CN107153594B - HA component main selection method and system of distributed database system - Google Patents
HA component main selection method and system of distributed database system Download PDFInfo
- Publication number
- CN107153594B CN107153594B CN201610124113.1A CN201610124113A CN107153594B CN 107153594 B CN107153594 B CN 107153594B CN 201610124113 A CN201610124113 A CN 201610124113A CN 107153594 B CN107153594 B CN 107153594B
- Authority
- CN
- China
- Prior art keywords
- component
- identifier
- node
- coordination
- module
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3003—Monitoring arrangements specially adapted to the computing system or computing system component being monitored
- G06F11/3006—Monitoring arrangements specially adapted to the computing system or computing system component being monitored where the computing system is distributed, e.g. networked systems, clusters, multiprocessor systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3055—Monitoring arrangements for monitoring the status of the computing system or of the computing system component, e.g. monitoring if the computing system is on, off, available, not available
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/27—Replication, distribution or synchronisation of data between databases or within a distributed database system; Distributed database system architectures therefor
Abstract
The invention discloses a HA component main selection method and a HA component main selection system of a distributed database system, wherein the HA component main selection method comprises the following steps: each HA component sends a write request to the coordination component; the coordination component gives unique identifiers to the HA components respectively according to the writing request, and obtains the associated information of the HA components and the corresponding identifiers; an HA component acquires the association information; the HA component acquires an identifier corresponding to the HA component according to the associated information; and judging whether the identification is a preset main node identification. By introducing a coordination component in the overall architecture of the distributed database, the efficiency of the HA components can be prevented from being reduced by realizing the mutual communication among the HA components through the coordination component; according to the characteristics of the coordination components such as sequence consistency and the like, unique identifiers are given to all the HA components, so that each HA component can determine a unique main node according to the uniqueness of the identifiers, and further, the split brain phenomenon is prevented.
Description
Technical Field
The invention relates to the field of distributed databases, in particular to a HA component selection method and system of a distributed database system.
Background
In the big data era, distributed databases become an essential infrastructure for data storage. Distributed database systems have some problems, of which the usability problem is a fundamental problem for most application services. Providing a certain degree of high availability services is one of the essential features of a distributed database system.
At present, a plurality of high-availability schemes of a distributed database system are adopted, and a single-node high-availability component is adopted for fault detection and fault processing. The scheme has a single point of failure, and when the high-availability component is abnormal, the high-availability service cannot be provided.
For a distributed high availability system with high availability components (HA components) deployed in multiple nodes, a single point problem can be solved. However, the primary-backup relationship must be determined between multiple high-availability component nodes, only the primary node can handle the failure, otherwise, the operations of multiple high-availability nodes are easy to cause conflicts. If a fixed primary and secondary role relationship is adopted, when the primary node is abnormal, other secondary nodes cannot perform fault processing, so that the common scheme adopts a dynamic primary and secondary role relationship. And determining the main node and the standby node through a certain algorithm, and selecting one node from the standby nodes as the main node when the main node is abnormal. The master selection algorithm is usually complex, and when the network is unstable, a split phenomenon easily occurs, that is, more than one master node exists at the same time. Algorithms which can guarantee consistency in theory, such as common paxos algorithms and raft algorithms, are difficult to realize and prone to deadlock.
In the patent publication No. 201410854727.6, a method for solving a 2PC model single point failure problem applied to a distributed database is proposed, which includes: establishing a list of slave coordinator queues, and storing a plurality of sites as slave coordinators, wherein the function of each slave coordinator is the same as that of the coordinator; the coordinator and the participants communicate through data packets according to a two-stage submission protocol; meanwhile, the coordinator and all the slave coordinators carry out communication at intervals; if the slave coordinator does not receive the message sent by the coordinator after the time period, the coordinator is considered to be blocked, and the system immediately selects one slave coordinator from the coordinator queue as a new coordinator to communicate with a plurality of participants according to a certain algorithm. However, this method cannot completely prevent the split brain phenomenon, and the coordinators need to communicate with each other at regular time, which may reduce efficiency.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: a HA component master selection method of a distributed database system and a system thereof are provided, which can determine a unique master node.
In order to solve the technical problems, the invention adopts the technical scheme that: a HA component master method of a distributed database system comprises the following steps:
each HA component sends a write request to the coordination component;
the coordination component gives unique identifiers to the HA components respectively according to the writing request, and obtains the associated information of the HA components and the corresponding identifiers;
an HA component acquires the association information;
the HA component acquires an identifier corresponding to the HA component according to the associated information;
and judging whether the identification is a preset main node identification.
