CN115102960A - Enterprise internet of things management system based on flexible deployment - Google Patents

Enterprise internet of things management system based on flexible deployment Download PDF

Info

Publication number
CN115102960A
CN115102960A CN202210696552.5A CN202210696552A CN115102960A CN 115102960 A CN115102960 A CN 115102960A CN 202210696552 A CN202210696552 A CN 202210696552A CN 115102960 A CN115102960 A CN 115102960A
Authority
CN
China
Prior art keywords
data
transmission
service
value
telescopic
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202210696552.5A
Other languages
Chinese (zh)
Other versions
CN115102960B (en
Inventor
李强
崔传建
李温静
刘柱
赵峰
陈武
冯笑
邓惠贤
翁章君
郭梦溪
李春阳
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
State Grid Information and Telecommunication Co Ltd
Great Power Science and Technology Co of State Grid Information and Telecommunication Co Ltd
Original Assignee
State Grid Information and Telecommunication Co Ltd
Great Power Science and Technology Co of State Grid Information and Telecommunication Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by State Grid Information and Telecommunication Co Ltd, Great Power Science and Technology Co of State Grid Information and Telecommunication Co Ltd filed Critical State Grid Information and Telecommunication Co Ltd
Priority to CN202210696552.5A priority Critical patent/CN115102960B/en
Publication of CN115102960A publication Critical patent/CN115102960A/en
Application granted granted Critical
Publication of CN115102960B publication Critical patent/CN115102960B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

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
    • H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
    • H04L67/1004—Server selection for load balancing
    • H04L67/1012—Server selection for load balancing based on compliance of requirements or conditions with available server resources
    • 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
    • H04L67/12—Protocols specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Health & Medical Sciences (AREA)
  • Computing Systems (AREA)
  • General Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

The invention relates to an enterprise internet of things management system based on telescopic deployment, which comprises an access balance server, a front-end service cluster, a middle balance server, an intermediate service cluster, a rear-end balance server, a rear-end service cluster and a deployment server, wherein the access balance server is connected with the front-end service cluster; according to the user characteristic information of each user side, the user characteristic information is firstly divided by the deployment server, after the division is completed, the user characteristic groups can be classified correspondingly, then the unified management is carried out according to the user characteristic groups, the number of the corresponding server service modules is configured through the telescopic group configuration module, the service modules are refined according to the service components in advance, so that when the server is on line or off line, the parameters of the corresponding server can be directly known, the dynamic server configuration is realized, and therefore, the telescopic configuration part can be isolated by matching with the three layers of load balancing servers, the service resources are dynamically adjusted based on the change of the user side, and the current power service equipment is compatible.

Description

Enterprise internet of things management system based on flexible deployment
Technical Field
The invention relates to the technical field of power internet of things, in particular to an enterprise internet of things management system based on telescopic deployment.
Background
At present, along with the intelligent upgrading of power services, various power service devices have functions of intelligent acquisition, intelligent control, intelligent analysis and the like, the upgrading of power intelligent devices is accompanied with the increase of data volume of each power service device, which also brings higher requirements on an internet of things management system, along with the increase of service types of the power devices, if processing resources of a server cluster are difficult to maintain, problems of data loss, processing delay and the like easily occur, but in the prior art, servers are continuously added according to the function conditions, so that the cost is undoubtedly increased, the adding difficulty and the cost are higher, and although a load balancing strategy for realizing the maximum utilization of the servers by allocating the server resources also exists in the prior art, the optimal configuration of the server resources by the converted strategy is still not suitable for the emerging power internet of things data management system to a greater extent, the reason is that the terminals belonging to the power internet of things management system are power devices, the types, functions and formats of the acquired files of the power devices may be different, and not all the power service devices have an interactive function, but the service content corresponding to the currently used load balancing strategy is single, such as data storage or data distribution, so that resource calling among servers is simple and easy to implement.
Disclosure of Invention
In view of this, the present invention aims to provide an enterprise internet of things management system based on scalable deployment.
