CN114513525B - Data consistency optimization method and system adopting cross-machine-room chain forwarding - Google Patents
Data consistency optimization method and system adopting cross-machine-room chain forwarding Download PDFInfo
- Publication number
- CN114513525B CN114513525B CN202210407575.XA CN202210407575A CN114513525B CN 114513525 B CN114513525 B CN 114513525B CN 202210407575 A CN202210407575 A CN 202210407575A CN 114513525 B CN114513525 B CN 114513525B
- Authority
- CN
- China
- Prior art keywords
- node
- main
- data
- group
- main node
- 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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1095—Replication or mirroring of data, e.g. scheduling or transport for data synchronisation between network nodes
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/06—Management of faults, events, alarms or notifications
- H04L41/0677—Localisation of faults
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a method and a system for optimizing data consistency by adopting cross-machine-room chain forwarding, belonging to the improvement of database technology. By adopting a cluster deployment grouping mode, the redundant occupation of network resources caused by communication between machine rooms can be saved. Specifically, the grouping management of the data nodes is combined with the writing function of the original auxiliary nodes, so that the network bandwidth of a computer room can be saved, the utilization efficiency of network resources is improved and the load of the Leader node is reduced from the original multiple identical data to the situation that only one data copy needs to be sent. The data node grouping management and the auxiliary node writing function are combined, and part of work of data copying of the first main node is distributed to the second main node to be executed, so that the problem that the first main node is overloaded due to the increase of the cluster scale is solved, and hot spots in the cluster are avoided.
Description
Technical Field
The invention relates to an improvement of a database technology, in particular to a method and a system for optimizing data consistency by adopting cross-machine room chain forwarding.
Background
Data consistency of a distributed database can be understood as transaction consistency and data copy consistency, and here we pay attention to the scenario of data copy consistency, that is, how to guarantee data consistency when data is stored in multiple copies. For data consistency, a consistency protocol is required, common distributed protocols include a two-stage submission protocol, a three-stage submission protocol, Paxos, Raft and the like, and the Raft protocol is mostly adopted to ensure the data consistency in engineering practice. Briefly, the Raft protocol mainly includes two processes: and selecting a Leader and copying the log, wherein the left maintains consistency of the copied log among the nodes by selecting a Leader and endowing the Leader with a function of managing the copied log. The Leader receives the log from the client, copies the log entry to other Follower nodes, and tells other Followerers to apply the log entry to their state machines on the premise of ensuring security. In the process, data flows from the Leader to the Follower in a unidirectional mode, and only after the copying of the nodes exceeding the half number is completed, the data is successfully copied, so that even if a few copies are lost, the latest data in the system can be guaranteed. Taking 5 nodes as an example, as shown in fig. 1, the client sends a request to the master node, the master node adds the request as a log to its log file, and simultaneously sends the log to the slave node in parallel. The secondary node informs the primary node of successful replication after successfully replicating log data, and after receiving most of secondary node responses, the primary node submits logs and sends responses to the client (here, the secondary node 1 and the secondary node 2 both respond to the primary node, 3 total nodes in the whole raft group complete log replication and can submit the logs, and the secondary node 3 and the secondary node 4 cannot respond to the primary node due to faults or cannot influence the submission of the logs due to response delay). In the design of the deployment scheme, the alternative schemes include single-machine room deployment and multi-machine room deployment. Because the availability of the database cluster is determined by the availability of the machine room according to the architectural characteristics of single machine room deployment, multi-machine room deployment is mostly adopted in actual engineering practice, and various deployment schemes such as same city double-activity, different places multiple-activity, two places and three centers and the like can be further divided according to the geographical position of the machine room.
In the design of the distributed database, data can be transmitted to the secondary node from the primary node in a unidirectional mode, and a data copying process does not exist between the secondary nodes. When the cluster scale is increased and the number of copies of data to be synchronized is increased, the number of secondary nodes which need to communicate with the primary node is increased, the occupied bandwidth resource is also increased, and a performance bottleneck is easily formed. Meanwhile, the deployment of the distributed database cluster is mostly multi-machine-room deployment across regions, so that the problems that the network delay exists in the communication between the machine rooms and the positive correlation relationship exists between the data volume transmitted between the machine rooms and the auxiliary nodes, and the bandwidth resources are greatly wasted are solved. There is thus a need for further research and improvement in the design of deployment optimizations for such distributed systems.