The invention also relates to an HA component selection system of a distributed database system, comprising:
a sending module, configured to send a write request to the coordination component by each HA component;
the assigning module is used for assigning unique identifiers to the HA components respectively by the coordination component according to the writing request and acquiring the associated information of the HA components and the corresponding identifiers;
a first obtaining module, configured to obtain the association information by an HA component;
the second acquisition module is used for acquiring the corresponding identifier of the HA component according to the associated information;
and the judging module is used for judging whether the identification is a preset main node identification.
The invention has the beneficial effects that: by introducing the coordination component into the overall architecture of the distributed database, the mutual communication between the HA components is realized through the coordination component, and the efficiency of the HA components can be improved; according to the characteristics of the coordination components such as sequence consistency and the like, unique identifiers are given to all the HA components, so that each HA component can determine a unique main node according to the uniqueness of the identifiers, and further, the split brain phenomenon is prevented.
Drawings
FIG. 1 is a flow chart of a HA component election method of a distributed database system of the present invention;
fig. 2 is a schematic structural diagram of a distributed database system according to a first embodiment of the present invention;
FIG. 3 is a flowchart of a method according to a first embodiment of the present invention;
FIG. 4 is a schematic structural diagram of an HA component host system of a distributed database system according to the present invention;
fig. 5 is a schematic system structure diagram according to a second embodiment of the present invention.
Description of reference numerals:
1. a database cluster; 2. an HA component cluster; 3. a database agent; 4. a coordination component;
101. a sending module; 102. a endowing module; 103. a first acquisition module; 104. a second acquisition module; 105. a judgment module; 106. a first determination module; 107. a first deletion module; 108. a notification module; 109. selecting a module; 110. a second determination module; 111. and a second deleting module.
Detailed Description
In order to explain technical contents, objects and effects of the present invention in detail, the following detailed description is given with reference to the accompanying drawings in conjunction with the embodiments.
The most key concept of the invention is as follows: and realizing role distribution among the HA components based on the coordination component, and determining a unique main node according to the unique identifier.
Referring to fig. 1, a HA component selection method of a distributed database system includes:
each HA component sends a write request to the coordination component;
the coordination component gives unique identifiers to the HA components respectively according to the writing request, and obtains the associated information of the HA components and the corresponding identifiers;
an HA component acquires the association information;
the HA component acquires an identifier corresponding to the HA component according to the associated information;
and judging whether the identification is a preset main node identification.
From the above description, the beneficial effects of the present invention are: the unique main node can be determined according to the unique identifier, and the split brain phenomenon is prevented.
Further, the write request includes HA component data that uniquely corresponds to the HA component.
As can be seen from the above description, each HA component can uniquely acquire the identifier corresponding to the HA component through the HA component data uniquely corresponding to the HA component, so that a situation that one HA component acquires multiple identifiers is prevented.
Further, the step of the coordination component giving unique identifiers to each HA component according to the write request and acquiring the association information between each HA component and the corresponding identifier thereof specifically includes:
and the coordination component gives a sequence number which is sequentially increased to each HA component according to the time sequence of receiving the write request of each HA component, and acquires the associated information of each HA component and the sequence number.
As can be seen from the above description, according to the characteristics of the coordination component, each HA component corresponds to a non-duplicate serial number, and the unique master node can be quickly determined by determining whether the unique serial number is the serial number preselected as the backup master node.
Further, after the determining whether the identifier is a preset host node identifier, the method further includes:
if the identifier is a preset host node identifier, determining that the HA component is a host node;
if the original main node fails, the coordination component deletes the associated information of the original main node and the corresponding identifier of the original main node;
the coordination component sends the abnormal event notice of the original main node to each HA component;
and selecting the HA component which is determined as the main node as a new main node.
Further, after the determining whether the identifier is a preset host node identifier, the method further includes:
if the identifier is not a preset host node identifier, determining that the HA component is a slave node;
and if the slave node fails, the coordination component deletes the slave node and the associated information of the corresponding identifier of the slave node.
As can be seen from the above description, when a node fails, the coordination component automatically deletes corresponding association information, thereby preventing the failed node from being selected as a master node; if the master node fails, a new master node is selected from the started and normal slave nodes, so that the problem of single-point failure is solved, and the high-availability function can be continuously executed.
