CN103581307A - Publishing/subscribing system based on clusters and method for guaranteeing reliability of publishing/subscribing system based on clusters - Google Patents
Publishing/subscribing system based on clusters and method for guaranteeing reliability of publishing/subscribing system based on clusters Download PDFInfo
- Publication number
- CN103581307A CN103581307A CN201310488048.7A CN201310488048A CN103581307A CN 103581307 A CN103581307 A CN 103581307A CN 201310488048 A CN201310488048 A CN 201310488048A CN 103581307 A CN103581307 A CN 103581307A
- Authority
- CN
- China
- Prior art keywords
- module
- cluster
- publish
- subscribe
- message
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Abstract
The invention discloses a publishing/subscribing system based on clusters and a method for guaranteeing the reliability of the publishing/subscribing system based on the clusters. The publishing/subscribing system comprises a client side, a route module, a data module and a manager module. The client side is connected with a publishing/subscribing agent module through JAVA message services and carries out publishing/subscribing operation through the publishing/subscribing agent module; the route module is used for receiving and forwarding publishing/subscribing massages; the data module is used for receiving subscription information and notices from the client side and sending and receiving messages; the manager module is used for recording cluster information, maintaining a tree topology formed by the clusters and querying and configuring parameter information of the system. By adopting the method, perfect cluster framework design, a multi-priority feedback queue based on a Bloom Filter and manager dual-system hot backup can be achieved, and therefore the high reliability of the system is guaranteed.
Description
Technical field
The present invention relates to network interconnection system and event-driven technology, relate in particular to a kind of publish/subscribe system and reliability guarantee method thereof based on cluster, comprise how organizational structures message minute priority forwards, and manager backs up the high available technology that guarantees.
Background technology
Network is increasing to the influence on development of modem computer systems, but the connectedness that network interconnection system day by day increases and scale have also been brought a series of problem.Just set up and safeguard with regard to these systems, key again basic requirement is to have the reliable stability of a system.This mechanism not only needs to support a large amount of assemblies, also will have the ability to face dynamically and the complicated applied environment often changing.The importance that another one need to be considered is the automation of data processing.System traditionally is generally designed to response user's interactive requirement, and in order to improve the functional and practicality of system, just should provide more autonomous data-handling capacity.
Web service notice (WSN, Web Service Notification) is a set of standard for publish/subscribe system, and it has defined the Web service standard that the publish/subscribe pattern by using based on theme is notified.Subscriber sends subscribe request to the message producer, and content comprises address, its interested notification categories and other relevant information of subscriber; The message producer, after receiving this request, subscribes to resource for it creates one, and returns to the address of service of this resource to it.When certain moment, when the message producer produces the notice of this subscription of coupling, it just sends to this notice to notify accordingly consumer.But the message mode of this Web service normalized definition is difficult to expansion among large-scale publish/subscribe system.Suppose to have M consumer, N the producer, and each consumer needs the notice of all producers' generations, has just produced the notice transmission network of a M * N.
WSN standard has proposed the message pass-through mode based on agency in its another one standard < < WS-Brokered Notification > >, under this scene, subscriber sends subscribe request to agency, notify the producer that notice is all sent to agency, by agency's judgement, described notice is transmitted to which consumer.Under which, producers and consumers does not need to know the other side's existence, thereby has played well the effect of decoupling zero.
Yet, be conceived to design one act on behalf of organizational form and notice routing mode, make the publish/subscribe system based on WSN standard notice can be routed to consumer from the producer, under this framework, although can be successfully completed tissue and notice route, but this organizational form exists defect aspect the reliability of system, cannot tackle the problems such as manager is surprisingly collapsed, system is out of touch, forwarding messages data volume increases substantially, partial data delay sensitive.Also exist and may cause that forward efficiency is low, message is blocked up, the significantly even reliability secret worry of system crash of message delay.
Summary of the invention
In view of this, main purpose of the present invention is to provide a kind of publish/subscribe system and reliability guarantee method thereof based on cluster, to realize the design of perfect aggregated structure, multipriority feedback queue and manager's two-node cluster hot backup based on Bloom filter (Bloom Filter), ensure the high reliability of this system, mainly comprise configuration and the design of cluster, how to design the multilevel feedback queue based on Bloom filter (Bloom Filter), and principal and subordinate manager how safeguards system safeguard several aspects such as normal.
