparatus for storing business data and database node |
| CN105260376A (en) * | 2015-08-17 | 2016-01-20 | 北京京东尚科信息技术有限公司 | Method, equipment and system used for cluster node contraction and expansion |
| WO2016116052A1 (en) * | 2015-01-21 | 2016-07-28 | Huawei Technologies Co., Ltd. | System and method for massively parallel processor database |
| CN105991769A (en) * | 2016-04-12 | 2016-10-05 | 上海帝联信息科技股份有限公司 | Data migration method and device and edge node server |
| CN106301938A (en) * | 2016-08-25 | 2017-01-04 | 成都索贝数码科技股份有限公司 | A kind of high availability and the data base cluster system of strong consistency and node administration method thereof |
| WO2017005191A1 (en) * | 2015-07-08 | 2017-01-12 | Huawei Technologies Co., Ltd. | System and method for data caching in processing nodes of massively parallel processing (mpp) database system |
| WO2017177800A1 (en) * | 2016-04-15 | 2017-10-19 | 中兴通讯股份有限公司 | Automatic expansion method and system for solr cluster, and computer storage medium |
| CN107870954A (en) * | 2016-09-28 | 2018-04-03 | 中国移动通信集团广东有限公司 | The extended method and device of a kind of distributed data base |
| CN108319623A (en) * | 2017-01-18 | 2018-07-24 | 华为技术有限公司 | A kind of fast resampling method, apparatus and data-base cluster |
| CN109525651A (en) * | 2018-10-29 | 2019-03-26 | 中译语通科技股份有限公司 | Application data node dynamic increases method and system |
| CN109587221A (en) * | 2018-11-09 | 2019-04-05 | 平安科技（深圳）有限公司 | Large data sets group managing means, device, storage medium and computer equipment |
| CN110168516A (en) * | 2016-12-09 | 2019-08-23 | 华为技术有限公司 | The dynamic calculate node based on cost for MPP is packet-optimized |
| CN111198782A (en) * | 2018-11-16 | 2020-05-26 | 中国移动通信集团辽宁有限公司 | Data redistribution method, device, equipment and storage medium |
| CN111291112A (en) * | 2018-12-07 | 2020-06-16 | 阿里巴巴集团控股有限公司 | Read-write control method and device for distributed database and electronic equipment |
| CN111723148A (en) * | 2020-04-10 | 2020-09-29 | 浙江大华技术股份有限公司 | Data storage method and device, storage medium and electronic device |
| US11151111B2 (en) | 2017-11-30 | 2021-10-19 | Futurewei Technologies, Inc. | Redistributing table data in a database cluster |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2004051474A2 (en) * | 2002-11-27 | 2004-06-17 | Oracle International Corporation | Clustering system and method having interconnect |
| CN101043398A (en) * | 2006-06-28 | 2007-09-26 | 华为技术有限公司 | Method and system for sharing connection dynamically |
| CN101187931A (en) * | 2007-12-12 | 2008-05-28 | 浙江大学 | Distribution type file system multi-file copy management method |
| US20100061375A1 (en) * | 2006-10-26 | 2010-03-11 | Jinsheng Yang | Network Data Storing System and Data Accessing Method |
| CN102033938A (en) * | 2010-12-10 | 2011-04-27 | 天津神舟通用数据技术有限公司 | Secondary mapping-based cluster dynamic expansion method |
| CN102158540A (en) * | 2011-02-18 | 2011-08-17 | 广州从兴电子开发有限公司 | System and method for realizing distributed database |
-
2011
- 2011-11-30 CN CN201110390930.9A patent/CN102521297B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2004051474A2 (en) * | 2002-11-27 | 2004-06-17 | Oracle International Corporation | Clustering system and method having interconnect |
| CN101043398A (en) * | 2006-06-28 | 2007-09-26 | 华为技术有限公司 | Method and system for sharing connection dynamically |
| US20100061375A1 (en) * | 2006-10-26 | 2010-03-11 | Jinsheng Yang | Network Data Storing System and Data Accessing Method |
| CN101187931A (en) * | 2007-12-12 | 2008-05-28 | 浙江大学 | Distribution type file system multi-file copy management method |
| CN102033938A (en) * | 2010-12-10 | 2011-04-27 | 天津神舟通用数据技术有限公司 | Secondary mapping-based cluster dynamic expansion method |
| CN102158540A (en) * | 2011-02-18 | 2011-08-17 | 广州从兴电子开发有限公司 | System and method for realizing distributed database |
Cited By (46)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102968503B (en) * | 2012-12-10 | 2015-10-07 | 曙光信息产业(北京)有限公司 | The data processing method of Database Systems and Database Systems |
| CN102968503A (en) * | 2012-12-10 | 2013-03-13 | 曙光信息产业(北京)有限公司 | Data processing method for database system, and database system |
| CN103106286A (en) * | 2013-03-04 | 2013-05-15 | 曙光信息产业(北京)有限公司 | Method and device for managing metadata |
| CN103106286B (en) * | 2013-03-04 | 2017-02-01 | 曙光信息产业(北京)有限公司 | Method and device for managing metadata |