Referring to fig. 4, the present invention further provides an HA component selection system of a distributed database system, including:
a sending module, configured to send a write request to the coordination component by each HA component;
the assigning module is used for assigning unique identifiers to the HA components respectively by the coordination component according to the writing request and acquiring the associated information of the HA components and the corresponding identifiers;
a first obtaining module, configured to obtain the association information by an HA component;
the second acquisition module is used for acquiring the corresponding identifier of the HA component according to the associated information;
and the judging module is used for judging whether the identification is a preset main node identification.
Further, the write request includes HA component data that uniquely corresponds to the HA component.
Further, the assigning module is specifically configured to assign, by the coordination module, a sequentially incremented serial number to each HA component according to the time sequence of receiving the write request of each HA component, and obtain association information between each HA component and the serial number thereof.
Further, still include:
the first judgment module is used for judging that the HA component is a main node if the identifier is a preset main node identifier;
the first deleting module is used for deleting the associated information of the original main node and the corresponding identifier if the original main node fails;
the notification module is used for sending the abnormal event notification of the original main node to each HA component by the coordination component;
a selecting module, configured to select the HA component determined as the master node as a new master node.
Further, still include:
the second judgment module is used for judging that the HA component is a slave node if the identifier is not a preset master node identifier;
and the second deleting module is used for deleting the associated information of the slave node and the corresponding identifier if the slave node fails.
Example one
Referring to fig. 1, a first embodiment of the present invention is: referring to fig. 2, the HA component election method of a distributed database system includes a database cluster 1, an HA component cluster 2, a database agent 3, and a coordination component 4. The database cluster 1 is composed of at least one database group, and one database group comprises a main database node and at least one standby database node. The HA components in the HA component cluster 2 provide the high availability functionality of the database cluster 1. The coordination component 4 stores the information of the database cluster 1, including the master-slave information of all database groups and database nodes therein, and actively informs the database agent 3 and the HA component when the information of the database cluster 1 changes. The database agent 3 reads the information of the whole database cluster 1 from the coordination component 4, and routes the request of the client to the corresponding database node according to a certain fragmentation rule. When the information of the database cluster 1 changes, the database proxy 3 is informed by the coordination component 4 of the update information. The HA component reads the database cluster 1 information from the coordination component 4, then periodically performs fault detection on the database nodes, and if a fault occurs, performs corresponding fault processing.
The HA component adopts distributed multi-node deployment, and the single-point problem is avoided. Each HA component HAs two role states, a master node and a slave node. In a distributed HA component cluster 2, only one HA component node is a master node at the same time, and the other HA component nodes are all slave nodes.
The coordination component 4 can be deployed in a multi-node distributed manner, and should have the following characteristics:
the sequence is consistent, and the data are updated according to the sequence of the requests sent by the client;
atomicity, the updating is either successful or failed, and partial updating can not occur;
the uniqueness is that the same view can be seen no matter which node the client is connected with;
reliability, once the data update is successful, the reliability is maintained until a new update is made;
and in time, the client can obtain the latest data within a certain time.
The solid lines in fig. 2 represent data transmission, the dotted lines represent heartbeats, and the dashed lines represent snooping.
The coordination component 4 may be developed autonomously, or may coordinate service software using an open-source distributed application program such as zookeeper, etc. The present embodiment takes zookeeper as the coordination component 4 as an example.
Referring to fig. 3, the method includes the following steps:
s1: starting the HA component; for example, three HA components with IDs of HA _ x, HA _ y, and HA _ z are started in the HA component cluster.
S2: and after the HA component is started, the zookeeper is automatically connected with the coordination component.
S3: each HA component sends a write request to a coordination component, wherein the write request comprises HA component data uniquely corresponding to the HA component; for example, each HA component writes its node ID in the form of a temporary node and a sequential node into the coordination component zookeeper.
S4: the coordination component gives unique identifiers to the HA components respectively according to the writing request, and obtains the associated information of the HA components and the corresponding identifiers; optionally, the coordination component assigns each HA component a sequentially increasing sequence number according to the time sequence of receiving the write request of each HA component, so as to obtain the association information between each HA component and the sequence number thereof; for example, the coordination component zookeeper writes data including the node ID and an incremented sequence number according to the time when the write request was actually received, that is, data that zookeeper finally writes are HA _ x _0001, HA _ y _0002, and HA _ z _ 0003.