For achieving the above object, technical scheme of the present invention is achieved in that
A publish/subscribe system based on cluster, comprises a plurality of clusters, also comprises client, routing module, data module and manager's module; Wherein,
Client, is connected by JAVA messenger service with publish/subscribe proxy module, by described publish/subscribe proxy module, carries out publish/subscribe operation;
Routing module, for carrying out reception and the forwarding of publish/subscribe message;
Data module, for receiving subscription and the notice from client, and carries out the sending and receiving of message;
Manager's module, for record clustering information, safeguards the tree topology being formed by cluster, the parameter information of inquiry and configuration-system.
Whether wherein, described routing module further comprises detection sub-module, send detect-message detect other that be connected with self and act on behalf of to exist and write to each other by this detection sub-module.
Described routing module also comprises routing table, is used to publish/subscribe notice to provide and sends or forward-path.
Described routing module further comprises subscription table, by JAVA database, is connected and is connected with database, completes the persistence of subscription information.
Described manager's module comprises master administrator module and backup management person's module of redundancy backup relation each other, and described master administrator module and backup management person's module are shared root node information and system configuration information.
Data module forwards the described message forwarding by the multipriority queue based on Bloom filter, described in each, queue has priority and timeslice separately.
A reliability guarantee method for the publish/subscribe system of cluster, the method comprises:
A, normal hosts in same local area network (LAN) is formed to cluster, the normal hosts in cluster is communicated by multicast mode, a cluster representative is set in each cluster, cluster representative communicates by publish/subscribe proxy module and other clusters;
B, the multipriority queue of use based on Bloom filter receive and forward publish/subscribe message and notification message, and the described message of the requirement of different priorities is carried out to classification processing;
The master administrator module that C, setting are connected with root node and backup management person's module, and adopt two-node cluster hot backup mechanism to carry out work.
Wherein, while utilizing cluster described in steps A to communicate, the agency in cluster can find whether the other side goes offline by detection; If cluster representative goes offline, start the operation of a gravity treatment representative.
Multipriority queue described in step B, each queue all has priority and timeslice separately, if timeslice is finished in queuing message repeating process, is switched to next priority query and forwards.
Described in step C, master administrator module and backup management person's module, obtain groups information by root node, described groups information with message structure regularly between described master administrator module and backup management person's module synchronously to keep up-to-date.
Publish/subscribe system and reliability guarantee method thereof based on cluster provided by the present invention, has the following advantages:
The present invention divides by residing local area network (LAN) participating in the agency of publish/subscribe, the agency who is in same local area network (LAN) is formed to a cluster, and get a name for cluster, in the agency of cluster inside, select an agency as the representative of cluster, between cluster and cluster, by representative, communicate simultaneously.Cluster will send registration request to manager while producing, and then from root node, starts to find insertion point, joins among tree-shaped topology.Various message in system, determine regular priority of giving processing according to the characteristic information of its setting and priority.This process utilizes Bloom Filter to carry out instantiation to rule, thereby formation rule determines engine, utilize message data classifying rules to carry out message data instantiation according to the characteristic information, then the data of instantiation are sent into Bloom Filter rule and determined engine, the priority that Auto-matching is different.Then different priorities message is put into corresponding priority query, each queue has the timeslice of oneself, from high priority to low priority, forwards successively corresponding message, the timeslice of each queue oneself is finished, and all will surrender the power of forwarding to next priority query.The present invention is on the basis of said system structure, provided routing module for the forwarding Reliability Assurance problem of big data quantity different priorities data and the solution of manager's module inefficacy backup problem, realized and having subscribed to and the route of the high efficient and reliable of notice, and there is message and forward the advantages such as high efficient and reliable, load balancing, system operator be reliable and stable.
Accompanying drawing explanation
Fig. 1 is aggregated structure schematic diagram of the present invention;
Fig. 2 is the structural representation of publish/subscribe system of the present invention;
Fig. 3 is the routing forwarding schematic diagram of message;
Fig. 4 is that master administrator and backup management person back up schematic diagram;
Fig. 5 is manager's backup messages content schematic diagram;
Fig. 6 is the hot standby switching schematic diagram of manager;
Fig. 7 multipriority queue and Bloom filter (Bloom Filter) structural representation;
Fig. 8 is Bloom filter (Bloom Filter) principle schematic;
Fig. 9 is that priority determines regular instantiation flow chart;
Figure 10 is priority deciding step flow chart.