| CN104035944A (en) * | 2013-03-08 | 2014-09-10 | 中兴通讯股份有限公司 | File system attribute synchronous control method, device and system |
| WO2014134905A1 (en) * | 2013-03-08 | 2014-09-12 | 中兴通讯股份有限公司 | File system attribute synchronization control method, device and system |
| CN104035944B (en) * | 2013-03-08 | 2018-11-09 | 南京中兴新软件有限责任公司 | Property synchronization control method, the device and system of file system |
| CN105814562B (en) * | 2013-12-13 | 2019-06-04 | 国际商业机器公司 | System and method for the redistribution of data in online shared-nothing database |
| US11341139B2 (en) | 2013-12-13 | 2022-05-24 | International Business Machines Corporation | Incremental and collocated redistribution for expansion of online shared nothing database |
| US10387422B2 (en) | 2013-12-13 | 2019-08-20 | International Business Machines Corporation | Incremental and collocated redistribution for expansion of online shared nothing database |
| CN105814562A (en) * | 2013-12-13 | 2016-07-27 | 国际商业机器公司 | Incremental and collocated redistribution for expansion of online shared nothing database |
| WO2015087237A1 (en) * | 2013-12-13 | 2015-06-18 | International Business Machines Corporation | Incremental and collocated redistribution for expansion of online shared nothing database |
| CN103944964A (en) * | 2014-03-27 | 2014-07-23 | 上海云数信息科技有限公司 | Distributed system and method carrying out expansion step by step through same |
| CN104378427A (en) * | 2014-11-14 | 2015-02-25 | 浪潮电子信息产业股份有限公司 | Method for keeping consistency of cluster key data |
| CN104615657A (en) * | 2014-12-31 | 2015-05-13 | 天津南大通用数据技术股份有限公司 | Expanding and shrinking method for distributed cluster with nodes supporting multiple data fragments |
| CN104484469A (en) * | 2014-12-31 | 2015-04-01 | 天津南大通用数据技术股份有限公司 | Expansion method for supporting non-halt of multi-hash map database cluster system |
| CN104484469B (en) * | 2014-12-31 | 2017-12-15 | 天津南大通用数据技术股份有限公司 | One kind supports more non-stop-machine expansion methods of hash map data base cluster systems |
| WO2016116052A1 (en) * | 2015-01-21 | 2016-07-28 | Huawei Technologies Co., Ltd. | System and method for massively parallel processor database |
| US9959332B2 (en) | 2015-01-21 | 2018-05-01 | Futurewei Technologies, Inc. | System and method for massively parallel processor database |
| WO2017005191A1 (en) * | 2015-07-08 | 2017-01-12 | Huawei Technologies Co., Ltd. | System and method for data caching in processing nodes of massively parallel processing (mpp) database system |
| US9875186B2 (en) | 2015-07-08 | 2018-01-23 | Futurewei Technologies, Inc. | System and method for data caching in processing nodes of a massively parallel processing (MPP) database system |
| CN105095417A (en) * | 2015-07-13 | 2015-11-25 | 小米科技有限责任公司 | Method and apparatus for storing business data and database node |
| CN105069109A (en) * | 2015-08-07 | 2015-11-18 | 北京思特奇信息技术股份有限公司 | Dilatation method and system of distributed database |
| CN105069109B (en) * | 2015-08-07 | 2018-04-03 | 北京思特奇信息技术股份有限公司 | A kind of method and system of distributed data base dilatation |
| CN105260376A (en) * | 2015-08-17 | 2016-01-20 | 北京京东尚科信息技术有限公司 | Method, equipment and system used for cluster node contraction and expansion |
| CN105260376B (en) * | 2015-08-17 | 2018-08-14 | 北京京东尚科信息技术有限公司 | Method, apparatus and system for clustered node reducing and expansion |
| CN105991769B (en) * | 2016-04-12 | 2019-04-16 | 上海帝联信息科技股份有限公司 | Data migration method and device, edge node server |
| CN105991769A (en) * | 2016-04-12 | 2016-10-05 | 上海帝联信息科技股份有限公司 | Data migration method and device and edge node server |
| WO2017177800A1 (en) * | 2016-04-15 | 2017-10-19 | 中兴通讯股份有限公司 | Automatic expansion method and system for solr cluster, and computer storage medium |
| CN106301938A (en) * | 2016-08-25 | 2017-01-04 | 成都索贝数码科技股份有限公司 | A kind of high availability and the data base cluster system of strong consistency and node administration method thereof |
| CN107870954B (en) * | 2016-09-28 | 2021-09-21 | 中国移动通信集团广东有限公司 | Method and device for expanding distributed database |
| CN107870954A (en) * | 2016-09-28 | 2018-04-03 | 中国移动通信集团广东有限公司 | The extended method and device of a kind of distributed data base |
| CN110168516A (en) * | 2016-12-09 | 2019-08-23 | 华为技术有限公司 | The dynamic calculate node based on cost for MPP is packet-optimized |
| CN110168516B (en) * | 2016-12-09 | 2021-09-07 | 华为技术有限公司 | Dynamic computing node grouping method and system for large-scale parallel processing |
| CN108319623A (en) * | 2017-01-18 | 2018-07-24 | 华为技术有限公司 | A kind of fast resampling method, apparatus and data-base cluster |
| WO2018133662A1 (en) * | 2017-01-18 | 2018-07-26 | 华为技术有限公司 | Data redistribution method and apparatus, and database cluster |
| CN108319623B (en) * | 2017-01-18 | 2021-10-22 | 华为技术有限公司 | Data redistribution method and device and database cluster |
| US11726984B2 (en) | 2017-01-18 | 2023-08-15 | Huawei Technologies Co., Ltd. | Data redistribution method and apparatus, and database cluster |
| US11151111B2 (en) | 2017-11-30 | 2021-10-19 | Futurewei Technologies, Inc. | Redistributing table data in a database cluster |
| CN109525651B (en) * | 2018-10-29 | 2021-08-27 | 中译语通科技股份有限公司 | Application program data node dynamic adding method and system |
| CN109525651A (en) * | 2018-10-29 | 2019-03-26 | 中译语通科技股份有限公司 | Application data node dynamic increases method and system |
| CN109587221A (en) * | 2018-11-09 | 2019-04-05 | 平安科技（深圳）有限公司 | Large data sets group managing means, device, storage medium and computer equipment |
| CN111198782A (en) * | 2018-11-16 | 2020-05-26 | 中国移动通信集团辽宁有限公司 | Data redistribution method, device, equipment and storage medium |
| CN111291112A (en) * | 2018-12-07 | 2020-06-16 | 阿里巴巴集团控股有限公司 | Read-write control method and device for distributed database and electronic equipment |
| CN111291112B (en) * | 2018-12-07 | 2023-04-28 | 阿里巴巴集团控股有限公司 | Read-write control method and device for distributed database and electronic equipment |
| CN111723148A (en) * | 2020-04-10 | 2020-09-29 | 浙江大华技术股份有限公司 | Data storage method and device, storage medium and electronic device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102521297B (en) | 2015-09-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102521297A (en) | Method for achieving system dynamic expansion in shared-nothing database cluster | |
| CN107005596B (en) | Replicated database allocation for workload balancing after cluster reconfiguration | |
| US9727273B1 (en) | Scalable clusterwide de-duplication | |
| US9996427B2 (en) | Parallel backup for distributed database system environments | |
| CN113535656B (en) | Data access method, device, equipment and storage medium | |
| US8954391B2 (en) | System and method for supporting transient partition consistency in a distributed data grid | |
| US11093468B1 (en) | Advanced metadata management | |
| CN102855284B (en) | The data managing method of a kind of cluster storage system and system | |
| US10176184B2 (en) | System and method for supporting persistent store versioning and integrity in a distributed data grid | |
| US11263236B2 (en) | Real-time cross-system database replication for hybrid-cloud elastic scaling and high-performance data virtualization | |
| Taranov et al. | Fast and strongly-consistent per-item resilience in key-value stores | |
| Moiz et al. | Database replication: A survey of open source and commercial tools | |
| EP3788489B1 (en) | Data replication in a distributed storage system | |
| CN114424181A (en) | Native persistent storage support for blockchains | |
| US8805797B2 (en) | Optimizing wide area network (WAN) traffic by providing home site deduplication information to a cache site | |
| US9747323B1 (en) | Method for reconstruction of a distributed lock state after a node addition or removal using a consistent hash | |
| Waqas et al. | Transaction management techniques and practices in current cloud computing environments: A survey | |
| CN102937964A (en) | Intelligent data service method based on distributed system | |
| US7069270B1 (en) | Automated method and mechanism for converting a single instance application to a multiple instance application | |
| CN114003657A (en) | Data processing method, system, device and storage medium for distributed database | |
| US8799569B2 (en) | Multiple enhanced catalog sharing (ECS) cache structure for sharing catalogs in a multiprocessor system | |
| US11461201B2 (en) | Cloud architecture for replicated data services | |
| US11449398B2 (en) | Embedded container-based control plane for clustered environment | |
| WO2023065868A1 (en) | Transaction execution method and apparatus, and computer device and storage medium | |
| US20230113180A1 (en) | Methods and systems for expanding gpu memory footprint based on hybrid-memory |
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 |