S5: each HA component acquires the associated information from the coordination component; the final written data is obtained, for example, from the coordinating component zookeeper.
S6: each HA component acquires an identifier corresponding to the HA component according to the associated information; for example, the HA component having the ID of HA _ x acquires HA _ x _0001, HA _ y _0002, and HA _ z _0003, and compares its own ID with HA _ x _0001, HA _ y _0002, and HA _ z _0003 to obtain its corresponding serial number of 001.
S7: each HA component judges whether the identifier is a preset main node identifier or not; for example, if the preset candidate master node is an HA component with the smallest sequence number, each HA component determines whether the sequence number corresponding to the HA component is the smallest, if so, S8 is executed, and if not, S9 is executed; optionally, the primary node may also be preset as the HA component with the largest sequence number.
S8: and judging that the HA component is a main node, and setting the role of the HA component as the main node to start executing the functions of the main node.
S9: determining that the HA component is a slave node.
S10: all the enabled HA components listen to the coordination component, i.e. both the master and slave nodes listen to the coordination component.
S11: if the node has a fault, judging whether the fault node is a main node, if not, executing S12, and if so, executing S13.
S12: and the coordination component deletes the associated information of the slave nodes and the corresponding identifications thereof.
S13: the coordination component deletes the association information of the master node and the corresponding identification thereof, and sends an event notification of the master node exception to the slave node, and the slave node executes step S5.
The embodiment is based on the sequence consistency characteristic of the coordination component, two HA components corresponding to the same serial number cannot exist at the same time, and therefore it is guaranteed that only one HA component corresponding to the minimum serial number exists, a unique main node is determined, and the split brain phenomenon is prevented.
Example two
Referring to fig. 5, the present embodiment is an HA component host system of a distributed database system corresponding to the method of the first embodiment, and includes:
a sending module 101, configured to send a write request to a coordination component by each HA component;
an assigning module 102, configured to assign unique identifiers to the HA components respectively by the coordination component according to the write request, and obtain association information between each HA component and the identifier corresponding to the HA component; the write request includes HA component data that uniquely corresponds to the HA component. Optionally, the giving module 102 is specifically configured to give, to each HA component, a sequentially incremented serial number according to the time sequence of receiving the write request of each HA component, and obtain association information between each HA component and the serial number thereof.
A first obtaining module 103, configured to obtain the association information by an HA component;
a second obtaining module 104, configured to obtain, by the HA component, an identifier corresponding to the association information according to the association information;
and the judging module 105 is configured to judge whether the identifier is a preset primary node identifier.
The system further comprises:
a first determining module 106, configured to determine that the HA component is a master node if the identifier is a preset master node identifier;
a first deleting module 107, configured to delete, by the coordination component, the associated information of the original primary node and the corresponding identifier thereof if the original primary node fails;
a notification module 108, configured to send an abnormal event notification of the original master node to each HA component by the coordination component;
a selecting module 109, configured to select the HA component determined as the master node as a new master node.
The system also includes
A second determining module 110, configured to determine that the HA component is a slave node if the identifier is not a preset master node identifier;
and a second deleting module 111, configured to delete, if the slave node fails, the associated information of the slave node and the corresponding identifier thereof by the coordinating component.
In summary, according to the HA component selection method and the HA component selection system for the distributed database system provided by the present invention, the coordination component is introduced into the overall architecture of the distributed database, and the coordination component realizes the mutual communication between the HA components, so that the efficiency of the HA component can be improved; through the HA component data uniquely corresponding to the HA components, each HA component can uniquely acquire the corresponding identifier, and the condition that one HA component acquires a plurality of identifiers is prevented; according to the characteristics of sequence consistency and the like of the coordination assemblies, each HA assembly corresponds to a non-repeated serial number, the unique main node can be quickly determined according to the fact that whether the unique serial number is the serial number pre-selected as the standby main node or not is judged, and the split phenomenon is prevented.
The above description is only an embodiment of the present invention, and not intended to limit the scope of the present invention, and all equivalent changes made by using the contents of the present specification and the drawings, or applied directly or indirectly to the related technical fields, are included in the scope of the present invention.