[critical piece symbol description]
1: cluster
2: cluster representative
3: normal hosts
4: client
5: the forward-path of notice
6: the cluster that needs notice
7: the cluster that produces notice.
Embodiment
Below in conjunction with accompanying drawing and embodiments of the invention, system of the present invention and reliability guarantee method thereof are described in further detail.
Basic thought of the present invention is: the proxy in network is divided into cluster 1 one by one, agency in cluster 1 (by publish/subscribe proxy module) subscribes to and the forwarding of notification message by the mode of multicast, between cluster, by cluster, represent that 2 communicate, and form the form of " tree ", subscribe message between cluster forwards in this " tree ", forward notice and according to the forwarding tree of notice, carry out, and by one independently manager carry out the maintenance of cluster information.Active and standby manager is set to two-node cluster hot backup, can automatically switch, to ensure the stability of this system.Message repeating process adopts the multipriority queue based on Bloom filter (Bloom Filter), and the message classification that different priorities is not required is processed, thereby maintain whole message, forwards reliability and high efficiency.
Fig. 1 is aggregated structure schematic diagram of the present invention, the structural representation that Fig. 2 is publish/subscribe system of the present invention.As shown in Figure 1 and Figure 2, client 4 is connected by JAVA messenger service (JMS) with publish/subscribe proxy module, carries out the operation of publish/subscribe by publish/subscribe proxy module.And the agency's (normal hosts 3) who is in same local area network (LAN) forms a cluster 1, and select an agency to represent 2 as cluster, between each agency, by multicast, intercom mutually.Between cluster 1, by cluster, represent that 2 communicate in point-to-point mode.Manager safeguards the essential information of cluster 1, and guides cluster to form one tree, whenever there being new node to require to add the fashionable root node that returns, for its request, inserts, and manager safeguards the information of this tree.
Cluster is a kind of method to agency's (normal hosts 3) tissue, and it forms a cluster 1 by the agency who is positioned within same local area network (LAN), and the agency of described cluster 1 inside communicates by the form of multicast.In cluster 1, there is an agency to represent that as cluster 2 communicate with other cluster.Cluster represents that 2 is automatically to select, and in representative goes offline, between agency, can be detected and be found that cluster represent that 2 have lost by " heartbeat ", can use in gravity treatment mechanism cluster, to produce a new cluster and represent.
The agency who is in same local area network (LAN) can form a cluster or a plurality of cluster, but agency of a certain moment can only be among a cluster, and cluster can be got a name in order to identify oneself.Cluster communicates by representative and other cluster, and cluster can abstractly be a proxy, and this main frame to be exactly cluster represent 2.So cluster represents the 2 necessary information of safeguarding other agencies in current cluster, this realizes by sending " heartbeat " message, and " heartbeat " is a kind of type of message of definition, is used for communicating by letter to know existing of the other side mutually between two nodes.Namely common agency is to representing that timed sending " heartbeat " message is to show existing of oneself, and cluster representative is also acted on behalf of timed sending " heartbeat " message to show existing of oneself to other.So represent the angle of management cluster home agent from cluster, this is a star-like topology.
Fig. 2 is the structural representation of publish/subscribe system of the present invention.As shown in Figure 2, this publish/subscribe system mainly comprises client, routing module, data module and manager's module.Wherein,
Routing module, take node as unit, for carrying out reception and the forwarding of publish/subscribe message.The subscription information that mainly the form maintenance system by table receives (be no matter from client or from other agency's), and when system is notified (be no matter client issue or from other agency) according to these notices of subscription table route.
Here, described routing module comprises routing table and subscription table, also comprise " heartbeat " detection sub-module, by this detection sub-module transmission " heartbeat " detect-message, detect other agencies that are connected with self and whether also exist and write to each other, and according to whether there being the processing of being correlated with.Described routing table, is used to publish/subscribe notice to provide and sends or forward-path.Described subscription table connects (JDBC, Java Data Base Connectivity) by JAVA database and is connected with database, completes the persistence of subscription information.
Data module, for receiving subscription and the notice from client, and carries out the sending and receiving of message.
Here, described data module comprises Web service notice (WSN) submodule, for completing the related communication of publish/subscribe.
Manager's module, for record clustering information, safeguards the tree topology being formed by cluster, the parameter information of inquiry and configuration-system.
Fig. 3 is the routing forwarding schematic diagram of message.As shown in Figure 3, A, B, C, H, I are for needing the cluster 6 of notice; D, E, F, G are for producing the cluster 7 of notice, and the cluster 7 of described generation notice sends to described notice by forward-path 5 cluster 6 that needs notice.