Disclosure of Invention
One of the objectives of the present invention is to provide a method and a system for optimizing data consistency by using cross-machine-room chain forwarding, so as to solve the technical problems of bandwidth resource occupation increase and bandwidth resource waste caused by the increase of the number of points of master node communication in the prior art.
In order to solve the technical problems, the invention adopts the following technical scheme:
the invention provides a method for optimizing data consistency by adopting cross-machine room chain forwarding on one hand, and the method comprises the following steps.
And step A, a first main node in the distributed database receives a data request and sends the data request to a first secondary node and a second main node in at least one secondary node group.
And B, the second main node sends the data request to a second auxiliary node in the auxiliary node group where the second main node is located.
And step C, after the first secondary node, the second main node and the second secondary node receive the data request, respectively feeding back a synchronization state to the first main node, thereby finishing the node data synchronization of the distributed database.
Preferably, the further technical scheme is as follows: in step a and step B, the second master node is generated in such a manner that all nodes in the slave node group where the second master node is located send requests to the first master node, the first master node outputs a node group lock to one of the nodes in the current slave node group, and the node that acquires the node group lock becomes the second master node of the current slave node group.
The further technical scheme is as follows: the method also comprises a step D that the second main node and the second auxiliary node are both in contact with the first main node, when the first main node fails to contact the second main node in a unit period, the second auxiliary node in the auxiliary node group where the current second main node is located sends a request to the first main node, the first main node outputs the node group lock again, and the other second auxiliary node acquires the node group lock to become the second main node in the current auxiliary node group.
The invention also provides a data consistency optimization system adopting cross-machine-room chain forwarding, which comprises a first main node and a first auxiliary node which are positioned in a main node group, and a second main node and a second auxiliary node which are positioned in an auxiliary node group, wherein: the first main node is used for receiving a data request in the distributed database and sending the data request to the first secondary node and a second main node in at least one secondary node group; the first secondary node is used for receiving a data request from the first primary node and completing data synchronization; the second main node is used for receiving the data request from the first main node, completing data synchronization and sending the data request to a second secondary node in a secondary node group where the second main node is located; the second secondary node is used for receiving the data request from the second main node and completing data synchronization; the first secondary node, the second primary node and the second secondary node are further used for respectively feeding back synchronization states to the first primary node after receiving the data requests, and therefore node data synchronization of the distributed database is completed.
Preferably, the further technical scheme is as follows: the number of the secondary node groups is multiple, and each secondary node group comprises a second main node and at least two second secondary nodes; and each secondary node group is positioned in a respective machine room and is accessed to the first main node through a network.
The further technical scheme is as follows: the second master node is generated in such a way that all nodes in the slave node group where the second master node is located send requests to the first master node, the first master node outputs a node group lock to one of the nodes in the current slave node group, and the node which acquires the node group lock becomes the second master node of the current slave node group.
The further technical scheme is as follows: and the second main node and the second auxiliary node are both in contact with the first main node, when the first main node fails to contact the second main node in a unit period, the second auxiliary node in the auxiliary node group where the current second main node is located sends a request to the first main node, the first main node outputs the node group lock again, and the other second auxiliary node acquires the node group lock to become the second main node in the current auxiliary node group.
The further technical scheme is as follows: the computer room where the at least one secondary node group is located and the computer room where the first main node is located are clustered and deployed in different urban areas.
Compared with the prior art, the invention has the beneficial effects of at least one of the following:
by adopting a cluster deployment grouping mode, the redundant occupation of network resources caused by communication between machine rooms can be saved. Specifically, the grouping management of the data nodes is combined with the writing function of the original auxiliary nodes, so that the network bandwidth of a computer room can be saved, and the utilization efficiency of network resources is improved from the original multiple identical data to the situation that only one data copy needs to be sent.