Claims (8)
1. A HA component main selection method of a distributed database system is characterized in that: the method comprises the following steps:
each HA component sends a write request to the coordination component; the HA component is a high availability component;
the coordination component gives unique identifiers to the HA components respectively according to the writing request, and obtains the associated information of the HA components and the corresponding identifiers;
an HA component acquires the association information;
the HA component acquires an identifier corresponding to the HA component according to the associated information;
judging whether the identifier is a preset main node identifier or not;
after the step of determining whether the identifier is the preset host node identifier, the method further includes:
if the identifier is a preset host node identifier, determining that the HA component is a host node;
if the original main node fails, the coordination component deletes the associated information of the original main node and the corresponding identifier of the original main node;
the coordination component sends the abnormal event notice of the original main node to each HA component;
and selecting the HA component which is determined as the main node as a new main node.
2. The HA component hosting method of a distributed database system of claim 1, wherein: the write request includes HA component data that uniquely corresponds to the HA component.
3. The HA component hosting method of a distributed database system of claim 1, wherein: the step of the coordination component giving unique identifiers to each HA component according to the write request and acquiring the associated information of each HA component and the corresponding identifier thereof specifically includes:
and the coordination component gives a sequence number which is sequentially increased to each HA component according to the time sequence of receiving the write request of each HA component, and acquires the associated information of each HA component and the sequence number.
4. The HA component hosting method of a distributed database system of claim 1, wherein: after the step of determining whether the identifier is the preset host node identifier, the method further includes:
if the identifier is not a preset host node identifier, determining that the HA component is a slave node;
and if the slave node fails, the coordination component deletes the slave node and the associated information of the corresponding identifier of the slave node.
5. An HA component election system for a distributed database system, characterized by: the method comprises the following steps:
a sending module, configured to send a write request to the coordination component by each HA component; the HA component is a high availability component;
the assigning module is used for assigning unique identifiers to the HA components respectively by the coordination component according to the writing request and acquiring the associated information of the HA components and the corresponding identifiers;
a first obtaining module, configured to obtain the association information by an HA component;
the second acquisition module is used for acquiring the corresponding identifier of the HA component according to the associated information;
the judging module is used for judging whether the identifier is a preset main node identifier or not;
further comprising:
the first judgment module is used for judging that the HA component is a main node if the identifier is a preset main node identifier;
the first deleting module is used for deleting the associated information of the original main node and the corresponding identifier if the original main node fails;
the notification module is used for sending the abnormal event notification of the original main node to each HA component by the coordination component;
a selecting module, configured to select the HA component determined as the master node as a new master node.
6. The HA component selection system of a distributed database system of claim 5, wherein: the write request includes HA component data that uniquely corresponds to the HA component.
7. The HA component selection system of a distributed database system of claim 5, wherein: the assigning module is specifically configured to assign, by the coordination module, a sequentially increasing serial number to each HA component according to the time sequence of receiving the write request of each HA component, and acquire association information between each HA component and the serial number thereof.
8. The HA component selection system of a distributed database system of claim 5, wherein: further comprising:
the second judgment module is used for judging that the HA component is a slave node if the identifier is not a preset master node identifier;
and the second deleting module is used for deleting the associated information of the slave node and the corresponding identifier if the slave node fails.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610124113.1A CN107153594B (en) | 2016-03-04 | 2016-03-04 | HA component main selection method and system of distributed database system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610124113.1A CN107153594B (en) | 2016-03-04 | 2016-03-04 | HA component main selection method and system of distributed database system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107153594A CN107153594A (en) | 2017-09-12 |
| CN107153594B true CN107153594B (en) | 2020-03-17 |
Family
ID=59792082