As, A, B, C, H, I have subscribed to respectively the message of a certain theme, form a binary tree, and E has produced the notification message of corresponding this theme, so first notification message is transmitted to B and A from E, then by B, is transmitted to I, and is transmitted to C and H by A.
Master administrator is as a module independently, and framework, outside ServiceMix-WSN system, is served the routing module of system, and routing module topological structure is managed and inquired about, and facilitates manager to grasp system dynamically, system relevant parameter is configured.
Fig. 4 is that master administrator and backup management person back up schematic diagram.In view of the status of gerentocratic particular importance, master and slave manager's redundancy backup mechanism is set, its master and slave relation can be exchanged, backup mutually, in case when master administrator is collapsed, can obtain the information of master administrator from manager, replace its position to proceed management and the inquiry of topological structure.
First, backup management person is set, the heartbeat forming between master administrator and backup management person is mutual.By User Datagram Protoco (UDP) (UDP) message, send heartbeat to the other side, and upgrade within a certain period of time heart time, know that surpassing heartbeat threshold value does not also receive heartbeat, carries out heartbeat crash handling.
Fig. 5 is manager's backup messages content schematic diagram.As shown in Figure 5, the information that manager need to obtain is cluster family (groups, in system all group) information, and these information can both obtain by root node, in order to dwindle synchronizing information amount, therefore synchronous root node information only.These information with message structure regularly between master administrator and backup management person synchronously to keep up-to-date.
Fig. 6 is the hot standby switching schematic diagram of manager.As shown in Figure 6, when finding that " heartbeat " information of a side is wherein overtime, can think that this manager has lost.Backup management person can read root node information from backup information, and to root node, sends the message of manager's change and master administrator inefficacy, and root node is forwarded to network by failure management person's information, is delivered to a side of subscribe message, waits for maintenance process.Failure management person after handling can re-establish heartbeat and synchronous with current master administrator again, as backup management person.Cycling backup, guarantees that system maintenance is stable.
Fig. 7 multipriority queue and Bloom filter (Bloom Filter) structural representation.As shown in Figure 7, the message that multipriority queue forwards needs is divided into k priority, and a corresponding k priority query, is respectively Q
1, Q
2..., Q
k.Each queue has priority and the timeslice of oneself, if timeslice is finished in each queuing message repeating process, is switched to next priority query and forwards.
Fig. 8 is Bloom filter (Bloom Filter) principle schematic.The fast finding algorithm of the uncommon Function Mapping in a kind of Doha that described Bloom filter proposes in 1970 based on Howard Bloom, this algorithm can judge that certain element is whether outside a set very fast.This detection only can be to the data misjudgement in set, and can be to not being that data in set are misjudged, and each detects request and has returned to " in set (possible errors) " and " not within gathering (anything but within gathering) " two kinds of situations like this.All position zero, safeguard k different hash function, and cryptographic Hash is floated between 1 ~ m, for each, belongs to the element of this priority, all can be by Hash k time, and position assignment corresponding to cryptographic Hash is 1.
Such as checking an element y
1whether belong to this priority, check so all h
i(y
1) whether be all 1, as shown in Figure 8.
Fig. 9 is that priority determines regular instantiation flow chart.Rule instantiation process is converted into Event Priority decision process exactly and judges whether an element belongs to certain set, change abstract priority rule into concrete regular example and form Bloom Filter, as shown in Figure 9, thereby the regular example generating forms through k hash function respectively or upgrades Bloom Filter, finally forms k priority.
In priority decision process, the characteristic information of input forms example, imports Bloom Filter, and after Hash k time, whether comparison result is hit, and namely whether belongs to current priority.As shown in figure 10.
The above, be only preferred embodiment of the present invention, is not intended to limit protection scope of the present invention.
Claims (10)
1. the publish/subscribe system based on cluster, comprises a plurality of clusters, it is characterized in that, this system mainly comprises client, routing module, data module and manager's module; Wherein,
Client, is connected by JAVA messenger service with publish/subscribe proxy module, by described publish/subscribe proxy module, carries out publish/subscribe operation;
Routing module, for carrying out reception and the forwarding of publish/subscribe message;
Data module, for receiving subscription and the notice from client, and carries out the sending and receiving of message;
Manager's module, for record clustering information, safeguards the tree topology being formed by cluster, the parameter information of inquiry and configuration-system.
2. whether the publish/subscribe system based on cluster according to claim 1, is characterized in that, described routing module further comprises detection sub-module, send detect-message detect other that be connected with self and act on behalf of to exist and write to each other by this detection sub-module.
3. the publish/subscribe system based on cluster according to claim 2, is characterized in that, described routing module also comprises routing table, is used to publish/subscribe notice to provide and sends or forward-path.
4. the publish/subscribe system based on cluster according to claim 2, is characterized in that, described routing module further comprises subscription table, by JAVA database, is connected and is connected with database, completes the persistence of subscription information.
5. the publish/subscribe system based on cluster according to claim 1, it is characterized in that, described manager's module comprises master administrator module and backup management person's module of redundancy backup relation each other, and described master administrator module and backup management person's module are shared root node information and system configuration information.
6. the publish/subscribe system based on cluster according to claim 1, is characterized in that, data module forwards the described message forwarding by the multipriority queue based on Bloom filter, and described in each, queue has priority and timeslice separately.
7. a reliability guarantee method for the publish/subscribe system based on cluster, is characterized in that, the method comprises:
A, normal hosts in same local area network (LAN) is formed to cluster, the normal hosts in cluster is communicated by multicast mode, a cluster representative is set in each cluster, cluster representative communicates by publish/subscribe proxy module and other clusters;
B, the multipriority queue of use based on Bloom filter receive and forward publish/subscribe message and notification message, and the described message of the requirement of different priorities is carried out to classification processing;
The master administrator module that C, setting are connected with root node and backup management person's module, and adopt two-node cluster hot backup mechanism to carry out work.
8. the reliability guarantee method of the publish/subscribe system based on cluster according to claim 7, is characterized in that, while utilizing cluster described in steps A to communicate, the agency in cluster can find whether the other side goes offline by detection; If cluster representative goes offline, start the operation of a gravity treatment representative.
9. the reliability guarantee method of the publish/subscribe system based on cluster according to claim 7, it is characterized in that, multipriority queue described in step B, each queue all has priority and timeslice separately, if timeslice is finished in queuing message repeating process, is switched to next priority query and forwards.
10. the reliability guarantee method of the publish/subscribe system based on cluster according to claim 7, it is characterized in that, master administrator module and backup management person's module described in step C, by root node, obtain groups information, described groups information with message structure regularly between described master administrator module and backup management person's module synchronously to keep up-to-date.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310488048.7A CN103581307A (en) | 2013-10-17 | 2013-10-17 | Publishing/subscribing system based on clusters and method for guaranteeing reliability of publishing/subscribing system based on clusters |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310488048.7A CN103581307A (en) | 2013-10-17 | 2013-10-17 | Publishing/subscribing system based on clusters and method for guaranteeing reliability of publishing/subscribing system based on clusters |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103581307A true CN103581307A (en) | 2014-02-12 |
Family
ID=50052215
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310488048.7A Pending CN103581307A (en) | 2013-10-17 | 2013-10-17 | Publishing/subscribing system based on clusters and method for guaranteeing reliability of publishing/subscribing system based on clusters |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103581307A (en) |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015117342A1 (en) * | 2014-07-18 | 2015-08-13 | 中兴通讯股份有限公司 | Notification message sending method, device and system |
| CN105208133A (en) * | 2015-10-20 | 2015-12-30 | 上海斐讯数据通信技术有限公司 | Server, load balancer as well as server load balancing method and system |
| CN105306559A (en) * | 2015-10-12 | 2016-02-03 | 国电南瑞科技股份有限公司 | Multi-partition message control system without central partition for regional power grid regulation and control system |
| CN105897827A (en) * | 2015-11-27 | 2016-08-24 | 乐视云计算有限公司 | Server node, local area network server cluster and realizing method thereof |
| CN106209468A (en) * | 2016-07-21 | 2016-12-07 | 广东电网有限责任公司中山供电局 | The method and system of synchronizing information are operated between the active and standby dispatch automated system of isomery |
| CN107222534A (en) * | 2017-05-25 | 2017-09-29 | 努比亚技术有限公司 | A kind of cluster timed task synchronous method, system and equipment |
| CN108199912A (en) * | 2017-12-15 | 2018-06-22 | 北京奇艺世纪科技有限公司 | Management, consuming method and the device of a kind of strange land distributed message mostly living |
| CN108197138A (en) * | 2017-11-21 | 2018-06-22 | 北京邮电大学 | The method and system for the matching subscription information that releases news in publish/subscribe system |
| CN108494815A (en) * | 2018-02-06 | 2018-09-04 | 北京极智感科技有限公司 | A kind of method and system of transmission message |
| CN110034983A (en) * | 2019-03-08 | 2019-07-19 | 佛山市云米电器科技有限公司 | Home equipment interlock method and system based on Publish-subscribe message processing facility |
| CN111488318A (en) * | 2020-04-16 | 2020-08-04 | 北京雷石天地电子技术有限公司 | File synchronization method and device based on multistage feedback queue |
| CN112637265A (en) * | 2020-11-25 | 2021-04-09 | 新华三技术有限公司 | Equipment management method, device and storage medium |
| CN112769885A (en) * | 2019-11-04 | 2021-05-07 | 千寻位置网络有限公司 | Proxy broadcasting method and device based on data point subscription |
| CN114338063A (en) * | 2020-09-29 | 2022-04-12 | 中移物联网有限公司 | Message queue system, service processing method, and computer-readable storage medium |
| WO2024026838A1 (en) * | 2022-08-05 | 2024-02-08 | Oppo广东移动通信有限公司 | Communication method, apparatus and device, and storage medium, chip, product and program |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102355413A (en) * | 2011-08-26 | 2012-02-15 | 北京邮电大学 | Method and system for unifying message space on large scale in real time |
| CN102984174A (en) * | 2012-12-21 | 2013-03-20 | 北京邮电大学 | Method and system for reliability guarantee in publish-subscribe system |
| CN103037000A (en) * | 2012-12-21 | 2013-04-10 | 北京邮电大学 | Publish-subscribe system and notification message processing method thereof |
-
2013
- 2013-10-17 CN CN201310488048.7A patent/CN103581307A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102355413A (en) * | 2011-08-26 | 2012-02-15 | 北京邮电大学 | Method and system for unifying message space on large scale in real time |
| CN102984174A (en) * | 2012-12-21 | 2013-03-20 | 北京邮电大学 | Method and system for reliability guarantee in publish-subscribe system |
| CN103037000A (en) * | 2012-12-21 | 2013-04-10 | 北京邮电大学 | Publish-subscribe system and notification message processing method thereof |
Non-Patent Citations (1)
| Title |
|---|
| 石瑞生 等: "EDSOA服务平台中发布订阅网络基础服务的设计", 《计算机集成制造系统》, vol. 18, no. 8, 15 August 2012 (2012-08-15), pages 1659 - 1666 * |
Cited By (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015117342A1 (en) * | 2014-07-18 | 2015-08-13 | 中兴通讯股份有限公司 | Notification message sending method, device and system |
| CN105306559B (en) * | 2015-10-12 | 2018-12-04 | 国电南瑞科技股份有限公司 | Non-stop layer subregion multi partition message control system for regional power grid regulator control system |
| CN105306559A (en) * | 2015-10-12 | 2016-02-03 | 国电南瑞科技股份有限公司 | Multi-partition message control system without central partition for regional power grid regulation and control system |
| CN105208133A (en) * | 2015-10-20 | 2015-12-30 | 上海斐讯数据通信技术有限公司 | Server, load balancer as well as server load balancing method and system |
| CN105208133B (en) * | 2015-10-20 | 2018-05-25 | 上海斐讯数据通信技术有限公司 | A kind of server, load equalizer and server load balancing method and system |
| CN105897827A (en) * | 2015-11-27 | 2016-08-24 | 乐视云计算有限公司 | Server node, local area network server cluster and realizing method thereof |
| CN106209468A (en) * | 2016-07-21 | 2016-12-07 | 广东电网有限责任公司中山供电局 | The method and system of synchronizing information are operated between the active and standby dispatch automated system of isomery |
| CN106209468B (en) * | 2016-07-21 | 2019-10-29 | 广东电网有限责任公司中山供电局 | The synchronous method and system of operation information between the active and standby dispatch automated system of isomery |
| CN107222534A (en) * | 2017-05-25 | 2017-09-29 | 努比亚技术有限公司 | A kind of cluster timed task synchronous method, system and equipment |
| CN107222534B (en) * | 2017-05-25 | 2020-11-17 | 太仓市昊博技术服务有限公司 | Cluster timing task synchronization method, system and equipment |
| CN108197138A (en) * | 2017-11-21 | 2018-06-22 | 北京邮电大学 | The method and system for the matching subscription information that releases news in publish/subscribe system |
| CN108199912A (en) * | 2017-12-15 | 2018-06-22 | 北京奇艺世纪科技有限公司 | Management, consuming method and the device of a kind of strange land distributed message mostly living |
| CN108199912B (en) * | 2017-12-15 | 2020-09-22 | 北京奇艺世纪科技有限公司 | Method and device for managing and consuming distributed messages of multiple activities in different places |
| CN108494815A (en) * | 2018-02-06 | 2018-09-04 | 北京极智感科技有限公司 | A kind of method and system of transmission message |
| CN110034983A (en) * | 2019-03-08 | 2019-07-19 | 佛山市云米电器科技有限公司 | Home equipment interlock method and system based on Publish-subscribe message processing facility |
| CN110034983B (en) * | 2019-03-08 | 2020-12-29 | 佛山市云米电器科技有限公司 | Home equipment linkage method and system based on publish-subscribe message processing mechanism |
| CN112769885A (en) * | 2019-11-04 | 2021-05-07 | 千寻位置网络有限公司 | Proxy broadcasting method and device based on data point subscription |
| CN111488318A (en) * | 2020-04-16 | 2020-08-04 | 北京雷石天地电子技术有限公司 | File synchronization method and device based on multistage feedback queue |
| CN114338063A (en) * | 2020-09-29 | 2022-04-12 | 中移物联网有限公司 | Message queue system, service processing method, and computer-readable storage medium |
| CN114338063B (en) * | 2020-09-29 | 2023-11-10 | 中移物联网有限公司 | Message queue system, service processing method and computer readable storage medium |
| CN112637265A (en) * | 2020-11-25 | 2021-04-09 | 新华三技术有限公司 | Equipment management method, device and storage medium |
| CN112637265B (en) * | 2020-11-25 | 2022-07-12 | 新华三技术有限公司 | Equipment management method, device and storage medium |
| WO2024026838A1 (en) * | 2022-08-05 | 2024-02-08 | Oppo广东移动通信有限公司 | Communication method, apparatus and device, and storage medium, chip, product and program |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103581307A (en) | Publishing/subscribing system based on clusters and method for guaranteeing reliability of publishing/subscribing system based on clusters | |
| CN102355413B (en) | A kind of method of extensive unified message space in real time and system thereof | |
| US9021065B2 (en) | Automated topology formation in dynamic distributed environments | |
| Pietzuch et al. | Hermes: A distributed event-based middleware architecture | |
| CN102404390B (en) | Intelligent dynamic load balancing method for high-speed real-time database | |
| Van Renesse et al. | Willow: DHT, aggregation, and publish/subscribe in one protocol | |
| JP2008533564A (en) | Method and apparatus for data management | |
| CN110071873A (en) | A kind of method, apparatus and relevant device sending data | |
| CN102209041A (en) | Scheduling method, device and system | |
| CN102984174B (en) | Reliability guarantee method and system in a kind of distribution subscription system | |
| CN110677282A (en) | Hot backup method of distributed system and distributed system | |
| CN102843432B (en) | Network communication system applicable to sharing and synchronous updating of data at multiple clients | |
| CN114189444A (en) | Method for managing industrial end equipment, time-sensitive network controller and system | |
| EP2645261B1 (en) | Management apparatus, management system, management method and set of an application source program, a first program and a second program | |
| CN101420327B (en) | Method for avoiding routing oscillation in smooth restarting process and smooth restarting equipment | |
| CN102118266A (en) | Industrial Ethernet dual-link redundancy technology | |
| Wang et al. | Reaching agreement among virtual subnets in hybrid failure mode | |
| CN103368693B (en) | A kind of method of power distribution network interexchange bus reliable data transmission | |
| CN113055461B (en) | ZooKeeper-based unmanned cluster distributed cooperative command control method | |
| CN102647424B (en) | Data transmission method and data transmission device | |
| Allawi et al. | Cost-effective topology design for HSR resilient mesh networks | |
| Jaeger et al. | Reconfiguring self-stabilizing publish/subscribe systems | |
| EP2669808A1 (en) | Management device, management method, and management program | |
| CN113518126A (en) | Cross fault-tolerant method for alliance chain | |
| Marzal et al. | Efficient event notification middleware for smart microgrids over P2P networks |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20140212 |
|
| RJ01 | Rejection of invention patent application after publication |