And reducing the load of the Leader node. The data node grouping management and the auxiliary node writing function are combined, and part of work of data copying of the first main node is distributed to the second main node to be executed, so that the problem that the first main node is overloaded due to the increase of the cluster scale is solved, and hot spots in the cluster are avoided.
Data copy delay is reduced. Each node in the secondary node group can automatically identify a fault node, and dynamically switch the second primary node according to the node availability without manual participation, thereby avoiding the delay of data replication caused by node faults.
Drawings
FIG. 1 is a flow diagram of prior art Raft log replication.
Fig. 2 is a schematic diagram of a Raft node distribution used to illustrate one embodiment of the invention.
FIG. 3 is a flow chart illustrating the replication of the Raft log in one embodiment of the invention.
Fig. 4 is a flow chart for illustrating the feedback data synchronization of the Raft node in one embodiment of the invention.
Fig. 5 is a handover flow diagram for illustrating the failure of the secondary master node in one embodiment of the present invention.
Detailed Description
The invention is further elucidated with reference to the drawing.
The method is characterized in that a concept of a secondary node group is added on the basis of a Raft protocol, the secondary node group is generated in the secondary nodes of the Raft protocol according to a certain strategy, each secondary node group elects a proxy main node (hereinafter, collectively referred to as a second main node) to be responsible for the synchronous work of data copying of other secondary nodes (hereinafter, collectively referred to as second secondary nodes) in the group, in the data copying process, the main node in the Raft protocol only needs to send data to the secondary nodes of the group and the second main nodes of other secondary node groups, and the data synchronous work of the second secondary nodes in the secondary node group is completed by the second main node of the group.
When the second main node fails, a new second main node is selected from other second secondary nodes in the group, and the data replication work of other nodes in the group is continuously undertaken, namely the second main node can be generated dynamically, and the failure of log replication of the whole secondary node group caused by the failure of the node which plays the role is avoided.
N data nodes are deployed in a distributed database cluster and distributed in machine rooms in different regions. After normal starting, a first main node and N-1 auxiliary nodes are selected through a Raft election process.
Based on the design thought, an embodiment of the present invention is a data consistency optimization node system adopting cross-room chain forwarding, as shown in fig. 2 and 3, after the first master node is determined, N-1 secondary nodes are grouped according to a database cluster topology, and for example, 9 data node clusters are deployed in 3 rooms (room a, room B, and room C) in 2 regions, so that the secondary nodes can be divided into two groups according to the rooms, that is, a secondary node group 1 and a secondary node group 2 shown in fig. 2, and the two groups are located in the room B and the room C, respectively.
Therefore, the system of the present embodiment includes a first primary node and a first secondary node in the primary node group, and a second primary node and a second secondary node in the secondary node group; and the nodes in the machine rooms A, B and C carry out data interaction through the network. The main node group where the first main node is located in the machine room A, the main node group further comprises two first auxiliary nodes, one auxiliary node group is deployed in each of the machine room B and the machine room C, and each auxiliary node group is provided with one second main node and two second auxiliary nodes. Each node in the secondary node group has access to the first primary node through the network.
In the node system of the above structure:
referring to fig. 3, the first master node is responsible for receiving a data request from the distributed database, that is, after the cluster has data written therein, a data replication process is started, and the first master node sends the data to the first slave node and the second master node in the two slave node groups, respectively.
And the first secondary node receives the data request from the first primary node and completes data synchronization.
And the second main node in the two secondary node groups is responsible for receiving the data request from the first main node, completing data synchronization and sending the data request to the second secondary node in the secondary node group where the second main node is located.
And the second secondary node receives the data request from the second primary node and completes data synchronization.
Referring to fig. 4, after receiving the data request, the first secondary node, the second primary node, and the second secondary node respectively feed back synchronization states to the first primary node, thereby completing node data synchronization of the distributed database. After receiving the data, the second secondary nodes in the secondary node group directly feed back to the first main node, the second main node and the second secondary nodes are not distinguished any more, and according to the Raft protocol, the submission of log data can be promoted as long as the number of voted nodes reaches a preset legal number, so that the safety of the log data is ensured.
And when the first main node receives the feedback and counts that most of the auxiliary nodes are completely copied, returning to an upper layer to inform that the data copying is successful, namely completing the data synchronization.
The second master node is generated at will in the node in the slave node group to which the second master node belongs, specifically, all nodes in the slave node group to which the second master node belongs send requests to the first master node, the first master node outputs a node group lock to one of the nodes in the current slave node group, and the node which acquires the node group lock becomes the second master node of the current slave node group.
In the systems shown in fig. 2 to 4, when a node failure fails to respond to a request in time due to a failure, the following cases can be roughly classified.
1) And the first secondary node of the same computer room as the first primary node is abnormal.
2) The second master node in the machine room B is abnormal.
3) The second secondary node in the machine room B is abnormal.
4) All nodes in the machine room B are abnormal.
For exception 1), the log data can be submitted as long as most of other nodes successfully copy the log in the same original Raft protocol exception handling.
For the exception 2), referring to fig. 5, when the original second master node is abnormal, a new second master node may be generated from the remaining second slave nodes, and take over the second master node to perform the log copy operation in the group. Specifically, the second main node and the second secondary node in each secondary node group are both in contact with the first main node, when the first main node fails to contact the second main node in a unit period, the second secondary node in the secondary node group where the current second main node is located sends a request to the first main node, the first main node outputs the node group lock again, and the other second secondary node acquires the node group lock to become the second main node in the current secondary node group.
And aiming at the exception 3), the exception of the second secondary node does not affect other nodes in the group, and log data can be submitted after most of the Raft group finishes log replication.
For exception 4), all nodes of the machine room B are abnormal, and only one secondary node group 2 in the machine room C is divided into groups in the whole Raft group. Firstly, the exception of the machine room B does not affect the nodes of the machine room C, and secondly, as long as most of the nodes still exist in the whole Raft group and can be copied by most of the nodes, the log copying process is normally carried out. For the abnormal condition of all the nodes in the machine room B, the total of five nodes are two first secondary nodes in the machine room A and three nodes in the machine room C, as long as any three nodes finish log replication, and the total of four nodes finish log replication by adding the first primary node in the machine room A, so that the first primary node can finish submission.
On the basis of the four abnormal conditions, the inexhaustible abnormal conditions are processed as the abnormal conditions in the original Raft protocol.
It should be noted that the above idea can also be divided into more sub-node groups on the basis of more data nodes, and the sub-node groups are deployed in more rooms, and the 9 data nodes in this embodiment are only exemplary illustrations of the present invention.
Based on the node system with the above structure, another embodiment of the present invention is a data consistency optimization method using cross-machine-room chain forwarding, which includes and preferably is performed according to the following steps:
step S1, the first master node in the distributed database receives the data request and sends it to the first slave node and the second master node in the at least one slave node group.
And step S2, the second main node sends the data request to a second secondary node in the secondary node group where the second main node is located. In this step, the second master node is generated in such a manner that all nodes in the slave node group where the second master node is located send requests to the first master node, the first master node outputs a node group lock to one of the nodes in the current slave node group, and the node that has acquired the node group lock becomes the second master node of the current slave node group.
And the second main node and the second secondary nodes are both in contact with the first main node, when the second main node is abnormal, namely when the first main node fails to contact the second main node in a unit period, a request is sent to the first main node by the second secondary node in the secondary node group where the current second main node is located, the first main node outputs the node group lock again, and the other second secondary node acquires the node group lock to become the second main node in the current secondary node group.
And step S3, after the first secondary node, the second main node and the second secondary node receive the data request, respectively feeding back the synchronization state to the first main node, thereby completing the node data synchronization of the distributed database.
In addition to the foregoing, it should be noted that reference throughout this specification to "one embodiment," "another embodiment," "an embodiment," or the like, means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment described generally throughout this application. The appearances of the same phrase in various places in the specification are not necessarily all referring to the same embodiment. Further, when a particular feature, structure, or characteristic is described in connection with any embodiment, it is submitted that it is within the scope of the invention to effect such feature, structure, or characteristic in connection with other embodiments.
Although the invention has been described with reference to a number of illustrative embodiments thereof, it should be understood that numerous other modifications and embodiments can be devised by those skilled in the art that will fall within the scope and spirit of the principles of this disclosure. More specifically, various variations and modifications are possible in the component parts and/or arrangements of the subject combination arrangement within the scope of the disclosure, the drawings and the appended claims. In addition to variations and modifications in the component parts and/or arrangements, other uses will also be apparent to those skilled in the art.
Claims (6)
1. A data consistency optimization method adopting cross-machine room chain forwarding is characterized by comprising the following steps:
a first main node in a distributed database receives a data request and sends the data request to a first secondary node and a second main node in at least one secondary node group;
the second main node sends the data request to a second secondary node in a secondary node group where the second main node is located;
after receiving the data requests, the first secondary node, the second primary node and the second secondary node respectively feed back synchronization states to the first primary node, so that node data synchronization of the distributed database is completed;
the second main node and the second secondary nodes are both in contact with the first main node, when the first main node fails to contact the second main node in a unit period, the second main node fails, and a new second main node is selected from other second secondary nodes in the secondary node group;
the first main node, the second auxiliary node, the second main node and the second auxiliary node are Raft nodes.
2. The method for optimizing data consistency by adopting cross-machine room chain forwarding according to claim 1, characterized in that: the generation mode of the second main node is that all nodes in the secondary node group where the second main node is located send requests to the first main node, the first main node outputs a node group lock to one node in the current secondary node group, and the node which acquires the node group lock becomes the second main node of the current secondary node group.
3. A data consistency optimization system adopting cross-machine room chain forwarding is characterized in that the system comprises:
the first main node is used for receiving the data request in the distributed database and sending the data request to the first secondary node and a second main node in at least one secondary node group;
the first secondary node is used for receiving the data request from the first primary node and completing data synchronization;
the second main node is used for receiving the data request from the first main node, completing data synchronization and sending the data request to a second secondary node in a secondary node group where the second main node is located;
the second secondary node is used for receiving the data request from the second main node and completing data synchronization;
the first secondary node, the second primary node and the second secondary node are further used for respectively feeding back synchronization states to the first primary node after receiving the data requests, so that node data synchronization of the distributed database is completed;
the second main node and the second secondary node are both in contact with the first main node, when the first main node fails to contact the second main node in a unit period, the second main node fails, and a new second main node is selected from other second secondary nodes in the secondary node group;
the first main node, the second auxiliary node, the second main node and the second auxiliary node are Raft nodes.
4. The system according to claim 3, wherein the system comprises: the number of the secondary node groups is multiple, and each secondary node group comprises a second main node and at least two second secondary nodes; and each secondary node group is positioned in a respective machine room and is accessed to the first main node through a network.
5. The system according to claim 3, wherein the data consistency optimization system using cross-room chain forwarding is characterized in that: the second master node is generated in such a way that all nodes in the slave node group where the second master node is located send requests to the first master node, the first master node outputs a node group lock to one of the nodes in the current slave node group, and the node which acquires the node group lock becomes the second master node of the current slave node group.
6. The system according to claim 3, wherein the data consistency optimization system using cross-room chain forwarding is characterized in that: the computer room where the at least one secondary node group is located and the computer room where the first main node is located are clustered and deployed in different urban areas.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210407575.XA CN114513525B (en) | 2022-04-19 | 2022-04-19 | Data consistency optimization method and system adopting cross-machine-room chain forwarding |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210407575.XA CN114513525B (en) | 2022-04-19 | 2022-04-19 | Data consistency optimization method and system adopting cross-machine-room chain forwarding |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114513525A CN114513525A (en) | 2022-05-17 |
| CN114513525B true CN114513525B (en) | 2022-07-05 |
Family
ID=81554732
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210407575.XA Active CN114513525B (en) | 2022-04-19 | 2022-04-19 | Data consistency optimization method and system adopting cross-machine-room chain forwarding |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114513525B (en) |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102025756A (en) * | 2009-09-09 | 2011-04-20 | 中兴通讯股份有限公司 | Distributed system and data synchronization method thereof |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8572031B2 (en) * | 2010-12-23 | 2013-10-29 | Mongodb, Inc. | Method and apparatus for maintaining replica sets |
| CN107295080B (en) * | 2017-06-19 | 2020-12-18 | 北京百度网讯科技有限公司 | Data storage method applied to distributed server cluster and server |
| CN109828722B (en) * | 2019-01-29 | 2022-01-28 | 中国人民大学 | Self-adaptive distribution method for Raft group data of heterogeneous distributed key value storage system |
| CN110474986B (en) * | 2019-08-20 | 2022-08-12 | 深圳前海微众银行股份有限公司 | Consensus method, device and system based on block chain system |
| WO2021147926A1 (en) * | 2020-01-20 | 2021-07-29 | Huawei Technologies Co., Ltd. | Methods and systems for hybrid edge replication |
| CN113127565A (en) * | 2021-04-28 | 2021-07-16 | 联通沃音乐文化有限公司 | Method and device for synchronizing distributed database nodes based on external observer group |
| CN113472566A (en) * | 2021-06-11 | 2021-10-01 | 北京市大数据中心 | Status monitoring method of union block chain and master node status monitoring system |
| CN113821563A (en) * | 2021-08-11 | 2021-12-21 | 贝格迈思(深圳)科技有限公司 | Asymmetric multi-copy distributed transaction processing method and system |
-
2022
- 2022-04-19 CN CN202210407575.XA patent/CN114513525B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102025756A (en) * | 2009-09-09 | 2011-04-20 | 中兴通讯股份有限公司 | Distributed system and data synchronization method thereof |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114513525A (en) | 2022-05-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US6977908B2 (en) | Method and apparatus for discovering computer systems in a distributed multi-system cluster | |
| CN102404390B (en) | Intelligent dynamic load balancing method for high-speed real-time database | |
| WO2016070375A1 (en) | Distributed storage replication system and method | |
| EP2643771B1 (en) | Real time database system | |
| JP5015965B2 (en) | Server management system and method | |
| CN109388481B (en) | Transaction information transmission method, system, device, computing equipment and medium | |
| JP2007518195A (en) | Cluster database using remote data mirroring | |
| CN114461438A (en) | Distributed database disaster recovery system and method of asymmetric center mode | |
| WO2014177085A1 (en) | Distributed multicopy data storage method and device | |
| CN114124650A (en) | Master-slave deployment method of SPTN (shortest Path bridging) network controller | |
| CN111158949A (en) | Configuration method, switching method and device of disaster recovery architecture, equipment and storage medium | |
| Dustdar et al. | Dynamic replication and synchronization of web services for high availability in mobile ad-hoc networks | |
| CN1452348A (en) | Method for detecting problem in route protocol module of distribution router | |
| CN109165122B (en) | Method for improving disaster recovery capability of application system same city multi-park deployment realized based on block chain technology | |
| CN114513525B (en) | Data consistency optimization method and system adopting cross-machine-room chain forwarding | |
| JP3139884B2 (en) | Multi-element processing system | |
| CN117061512A (en) | Distributed information management method and system based on big data | |
| CN114564340B (en) | High availability method for distributed software of aerospace ground system | |
| CN114039978B (en) | Decentralized PoW computing power cluster deployment method | |
| CN114363350A (en) | Service management system and method | |
| JP2001166967A (en) | Method for duplicating data of distributed control system | |
| CN115378557B (en) | Hot standby implementation method, device, system, electronic equipment and storage medium | |
| WO2024021746A1 (en) | Data processing method, communication system, and related device | |
| CN116483904A (en) | Data processing system | |
| CN116319986A (en) | Trusted and efficient cross-region block chain data parallel transfer model based on relay mechanism |
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 |