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610124113.1A Active CN107153594B (en) | 2016-03-04 | 2016-03-04 | HA component main selection method and system of distributed database system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107153594B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112732674B (en) * | 2021-01-08 | 2022-08-19 | 烽火通信科技股份有限公司 | Cloud platform service management method, device, equipment and readable storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6272491B1 (en) * | 1998-08-24 | 2001-08-07 | Oracle Corporation | Method and system for mastering locks in a multiple server database system |
| US7120690B1 (en) * | 2001-09-27 | 2006-10-10 | Emc Corporation | Managing a distributed directory database |
| CN1976350A (en) * | 2005-10-27 | 2007-06-06 | 国际商业机器公司 | Method and device for distribution computing system |
| US7269648B1 (en) * | 2001-09-27 | 2007-09-11 | Emc Corporation | Resolving multiple master node conflict in a DDB |
| CN102055779A (en) * | 2009-10-30 | 2011-05-11 | 国际商业机器公司 | Method, device and system for generating HA (High Availability) group |
| CN104679907A (en) * | 2015-03-24 | 2015-06-03 | 新余兴邦信息产业有限公司 | Realization method and system for high-availability and high-performance database cluster |
| CN105338028A (en) * | 2014-07-30 | 2016-02-17 | 浙江宇视科技有限公司 | Method and device for selecting master and slave nodes in distributed server cluster |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9087085B2 (en) * | 2012-12-10 | 2015-07-21 | International Business Machines Corporation | Pre-assimilation values and post-assimilation values in hardware instance identifiers |
| US9519510B2 (en) * | 2014-03-31 | 2016-12-13 | Amazon Technologies, Inc. | Atomic writes for multiple-extent operations |
-
2016
- 2016-03-04 CN CN201610124113.1A patent/CN107153594B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6272491B1 (en) * | 1998-08-24 | 2001-08-07 | Oracle Corporation | Method and system for mastering locks in a multiple server database system |
| US7120690B1 (en) * | 2001-09-27 | 2006-10-10 | Emc Corporation | Managing a distributed directory database |
| US7269648B1 (en) * | 2001-09-27 | 2007-09-11 | Emc Corporation | Resolving multiple master node conflict in a DDB |
| CN1976350A (en) * | 2005-10-27 | 2007-06-06 | 国际商业机器公司 | Method and device for distribution computing system |
| CN102055779A (en) * | 2009-10-30 | 2011-05-11 | 国际商业机器公司 | Method, device and system for generating HA (High Availability) group |
| CN105338028A (en) * | 2014-07-30 | 2016-02-17 | 浙江宇视科技有限公司 | Method and device for selecting master and slave nodes in distributed server cluster |
| CN104679907A (en) * | 2015-03-24 | 2015-06-03 | 新余兴邦信息产业有限公司 | Realization method and system for high-availability and high-performance database cluster |
Non-Patent Citations (1)
| Title |
|---|
| zookeeper的选主流程（源码分析）;qinghua9;《https://blog.csdn.net/qinghua9/article/details/22858033》;20140403;第1-8页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107153594A (en) | 2017-09-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107153660B (en) | Fault detection processing method and system for distributed database system | |
| CN109729111B (en) | Method, apparatus and computer program product for managing distributed systems | |
| CN107295080B (en) | Data storage method applied to distributed server cluster and server | |
| US8671151B2 (en) | Maintaining item-to-node mapping information in a distributed system | |
| JP6382454B2 (en) | Distributed storage and replication system and method | |
| US11736371B2 (en) | Heartbeat information sending method and apparatus, and heartbeat sending node | |
| US7139925B2 (en) | System and method for dynamic cluster adjustment to node failures in a distributed data system | |
| US20190075084A1 (en) | Distributed Lock Management Method, Apparatus, and System | |
| JP2017525008A (en) | Arbitration processing method, quorum storage device, and system after cluster brain division | |
| JP3798661B2 (en) | Method for processing a merge request received by a member of a group in a clustered computer system | |
| EP3817338B1 (en) | Method and apparatus for acquiring rpc member information, electronic device and storage medium | |
| AU2015408848A1 (en) | Method for processing acquire lock request and server | |
| CN111259072A (en) | Data synchronization method and device, electronic equipment and computer readable storage medium | |
| WO2022134797A1 (en) | Data fragmentation storage method and apparatus, a computer device, and a storage medium | |
| CN107153595B (en) | Fault detection method and system for distributed database system | |
| CN110971702A (en) | Service calling method and device, computer equipment and storage medium | |
| CN111708668A (en) | Cluster fault processing method and device and electronic equipment | |
| KR20200062889A (en) | Method and Apparatus for performing block agreement in block chain system | |
| CN107153594B (en) | HA component main selection method and system of distributed database system | |
| CN111147312B (en) | Resource allocation management method and device, resource allocation cache management method and device, and allocation management system | |
| JP2010044553A (en) | Data processing method, cluster system, and data processing program | |
| CN112865995B (en) | Distributed master-slave system | |
| CN112631756A (en) | Distributed regulation and control method and device applied to space flight measurement and control software | |
| CN110324262A (en) | A kind of method and device that resource is seized | |
| EP3341853B1 (en) | Message processing node and database in a message processing system and methods of operating the same |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |