CN110149378A - A kind of people's air defense data server cluster based on WLC algorithm - Google Patents
A kind of people's air defense data server cluster based on WLC algorithm Download PDFInfo
- Publication number
- CN110149378A CN110149378A CN201910364337.3A CN201910364337A CN110149378A CN 110149378 A CN110149378 A CN 110149378A CN 201910364337 A CN201910364337 A CN 201910364337A CN 110149378 A CN110149378 A CN 110149378A
- Authority
- CN
- China
- Prior art keywords
- server
- node
- algorithm
- people
- air defense
- 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
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
-
- 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
- H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
Abstract
A kind of people's air defense data server cluster based on WLC algorithm, belongs to data communication technology field.Hardware includes clustered node server, load balance scheduling server and message queue telemetering transport protocol client-server；By six type shielding grid line physical connections for realizing the publication of high-speed data interaction message between these three types of server hardwares.Clustered node server is 2 Dao more when in use, when how many message queue telemetering transport protocol client-server just needs to construct how many a clustered node servers, is communicated between clustered node server using bus communication mode.Multiple software modules are installed in each clustered node server.The advantage is that solving the problems, such as that people's air defense institute data server load capacity and communication imbalance and people's air defense command post's equipment connection communication are ropy；Using low-cost common server, and dispose load balance scheduling device software, so that it may realize the comparable network throughput of performance, while there is flexible scalability, the compatible and very high cost performance of height.
Description
Technical field
The invention belongs to data communication technology field, in particular to a kind of people's air defense data server collection based on WLC algorithm
Group.
Background technique
People's air defense command post is advanced war preparedness place, has very high security level, to the automated communication of equipment
Propose higher requirement.The equipment of people's air defense command post is mainly divided to two classes, and one kind is Informationization Equipment, for commanding troops；Separately
One kind is the guarantee infrastructure device of people's air defense command post, such as ventilation, water supply, distribution.
In order to ensure the normal operation of people's air defense command post, Automatic monitoring systems are equipped in duty room, to command of people's air defense
All devices carry out real-time control and monitoring, mainly include a few class equipment such as wind, water, electricity, door.With common building automatic
Unlike system, people's air defense command post is the autonomous system established one's own system, and complete isolation operation is realized, to guarantee
The safety in wartime.
People's air defense automated system includes three proofings ventilating system, water circulation hot and cold system, distribution electricity generation system, environmental monitoring system
System four is most of, and the equipment communication of each section has unique operation characteristic.Monitored object such as three proofings ventilating system is main
It is valve and blower, it is desirable that real-time is high, safety is timely.Environmental monitoring system number of sensors is more, type is more, it is desirable that compatibility
Good, extension is conveniently.Ventilating system and electricity generation system require reliable and stable.
People's air defense field device and data server between communication mode it is extremely important for data server.Tradition
Communication mode be generally fieldbus mode, including 485 modes, Lon bus, DP bus, CIP bus etc., these are total
Line is proprietary system agreement, can not be compatible, and when realizing data monitoring, data server will adapt to various protocols, installation
A variety of physical interfaces, use are more inconvenient.As Internet of Things is universal, Ethernet gradually occupies the mainstream of equipment communication, it will
All automation equipments are accordingly to be regarded as the node of Internet of Things, using unified data-interface and agreement.Internet of Things communication protocol is answered
With most widely mainly having SIMPlE agreement, XMPP protocol and MQTT agreement, wherein MQTT is obtained with binary light-type agreement
The extensive support of industrial equipment producer is arrived.MQTT is the real-time communication protocol based on publication and subscribing mode, Netowrk tape
Width occupancy is small, and the deployment of all platforms is supported to implement.The communication mechanism of MQTT only needs a proxy server and a client
The live field bus device of people's air defense, is converted to the client of MQTT bus, and easy to implement and price is lower by end.
Existing people's air defense data server generally takes active and standby method to ensure the communication between equipment, this data communication side
Method is referred to as single-point communication, and when data volume is larger or connection equipment is more, data server can not timely respond to processing requirement,
The hardware facility of higher performance server is needed, and when data volume is less, the low carrying row of expensive hardware resource, and will cause
The waste of resource and fund.
Summary of the invention
The purpose of the present invention is to provide a kind of people's air defense data server clusters based on WLC algorithm, solve people's air defense institute
Data server load capacity and communication imbalance problem.Based on current people's air defense device data server architecture it is backward, suitable
The problem of network load ability difference is answered, present invention seek to address that people's air defense command post's equipment communication high concurrent connection communication is ropy
Problem.For people's air defense device data communicate feature, use distributed message distribution mechanism, can be to the task of user
Running priority grade is allocated, and is connected according to the real time load rate equilibrium assignment of server, and the efficiency of message server is improved,
For people's air defense data safety sound assurance is provided.When network load is lighter, system is distributed a small amount of node according to algorithm and is held
Row task, and when size of message or connection number of concurrent sharply increase, user's request is reasonably distributed to each by load balance system
Server node reduces the request response time of user, improves the handling capacity of network data.
As shown in Figure 1, hardware of the present invention includes clustered node server 1, load balance scheduling server 2 and message queue
Telemetering transport protocol client-server 3 is used between these three types of server hardwares by six type shielding grid line physical connections
Realize the publication of high-speed data interaction message.Clustered node server 1 is 2 Dao more when in use, when how many message queue
Telemetering transport protocol client-server 3 just needs to construct how many a clustered node servers 1, between clustered node server 1
It is communicated using bus communication mode.Multiple software modules are installed in each clustered node server 1, software module includes:
As MQTT data communication, major function includes distribution subscription, counts weight for message queue telemetering transport protocol Message Agent service
Send out function, caching etc.；Storage and log services, using Redis database, major function includes data storage, log recording；It adjusts
Agency service is spent, the management and running of more message task publications are realized as Message Agent, using based on nginx algorithm actuator
Dispatcher software, major function are the load balances realized between clustered node server 1；Bus client service, as high speed
Data interaction bus communication management, using the queue service bus service software of RocketMQ, major function be realize node it
Between information realization transmission.Management and running software is installed, software uses nginx conduct in load balance scheduling server 2
Algorithm actuator；And Redis database, Redis build host-guest architecture and carry out state synchronized between each node, thus real
The caching now dispatched with it is synchronous.Each people's air defense field device be used as a message queue telemetering transport protocol client
Server 3.
People's air defense data communication and common building automation have significant difference, it is characterized in that communication equipment
Geographical dispersibility and traffic inhomogeneities.For people's air defense institute, it is ensured that facilities for preparation against war safe operation, it is most important that
The three proofings equipment of people's air defense, including wind, water, electric three parts real time monitoring, followed by commander's safety equipment and other subrings
Border monitoring device etc..The communication of all devices must be timely and effective, if data server cannot respond most urgent most important information,
Serious safety accident or even threat to life safety will be will cause.Traditional server takes performance redundancy generally to guarantee to respond
Timeliness, but network congestion once occurs, server can not processing equipment connection cause system failure.Number proposed by the present invention
Traditional data-direct connection mode is changed to distributed according to service cluster system, thoroughly solves the problems, such as network congestion.
The data of people's air defense take device cluster be for mechanical floor it is transparent, client is it is not necessary that how many current cluster section known
Point, its message will be handled by which node, these work are completed by load balance scheduling device.Load balance scheduling device prison
It surveys the message content of client and the loading condition of clustered node and makes a weight judgement, then dispatch a cluster section
The connection at point customer in response end.Load balance scheduling device there are two types of deployment way, the first, with an individual server node
Scheduling feature is completed, all data connections first pass through the processing of load balance scheduling device node, relay to MQTT agency's clothes
Business device, in this manner, system performance bottleneck is present on this single-point of scheduler；Second is in each clustered node
Scheduling broker service software is all disposed on server, the dispatch state of each node is that height is consistent, each node has solely
The failure of vertical connection forwarding capability, some node will not cause whole collapse.Present invention employs second of dispatching methods.
The administrator of group system can according to need increase or delete the node in cluster, take by load balance scheduling
The reasonable distribution of business device, the operating status of each node are in lighter load level.Common load balancing algorithm has rotation
Method, random hash method, minimum connection method, fastest response method etc..From the point of view of the characteristics of people's air defense institute's equipment communicates, these algorithms are not
It is optimal.By taking most common round robin as an example, all nodes are regarded as a queue and successively divided by being intended that of this algorithm
Hair task, at this moment the number of tasks of each node is almost identical.But problem is, although the number of tasks of each node is identical,
But the number of topics and message processing time for including in each connection are different, and server cluster still will appear load tilts
Situation.The invention proposes a kind of WLC (the Weighted Least-Connection based on time response mechanism
Scheduling) algorithmic dispatching mechanism, WLC algorithm are a kind of dynamic weight assignment algorithms, it considers server in cluster
The CPU usage of node, memory percentage, network bandwidth amount actual use situation, with the method for weight to all nodes into
Row is lined up, and the availability highest of this node of weighted value minimum specification, on the basis of WLC algorithm, a kind of time is added in the present invention
The mechanism of response records the time of this message connection processing that is, after the completion of each connection response, and is synchronized to each node
On interior dispatch list, according to this record, chosen in fixed weight distribution basis when load dispatch server is allocated again
One most fast node of response is handled.The treatment process of loads-scheduling algorithm is as shown in Figure 3.
People's air defense data server cluster proposed by the present invention is not needed using high-performance, the huge service of high-cost multicore
Device, it is only necessary to use low-cost common server, and dispose load balance scheduling device software, so that it may realize that performance is comparable
Network throughput, while there is flexible scalability, the compatible and very high cost performance of height.
Detailed description of the invention
Fig. 1 people's air defense data server cluster architecture diagram.
The internal structure chart of Fig. 2 scheduling broker service.
Fig. 3 loads-scheduling algorithm procedure chart.
One instance graph of Fig. 4 people's air defense data server cluster.
Specific embodiment
The present invention provides a kind of people's air defense data server cluster systems, to make the purpose of the present invention, technical solution and work
More accurate clear with effect, comparison attached drawing is further analyzed and is illustrated, the research people as this profession by part below
Member is it should be appreciated that attached drawing and embodiment proposed by the present invention can not be used to limit the content of invention, think of proposed by the present invention
The content that road and technical essential are included, all within protection scope.
The present invention provides a kind of data server cluster system based on WLC algorithm, total system is divided into three parts,
As shown in Figure 1, being clustered node server 1, load balance scheduling server 2 and message queue telemetering transport protocol client respectively
Hold server 3.
The internal structure of the scheduling broker service of clustered node server is as shown in Figure 2.The realization of load balance scheduling device
It is divided into four parts such as LVS virtual server, cluster service, WLC algorithm, external interface.LVS is provided to server node hardware
Management, such as multi-core CPU, redundance type disk, large capacity memory, high-speed bus driving, realized by LVS frame to difference
Server hardware shielding, the server node of cluster no longer limited by hardware driving.Cluster service software is server
The core content of node, the major function that it is realized have clustered node registration, clustered node parsing, clustered node scanning, cluster
The functions such as node processes are guarded, clustered node routes, clustered node log, provide complete cluster service.WLC algorithm is real
The method now dispatched, for the present invention using the dynamic adjustment algorithm based on time response, it passes through allocation list and cluster service
It is common to complete scheduling.External interface is used to be communicated with other clustered nodes, completes data exchange.
WLC algorithm proposed by the present invention has fully considered the performance difference of each node of cluster server, passes through the side of weight
Formula is evaluated.It is implemented as, and is already existing server section in cluster in cluster after service node registration
Point is numbered,
N={ N0, N1... Nn-1(formula 1)
Enable C indicate the load value of each node, then the total load in cluster is,
Csum=∑ C (Ni) (i=0,1 ... n-1) (formula 2)
W is enabled to indicate the weight of each node
If cluster is available, there is C (Nm)/W(Sm)=min (C (Ni)/W(Ni)) (i=0,1 ... n-1) (formula 3)
It enables L indicate the load value of each node, then has
Li=R1*LOAD+R2*DISK+R3*MEM+R4*RES+R5* BAND (formula 4)
R is the weight coefficient adjusted in real time, and LOAD, DISK, MEM, RES, BAND are respectively CPU, disk, memory, response
Time and bandwidth.
Enabling T is the message processing time of the MQTT of current last time.
After introducing time response, the adjustment of WLC algorithm are as follows:
Wi=Wi-1+T*(0.95-Li) (formula 5)
The scheduling of server cluster node can be carried out according to weight sequence.
Loads-scheduling algorithm process, that is, WLC calculating and storing process are as shown in Figure 3.
As a special example of the invention, Fig. 4 gives the application of a people's air defense server cluster based on WLC.
In certain people's air defense institute, field device is divided into three battery limits, living area and command area areas, each region is provided with one
Data server cluster, the data distributed data base Redis storing data between cluster.The data service that each area is equipped with
Device cluster scale is different, wherein maximum with the cluster scale of battery limits.In this cluster, other than conventional message communicating,
It is further provided with one group of offline service device cluster, for the history message of the equipment when storing in network interruption, once network recovery,
Offline service device will send the data to message server without scheduling by the way of pushing away.In order to guarantee in each region
Message processing locality as far as possible, this cluster is provided with a system load balancing module, only need to be when local area processing capacity is enough
Local area processing, and when local area processing capacity deficiency, system equalization module can call the calculating power in other regions to carry out
Processing.
The present invention for people's air defense institute device data server architecture is backward, problem of adaptation network load ability difference, mention
A kind of message service cluster system schema is gone out.The framework and data exchange ways of system are first defined, next illustrates cluster
The specific implementation of server each section, and WLC changeable weight algorithm is improved, finally build an application example.
It is to be understood that for the professional for understanding or understanding this invention thinking, it is any in embodiment party
Made in method equivalent replacement or division name change, all will not influence the present invention claims dedicated right, these be replaced or
The content of change belong to the present invention claims rights protection scope.
Claims (2)
1. a kind of people's air defense data server cluster based on WLC algorithm, which is characterized in that hardware includes clustered node server
(1), load balance scheduling server (2) and message queue telemetering transport protocol client-server (3), these three types of servers are hard
By six type shielding grid line physical connections between part, for realizing the publication of high-speed data interaction message；Clustered node server
It (1) is when in use 2 Dao more, when how many message queue telemetering transport protocol client-server (3) just needs to construct
How many a clustered node servers (1) are communicated using bus communication mode between clustered node server (1)；Each cluster
Software module is installed, message queue telemetering transport protocol Message Agent service is logical as MQTT data in node server (1)
News, function include distribution subscription, count retransmitting function, caching；Storage and log services use Redis database, and function includes
Data storage, log recording；Scheduling broker service is realized the management and running of more message task publications as Message Agent, is used
Based on the dispatcher software of nginx algorithm actuator, function is the load balance realized between clustered node server (1)；Bus
Client service is as high-speed data interaction bus communication management, using the queue service bus service software of RocketMQ, function
It can be the realization transmission for realizing the information between node；Management and running software is installed in load balance scheduling server (2), it is soft
Part is using nginx as algorithm actuator；And Redis database, Redis build host-guest architecture and carry out between each node
State synchronized, thus realize the caching of scheduling with it is synchronous；Each people's air defense field device be used as a message queue distant
It surveys transport protocol client-server (3).
2. the people's air defense data server cluster according to claim 1 based on WLC algorithm, which is characterized in that the pipe of cluster
Reason person increases as needed or deletes the node in cluster, by the reasonable distribution of load balance scheduling server, each node
Operating status be in lighter load level；Common load balancing algorithm has round robin, random hash method, minimum connection
Method, fastest response method；A kind of WLC algorithmic dispatching mechanism based on time response mechanism is proposed, WLC algorithm is a kind of dynamic
Weight assignment algorithm, it consider the CPU usage of server node in cluster, memory percentage, network bandwidth amount reality
Service condition is lined up all nodes in the method for weight, the availability highest of this node of weighted value minimum specification,
On the basis of WLC algorithm, the mechanism of time response a kind of is added, i.e., after the completion of each connection response, records this message
The time of processing is connected, and on the dispatch list being synchronized in each node, according to this record, load dispatch server is divided again
Timing is chosen the most fast node of response in fixed weight distribution basis and is handled.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910364337.3A CN110149378A (en) | 2019-04-30 | 2019-04-30 | A kind of people's air defense data server cluster based on WLC algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910364337.3A CN110149378A (en) | 2019-04-30 | 2019-04-30 | A kind of people's air defense data server cluster based on WLC algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110149378A true CN110149378A (en) | 2019-08-20 |
Family
ID=67594544
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910364337.3A Pending CN110149378A (en) | 2019-04-30 | 2019-04-30 | A kind of people's air defense data server cluster based on WLC algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110149378A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111131486A (en) * | 2019-12-30 | 2020-05-08 | 北京三快在线科技有限公司 | Load adjustment method and device of execution node, server and storage medium |
| CN111541599A (en) * | 2020-04-24 | 2020-08-14 | 山东山大电力技术股份有限公司 | Cluster software system and method based on data bus |
| CN112073506A (en) * | 2020-09-04 | 2020-12-11 | 哈尔滨工业大学 | IPv6 and message queue-based complex electromagnetic data acquisition method |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105282259A (en) * | 2015-11-13 | 2016-01-27 | 深圳联友科技有限公司 | Load balancing allocation method, agent and system used for background cluster service |
| CN105450994A (en) * | 2015-12-03 | 2016-03-30 | 河北远东通信系统工程有限公司 | Video commanding and dispatching system and method based on RTSP |
| CN108183945A (en) * | 2017-12-25 | 2018-06-19 | 冶金自动化研究设计院 | Civil air defense constructions and installations device intelligence management system based on technology of Internet of things |
| CN108696553A (en) * | 2017-04-06 | 2018-10-23 | 北京算云联科科技有限公司 | A kind of Internet of Things network control system based on mixed cloud |
| CN109361753A (en) * | 2018-11-02 | 2019-02-19 | 上海帆尚行科技有限公司 | A kind of Internet of things system framework and encryption method |
-
2019
- 2019-04-30 CN CN201910364337.3A patent/CN110149378A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105282259A (en) * | 2015-11-13 | 2016-01-27 | 深圳联友科技有限公司 | Load balancing allocation method, agent and system used for background cluster service |
| CN105450994A (en) * | 2015-12-03 | 2016-03-30 | 河北远东通信系统工程有限公司 | Video commanding and dispatching system and method based on RTSP |
| CN108696553A (en) * | 2017-04-06 | 2018-10-23 | 北京算云联科科技有限公司 | A kind of Internet of Things network control system based on mixed cloud |
| CN108183945A (en) * | 2017-12-25 | 2018-06-19 | 冶金自动化研究设计院 | Civil air defense constructions and installations device intelligence management system based on technology of Internet of things |
| CN109361753A (en) * | 2018-11-02 | 2019-02-19 | 上海帆尚行科技有限公司 | A kind of Internet of things system framework and encryption method |
Non-Patent Citations (1)
| Title |
|---|
| 刘玉艳，沈明玉: ""一种LVS负载均衡调度算法WLC的改进"", 《制造业自动化》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111131486A (en) * | 2019-12-30 | 2020-05-08 | 北京三快在线科技有限公司 | Load adjustment method and device of execution node, server and storage medium |
| CN111541599A (en) * | 2020-04-24 | 2020-08-14 | 山东山大电力技术股份有限公司 | Cluster software system and method based on data bus |
| CN111541599B (en) * | 2020-04-24 | 2021-08-06 | 山东山大电力技术股份有限公司 | Cluster software system and method based on data bus |
| CN112073506A (en) * | 2020-09-04 | 2020-12-11 | 哈尔滨工业大学 | IPv6 and message queue-based complex electromagnetic data acquisition method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110149378A (en) | A kind of people's air defense data server cluster based on WLC algorithm | |
| EP2710784B1 (en) | A method for load balancing of requests' processing of diameter servers | |
| US9509529B1 (en) | Assured messaging system with differentiated real time traffic | |
| US8891403B2 (en) | Inter-cluster communications technique for event and health status communications | |
| CN113011678B (en) | Virtual operation platform operation control method based on edge calculation | |
| US11743333B2 (en) | Tiered queuing system | |
| CN105812287A (en) | Effective circuits in packet-switched networks | |
| US8606908B2 (en) | Wake-up server | |
| CN107924332A (en) | The method and system of ICT service provisions | |
| KR102345082B1 (en) | Cloud based iec61850 information processing method | |
| WO2021120633A1 (en) | Load balancing method and related device | |
| CN104618466A (en) | System for balancing load and controlling overload based on message transfer and control method of system | |
| El-Zoghdy | A hierarchical load balancing policy for grid computing environment | |
| CN110476154A (en) | Proxy server device and method for data collection | |
| CN110855739A (en) | Container technology-based remote and heterogeneous resource unified management method and system | |
| WO2021032269A1 (en) | Distribution of constrained device data | |
| Slastikhin et al. | Redundant priority maintenance in the multi-channel systems | |
| Gerakos et al. | Motive-time-optimized contextual information flow on unmanned vehicles | |
| Mori et al. | Queueing theoretic approach to job assignment strategy considering various inter-arrival of job in fog computing | |
| CN108920111A (en) | Data sharing method and Distributed data share system | |
| CN106210624A (en) | A kind of method of electrically-based industry multi-stage cascade Streaming Media intelligent scheduling | |
| Kliushnikov et al. | Swarm of Unmanned Aerial Vehicles as a Multi-State Queueing System with Non-Controlled and Controlled Degradation | |
| CN115225664A (en) | Construction method of meteorological information infrastructure resource cloud platform | |
| CN114936071A (en) | Civil aircraft airborne distributed simulation system based on edge calculation | |
| CN111818310A (en) | Public safety management platform |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190820 |