In order to solve the technical problems, the technical scheme of the invention is as follows:
an enterprise Internet of things management system based on telescopic deployment comprises an access balance server, a front-end service cluster, a middle balance server, an intermediate service cluster, a rear-end balance server, a rear-end service cluster and a deployment server;
the access balancing server is provided with an access acquisition module, and the access acquisition module is used for acquiring data characteristic information of an access user terminal and generating user characteristic information;
the deployment server is configured with a business service model, the business service model comprises a plurality of different user characteristic groups, the user characteristic groups are divided through a preset characteristic division strategy based on user characteristic information, the deployment server is configured with a user solid state load calculation module and a division updating module, and the user solid state load calculation module calculates the user solid state load of each user characteristic group according to the division result; when new user characteristic information is generated, the division updating module executes the characteristic division strategy again to divide the business service model so as to generate a new user characteristic group;
the inter-service cluster is provided with a telescopic group configuration module, the telescopic group configuration module is provided with a telescopic index table, a user characteristic index and corresponding service components are stored in the telescopic index table in advance, the telescopic group configuration module generates corresponding user characteristic indexes according to user characteristic groups to determine corresponding telescopic service groups, each telescopic service group comprises a plurality of service components, and the telescopic group configuration module configures the initial number of the corresponding service modules of each service component according to the solid-state load of a user;
the inter-service cluster is further provided with a telescopic response module, the telescopic response module obtains a corresponding telescopic target and an adjusting parameter from the received dynamic telescopic instruction, adjusts the number of the service modules of the telescopic service group corresponding to the telescopic target according to the adjusting parameter, and updates a service address list corresponding to the service modules to the rear-end balancing server.
Further, the deployment server is configured with a category database, the category database stores a plurality of category information, each category information includes a plurality of category keywords, and the characteristic division strategy includes
A1, extracting user keywords from the user characteristic information;
a2, matching the user keywords with the category keywords to obtain a category matching value;
a3, determining an initial group for each user characteristic information according to the category matching value, wherein each category information corresponds to one initial group;
a4, calculating the communication characteristic value of each user characteristic information through a preset communication characteristic algorithm, and calculating the data characteristic value of each user characteristic information through a preset data characteristic algorithm;
a5, calculating the intra-group difference value of the communication characteristic value and the data characteristic value of the user characteristic information in the group through a preset difference evaluation algorithm, and if the intra-group difference value is larger than a preset group compatibility threshold value, entering A6; if the intra-group difference value is smaller than the preset group compatibility threshold, repeating the step A5 for the next group until all groups are compared;
a6, re-dividing the data into new groups according to the communication characteristic value and the data characteristic value by a cluster analysis algorithm, updating the user characteristic information in the groups after waiting for a first preset time, and returning to the step A4.
Further, the communication characteristic algorithm is
Figure BDA0003702830220000031
Wherein A is a communication characteristic value, a 2 For the communication protocol solid state redundancy value, a 1 For communication protocol static redundancy values, T c For a data sampling period, T 1 Is a preset lower limit cycle threshold, C i The duty ratio corresponding to the ith sampling data is set, i is the number of the sampling data, and Delta C is the average duty ratio of the sampling data; the user characteristic information comprises communication protocol data, the deployment server is configured with a protocol communication table, each communication protocol data in the protocol communication table has a corresponding communication protocol solid-state redundancy value and a corresponding communication protocol static redundancy value, and the corresponding protocol solid-state redundancy value and the corresponding communication protocol static redundancy value are obtained through the communication protocol data in the user characteristic information.
Further, the data characteristic algorithm is
Figure BDA0003702830220000032
Wherein B is a data characteristic value, S i The deployment server is configured with a data type table for the data priority value corresponding to the ith sampling data, each data type index in the data type table corresponds to a data priority value, the data priority value corresponding to the acquired data is called through the data type index in the user characteristic information, and if the data priority value corresponding to the acquired data does not exist in the data type index table, the S is calculated through the following formula k ＝b 1 S n +b 2 S m ，S k For the data priority value corresponding to the kth sample data, b 1 Adjusting the weight for a preset length, b 2 Adjusting weights for preset features, having b 1 +b 2 ＝1，S n Is the data priority value of the sample data closest to the effective data length of the sample data, S m Is closest to the data characteristics of the sampled dataThe data priority value of the sampled data.
Further, the discrepancy evaluation algorithm comprises
Figure BDA0003702830220000033
Wherein W is the intra-group difference value, A j Is the communication characteristic value corresponding to the jth user characteristic information, delta A is the average value of the data characteristic values in the group, B j And the data characteristic value is corresponding to the jth user characteristic information, the delta B is the average value of the data characteristic values in the group, and the j is the number of the user characteristic information in the group.
Further, the user solid-state load is the sum of the solid-state sub-loads of each piece of user characteristic information, the deployment server is configured with a load index database, the load index database stores load analysis conditions and corresponding solid-state sub-loads, and the deployment server judges the user characteristic information which is met through the load analysis conditions to configure the corresponding solid-state sub-loads;
each service component is configured with absolute service weight, when the service components form a telescopic service group, the relative service weight is obtained through calculation according to the absolute service weight corresponding to each service component, and the number of the corresponding service modules is obtained through calculation according to the solid load of the user and the corresponding relative service weight;
the intervenient service cluster is also provided with a load feedback module, the load feedback module comprises a state monitoring unit configured on the service module, the state monitoring unit acquires the load state of the corresponding service module to generate an actual load value, when the actual load value is larger than an upper limit adjusting threshold value, the absolute service weight of the corresponding service assembly is increased, and when the actual load value is smaller than a lower limit adjusting threshold value, the absolute service weight of the corresponding service assembly is reduced.
Further, the inter-service cluster further comprises a channel construction module and a channel selection module, wherein the channel construction module configures transmission characteristics for each transmission channel, and the transmission characteristics comprise a transmission protocol, a transmission success rate, a transmission security level and a transmission bandwidth; the channel selection module is used for matching a corresponding transmission channel for the telescopic service group according to the transmission characteristics and service requirement characteristics generated by the telescopic service group in real time, wherein the service requirement characteristics comprise a transmission protocol, a data priority, a data transmission requirement and a data security level;
each transmission channel is provided with a transmission monitoring unit, the transmission monitoring unit monitors the transmission load of the transmission channel in real time, and when the transmission load exceeds a preset load threshold value, the corresponding telescopic service group is accessed to a standby transmission channel.
Further, the deployment server includes a transmission balancing module configured with a load coordination policy, where the load coordination policy includes
B1, acquiring a transmission protocol from the service requirement characteristics;
b2, screening the corresponding transmission channel according to the transmission protocol, and calculating the transmission capacity value of the obtained transmission channel by a channel load formula, wherein the channel load formula is
Figure BDA0003702830220000041
Where R is the transmission capacity value of the transmission channel,
Figure BDA0003702830220000042
in order to have a success rate of the transmission,
Figure BDA0003702830220000043
in order to transmit the security level,
Figure BDA0003702830220000044
in order to be able to transmit the bandwidth,
Figure BDA0003702830220000045
the index security level is obtained by inquiring an index security level table according to the obtained average value of the transmission security levels of the transmission channels, and the index security level table takes the average value of the transmission security levels and the transmission security level of the current transmission channel as indexes to record the index security level;
b3, summing the obtained transmission capacity values to obtain a total transmission capacity value, and calculating the current transmission redundancy when waiting to receive a transmission coordination request, wherein the current transmission redundancy is the difference between the total transmission capacity value and the total current load value;
b4, acquiring a solid state transmission load and a dynamic transmission load table in the transmission coordination request, where the solid state transmission load reflects the transmission load of the fixed task data of the user side corresponding to the scalable service set, the dynamic transmission load table uses the predicted transmission redundancy as an index, each predicted transmission redundancy corresponds to a time-sharing transmission condition, and the predicted transmission redundancy is the difference between the current transmission redundancy and the solid state transmission load;
and B5, sending the time-sharing transmission condition to an access balancing server, wherein the access balancing server configures transmission time for the dynamic data corresponding to the access user terminal through the time-sharing transmission condition.
Furthermore, the inter-service cluster is configured with a telescopic prediction module, the telescopic prediction module is connected with a historical telescopic database, historical telescopic data of each telescopic service group are stored in the historical telescopic database, the historical telescopic data comprise telescopic targets and corresponding adjusting parameters, and the telescopic prediction module generates corresponding dynamic receiving instructions according to the historical telescopic data and sends the instructions to the telescopic response module.
Further, the intervening service cluster includes an index marking module, and the index marking module is configured to mark a new user feature index to the user feature group or delete a mark of the user feature index from the user feature group.
The technical effects of the invention are mainly embodied in the following aspects: through the setting, according to the user characteristic information of each user side, the user characteristic information is firstly divided by the deployment server, after the division is completed, the user characteristic groups can be classified correspondingly, then the unified management is carried out according to the user characteristic groups, the number of the corresponding server service modules is configured through the telescopic group configuration module, the service modules are refined according to the service components in advance, so that when the server is on line or off line, the parameters of the corresponding server can be directly known, the dynamic server configuration is realized, and therefore, the telescopic configuration part can be isolated by matching with the three layers of load balancing servers, the service resources are dynamically adjusted based on the change of the user side, and the current power service equipment is compatible.
Drawings
FIG. 1: the invention discloses a schematic diagram of a system architecture;
FIG. 2: the invention provides an intermediate service cluster schematic diagram;
FIG. 3: the present invention deploys a server architecture diagram.
Reference numerals are as follows: 100. accessing a balance server; 110. accessing an acquisition module; 200. a front-end service cluster; 300. an intermediate balancing server; 400. an intervening service cluster; 401. a service component; 410. a telescopic group configuration module; 420. a telescopic response module; 430. a load feedback module; 440. a telescoping prediction module; 450. an index marking module; 460. a channel construction module; 500. a back-end equalization server; 600. a back-end service cluster; 700. deploying a server; 701. a category database; 702. a load index database; 710. a user solid state load calculation module; 720. dividing an updating module; 730. and a transmission balancing module.
Detailed Description
The following detailed description of the present invention is provided to facilitate the understanding and appreciation of the technical aspects of the present invention in connection with the accompanying drawings.
Referring to fig. 1, an enterprise internet of things management system based on flexible deployment is firstly a whole system architecture logic, including an access equilibrium server 100, a front-end service cluster 200, an intermediate equilibrium server 300, an intermediate service cluster 400, a back-end equilibrium server 500, a back-end service cluster 600, and a deployment server 700; the access balancing server 100, the middle balancing server 300 and the back-end balancing server 500 all have a load balancing algorithm to realize load balancing, and the system architecture is divided into three layers: the front-end service cluster 200, the intermediate service cluster 400 and the back-end service cluster 600, because the front-end service cluster 200 is directly connected to the user terminal, and this part can directly balance the server resources through the load balancing algorithm, there is no need to construct a flexible logic, and the back-end service cluster 600 mainly serves in the aspects of data storage, and can also complete resource configuration through the load balancing algorithm, but the intermediate service cluster 400 is a service item generated based on business logic, and may need flexible deployment according to business upgrading or bursty data requirements, and a great waste will occur only through a fixed server cluster, so the present invention separates the intermediate service cluster through three layers of load balancing to realize flexible deployment, and the flexible deployment is based on the following logic: the whole architecture is divided into two parts of a non-state part and a state part during design, and the scalable deployment target is realized based on the separation of the two states. The part of the business logic is used as a stateless part and can be freely expanded horizontally, when a request is distributed, the request can be distributed to a new process for processing, after an application service is designed into a stateless mode, the request can be distributed to different application servers through a load balancing strategy, the load balancing server can sense or configure the number of the servers of the cluster, the request can be distributed to a newly on-line server, the off-line server is stopped, and the problem of flexibility of the application server cluster is solved, so that the business logic part is configured into an intermediate service cluster 400, and then the refined component splitting logic realizes dynamic server configuration. The data state information is stored in backend stateful middleware such as a cache, a database, an object store, a big data platform, a message queue, etc., and is therefore configured as a backend service cluster 600.
The access balancing server 100 is configured with an access acquisition module 110, where the access acquisition module 110 is configured to acquire data characteristic information of an access user terminal and generate user characteristic information; first, the access acquisition module 110 is configured to acquire information of each user terminal, where the information includes data in text formats such as the type, brand, name, service content, address, interface and protocol used for communication, and data in numerical forms such as communication parameters and real-time communication fields of the user terminal, and all of the information is used as user characteristic information.
The deployment server 700 is configured with a business service model, the business service model includes a plurality of different user feature groups, the user feature groups are divided according to user feature information through a preset feature division strategy, firstly, a category database 701 is configured according to the deployment server 700, the category database 701 stores a plurality of category information, each category information includes a plurality of category keywords, the category information in the category database 701 is configured in advance, because the text information of each user end cannot be collected in a formatting mode, the type of the user end is determined firstly, and the feature division strategy includes a plurality of feature division strategies
A1, extracting user keywords from the user characteristic information; the strategy for extracting the keywords is not repeated, and the extraction of the keywords of the user is completed by taking the category keywords as the basis and combining the text information extraction strategy and means.
A2, matching the user keywords with the category keywords to obtain a category matching value; since the category information includes a plurality of category keywords, the category keywords of different category information are different, and the matching degree corresponding to each category keyword is different, for example, if the user keyword has a current collecting device, the category keywords are given different matching degrees in different category information, so that the category information of each category keyword having current, collecting and device, current device, collecting device, and the like will generate different matching degrees, and the matching degree of each category information is summed to generate the corresponding category matching value corresponding to the user characteristic information.
A3, determining an initial group for each user characteristic information according to the category matching value, wherein each category information corresponds to one initial group; if the matching value of the user characteristic information corresponding to one category information is the highest, the category information is preferentially used as a category name to create an initial category, and other factors are also needed for creating the initial category of the category information, for example, data in the initial category cannot be too little, for example, if one category information only has one matched user characteristic information, the user characteristic information can only be matched with other category information even if the matching value is the highest, and the category information with the highest matching value is selected for creating the initial category.
A4, calculating the communication characteristic value of each user characteristic information through a preset communication characteristic algorithm, and calculating the data characteristic value of each user characteristic information through a preset data characteristic algorithm; then, calculating characteristic values of two dimensions of each user characteristic information, wherein the communication characteristic values reflect the requirements and expected loads of the user on communication, and the data characteristic information reflects the redundancy of the data transmission logic of the user
Figure BDA0003702830220000081
Wherein A is a communication characteristic value, a 2 For the communication protocol solid state redundancy value, a 1 For the static redundancy value, T, of the communication protocol c For a data sampling period, T 1 Is a preset lower limit cycle threshold, C i The duty ratio corresponding to the ith sampling data is set, i is the number of the sampling data, and delta C is the average duty ratio of the sampling data; the user characteristic information comprises communication protocol data, the deployment server is configured with a protocol communication table, each piece of communication protocol data in the protocol communication table has a corresponding communication protocol solid-state redundancy value and a corresponding communication protocol static redundancy value, and the corresponding protocol solid-state redundancy value and the corresponding communication protocol static redundancy value are called through the communication protocol data in the user characteristic information. Firstly, user characteristic information records corresponding communication protocol data, then a solid redundant value and a static redundant value of the communication protocol data are determined according to the communication protocol data, the solid redundant value reflects the load generated by the communication logic to service resources, the static redundant value reflects the load corresponding to each unit of data, then the equivalent load amount of a user terminal in unit time is judged according to the rule of user data sending and recorded as the communication characteristic value, and therefore the communication characteristic value also judges the communication load possibly generated by the user terminal by collecting actually generated data information.
The data characteristic algorithm is
Figure BDA0003702830220000091
Wherein B is a data characteristic value, S i The deployment server is configured with a data type table for the data priority value corresponding to the ith sampling data, each data type index in the data type table corresponds to a data priority value, the data priority value corresponding to the acquired data is called through the data type index in the user characteristic information, and if the data priority value corresponding to the acquired data does not exist in the data type index table, the S is calculated through the following formula k ＝b 1 S n +b 2 S m ，S k For the data priority value corresponding to the kth sample data, b 1 Adjusting the weight for a preset length, b 2 Adjusting weights for preset features, having b 1 +b 2 ＝1，S n Is the data priority value of the sample data closest to the effective data length of the sample data, S m The data priority value of the sample data closest to the data characteristic of the sample data. The data characteristic algorithm is used for the comprehensive priority values of different sections of sampled data, because each data packet may have different corresponding tasks and different task requirements, the corresponding priority values are called in a table look-up mode according to the task requirements of each data packet, the corresponding data fields of the data packets where the task requirements can be located are identified, and therefore the overall data transmission requirements of the equipment in the sampling period can be obtained, and the data transmission requirements are high, so that the data are important. The data priority value may not be in a form of a mark or not recorded in a data type table, and an equivalent data priority value can be calculated by finding similar data, which requires that corresponding data priority values of all data capable of being subjected to table lookup are calculated first, then a limited value of the data incapable of being subjected to table lookup is calculated according to the known data priority value, and through the length of the data and the characteristics of the data, the characteristics of the data including data change rate, data peak value, key data encoding and other information can be used as data characteristics, and an unknown data priority value is calculated according to the two closest data priority values.
A5, calculating the intra-group difference value of the communication characteristic value and the data characteristic value of the user characteristic information in the group through a preset difference evaluation algorithm, and if the intra-group difference value is larger than a preset group compatibility threshold value, entering A6; if the intra-group difference value is smaller than the preset group compatibility threshold, repeating the step A5 for the next group until all groups are compared; the discrepancy evaluation algorithm comprises
Figure BDA0003702830220000101
Wherein W is the intra-group difference value, A j Is the communication characteristic value corresponding to the jth user characteristic information, delta A is the average value of the data characteristic values in the group, B j And the data characteristic value is corresponding to the jth user characteristic information, the delta B is the average value of the data characteristic values in the group, and the j is the number of the user characteristic information in the group. The purpose of this step is to calculate whether to perform grouping again in the group, and since the meaning of directly distributing the scalable groups is not too large if the number in the group is large, and the scalable groups cannot be quickly adjusted according to the actual change situation, a group compatibility threshold is set, and when the intra-group difference value in the group is large, it indicates that grouping needs to be performed, then step a6 is performed. And otherwise, grouping is carried out until all the user characteristic information has the corresponding group.
A6, re-dividing the data into new groups according to the communication characteristic value and the data characteristic value by a cluster analysis algorithm, updating the user characteristic information in the groups after waiting for a first preset time, and returning to the step A4. If the grouping is to be performed again, the communication characteristic value is taken as an abscissa for analysis and the data characteristic value is taken as an ordinate for analysis through a cluster analysis algorithm, preferably through a DBScan or K-MEAN cluster analysis algorithm, and the number of the grouping can be based on the actual density distribution and the total number in the group, which is not limited herein.
The deployment server 700 is configured with a user solid state load calculation module 710 and a division update module 720, where the user solid state load calculation module 710 calculates a user solid state load of each user feature group according to a division result; when new user feature information is generated, the partition updating module 720 re-executes the feature partition strategy to partition the business service model to generate a new user feature group; the user solid load is the sum of the solid sub-loads of each user characteristic information, the deployment server 700 is configured with a load index database 702, the load index database 702 stores load analysis conditions and corresponding solid sub-loads, and the deployment server 700 judges the satisfied user characteristic information according to the load analysis conditions to configure the corresponding solid sub-loads; the load corresponding to the user characteristic information is analyzed through the load analysis condition, for example, the load analysis condition can judge the number of repeated fields in the collected data of the user characteristic information and the proportion of the length in the total data volume, on the other hand, the proportion of the total load of the data station with fixed length is considered, whether the data fall into the corresponding data range is judged to judge whether the load analysis condition is met, whether the information corresponding to the key word is relevant or not is judged, then, the solid-state sub-load can be configured for the data, and the solid-state total load can be obtained by adding the solid-state sub-load.
The inter-service cluster 400 is configured with a scalable group configuration module 410, the scalable group configuration module 410 is configured with a scalable index table, the scalable index table pre-stores a user characteristic index and a corresponding service component 401, the scalable group configuration module 410 generates a corresponding user characteristic index according to a user characteristic group to determine a corresponding scalable service group, each scalable service group includes a plurality of service components 401, and the scalable group configuration module 410 configures an initial number of corresponding service modules of each service component 401 according to a user solid-state load; first, the flexible group configuration module 410 can generate the configuration of the corresponding server resource according to each flexible group, because the requirement of the user feature group is generated in a user feature index manner, and the service component 401 is divided in advance, based on the distributed componentization splitting principle, the flexible component layering splitting is realized, and firstly, based on the componentization splitting principle and the splitting strategy, the inter-component splitting is performed on the internet-of-things management platform. Comprises a front-end component, a business service component 401, a message component, a database component and a cache component. Distributed cluster deployment can be performed among the large-scale components. The business service component 401 divides components according to coarse granularity, and comprises a basic component class, a platform management class component, a connecting component, a data processing component and the like, and then performs component splitting in a finer step through the coarse-granularity components, so that high cohesion in the components and low coupling among the components are realized, reusability and expansibility of the components are improved, and scalability on a platform component level is met. The basic component class is mainly divided into components such as a flow engine component, a protocol analysis component, a timing task component, a file service component 401, a search engine and the like. The platform management component is mainly divided into an equipment management component, an application management component, an intelligent operation and maintenance component, a model management component, a remote operation and maintenance management component, a safety management component, an operation monitoring component, a system management component, a capacity opening component, a message processing component and the like. Therefore, what the user characteristic group has needs, the corresponding content is determined from the service components 401, then the quantity of each component can be configured intelligently, so that the flexible group can determine the quantity required by each service component 401 corresponding to the user characteristic index, and simultaneously determine the data of the corresponding service module, and the proportion is known, then the servers can be dynamically online and offline under the condition that the quantity and the parameters of the servers corresponding to the service components 401 are known, and the service module can be directly understood as a server instance, so that the servers of a server cluster can be configured, and flexible deployment is completed.
The inter-service cluster 400 is further configured with a telescopic response module 420, where the telescopic response module 420 obtains a corresponding telescopic target and an adjustment parameter from the received dynamic telescopic instruction, adjusts the number of service modules of a telescopic service group corresponding to the telescopic target according to the adjustment parameter, and updates a service address list corresponding to the service module to the back-end balancing server 500. By researching a dynamic scaling mechanism, the platform can set scaling rules according to business requirements and strategies, automatically increase application server instances to ensure computing capacity when the business volume increases, and automatically reduce application server instances to save cost when the business volume decreases. The dynamic expansion mainly comprises dynamic expansion and dynamic contraction. A scalability group is the basic management unit when using application service instances for dynamic scalability management services. The scalable group is used for managing application service instances having the same application scenario and supporting associating multiple load balancing instances. After the telescopic group associates the load balancing instance with the database service instance, the application service instance is automatically added as a back-end server of the load balancing instance when joining the telescopic group, and the Internet Protocol (IP) of the application service instance is automatically added into an access white list of the database service instance. A scaled configuration is an instance template used when dynamically scaling automatically creating application service instances. A scalability group supports the creation of multiple scalability configurations, but only allows one scalability configuration to be in effect at a time.
Dynamic expansion: when some services of the platform, such as connection service and data processing service, need to be upgraded and concurrency and processing capacity is improved, dynamic expansion can automatically complete bottom layer resource upgrade, and access and processing delay and overload operation of resources are avoided. The platform pays attention to the resource use index of the application server instance in real time based on the monitoring component, when the resource use condition of all instances is larger than a set threshold value, a corresponding dynamic telescopic instruction is generated, the application server resource is expanded according to a dynamic rule configured in advance, a proper number of application server instances are automatically created, and the application server instances are added to load balancing.
Dynamic shrinkage: when the service volume is reduced, the dynamic expansion automatically completes the release of bottom layer resources, and the resource waste is avoided. The platform monitors the service condition of the application service instances on the basis of the monitoring component, when the index monitoring detects that the CPU utilization rate threshold of the application service instances in the telescopic group generates a corresponding dynamic telescopic instruction, the dynamic telescopic component contracts application service resources according to the configured telescopic rule, automatically releases the application service instances with proper quantity, and automatically removes the application service instances from an access white list of a rear-end server of the load balancing instances.
Each service component 401 is configured with an absolute service weight, when the service components 401 form a scalable service group, a relative service weight is calculated according to the absolute service weight corresponding to each service component 401, and the number of corresponding service modules is calculated according to the user solid load and the corresponding relative service weight;
the inter-service cluster 400 is further configured with a load feedback module 430, where the load feedback module 430 includes a state monitoring unit configured to the service module, and the state monitoring unit collects a load state of the corresponding service module to generate an actual load value, when the actual load value is greater than an upper limit adjustment threshold, the absolute service weight of the corresponding service component 401 is increased, and when the actual load value is less than a lower limit adjustment threshold, the absolute service weight of the corresponding service component 401 is decreased. The service condition of each service component 401 is monitored in real time, the weight is dynamically adjusted, and the adjustment of the number of servers during distribution is realized, so that the configuration of a server cluster is continuously optimized under the condition that the system continuously works, and the utilization rate is pertinently improved.
The inter-service cluster 400 is configured with a telescopic prediction module 440, the telescopic prediction module 440 is connected to a historical telescopic database, the historical telescopic database stores historical telescopic data of each telescopic service group, the historical telescopic data includes a telescopic target and corresponding adjustment parameters, and the telescopic prediction module 440 generates a corresponding dynamic receiving instruction to the telescopic response module 420 according to the historical telescopic data. By analyzing historical expansion and contraction data, when the analysis rule is met, such as historical expansion and contraction delay or a historical expansion and contraction fixed rule, the dynamic deployment of the expansion and contraction is carried out in advance, and the hysteresis caused by the dynamic receiving instruction input delay or the analysis delay and the like is avoided.
The intermediary service cluster 400 includes an index tagging module 450, and the index tagging module 450 is configured to tag a new user feature index to a user feature group or delete a tag of the user feature index from the user feature group. Through the setting of the index marking module 450, the background can add service functions to different users, and the change can be directly fed back to the whole logic of telescopic deployment, so that the users can dynamically mark, the user characteristic index is a service mark, for example, you need to select a cloud monitoring index and specify a target value. The elastic expansion and contraction automatically calculates the required number of instances and expands and contracts, so that the cloud monitoring index is maintained to be close to the target value. The basic requirements in the subscriber profile, such as the service of the file delivery service, are the index of the subscriber profile that can be identified when the subscriber accesses.
The southbound data transmission is characterized by large data volume and high concurrency, and once a data transmission channel is unstable or unavailable, massive data can be lost, so that the use of the whole platform is influenced. Therefore, a stable and highly reliable data transmission channel is important, so the following is designed: the inter-service cluster 400 further includes a channel construction module 460 and a channel selection module, wherein the channel construction module 460 configures transmission characteristics for each transmission channel, and the transmission characteristics include a transmission protocol, a transmission success rate, a transmission security level, and a transmission bandwidth; the channel selection module is used for matching a corresponding transmission channel for the telescopic service group according to the transmission characteristics and service requirement characteristics generated by the telescopic service group in real time, wherein the service requirement characteristics comprise a transmission protocol, a data priority, a data transmission requirement and a data security level;
each transmission channel is provided with a transmission monitoring unit, the transmission monitoring unit monitors the transmission load of the transmission channel in real time, and when the transmission load exceeds a preset load threshold value, the corresponding telescopic service group is accessed to a standby transmission channel. The configuration of the channel attributes is realized through the division of different dimensions of the data, and the overall management of the channels is finally realized. The importance level of the channel is divided based on the importance degree of the data: data with high importance level walk a channel with high importance level, the configuration of the construction of the channel is relatively high, the security level is high, and resources are preferentially distributed. And common data can be configured and maintained through channel management when going through a common channel. The types of the channels are divided based on different transmission protocols of the channels: the protocol support of the data transmission channel is various, and the data transmission channel comprises various open source message queues and self-defined message transmission components. And the protocol type of the channel can be maintained and configured when the channel is newly built. Dividing the scale of the channel based on the scale of data transmission: according to the transmission data with different scales, data transmission channels with different scales are created to adapt to the data transmission with different scale degrees. The channels are finely divided, and resources are reasonably utilized. And a certain number of channels are created in advance according to the actual situation by maintaining the configuration of the channel pool and are placed into the channel pool. And constructing a channel distribution strategy center, and distributing a data transmission channel for each connected device through the strategy center. In the data transmission process, the stability and the performance of data are monitored, the data are higher than a preset threshold value for a long time, and an idle adaptive channel is immediately allocated from a channel pool for switching, so that the high reliability of data transmission is ensured.
Because the data of the user side is divided into several cases, one is fixed data, for example, real-time interactive data, for example, the a device updates the acquisition information every minute, such data can calculate the actual fixed load by pre-judgment, and there is summary data generated every other day, and the data transmission time can be flexibly set, and also burst data, for example, data such as abnormal alarm, can occur, and if the data at each timing can be transmitted by staggering the processing peak of the server through the policy requirement, the load of the server can be reduced, and the specific mode is as follows: the deployment server 700 includes a transmission balancing module 730, the transmission balancing module 730 is configured with a load coordination policy, the load coordination policy includes B1, and a transmission protocol is obtained from a service demand characteristic;
b2, screening the corresponding transmission channel according to the transmission protocol, and calculating the transmission capacity value of the obtained transmission channel by a channel load formula, wherein the channel load formula is
Figure BDA0003702830220000151
Where R is the transmission capacity value of the transmission channel,
Figure BDA0003702830220000152
in order to have a success rate of the transmission,
Figure BDA0003702830220000153
in order to transmit the security level of the communication,
Figure BDA0003702830220000154
in order to be able to transmit the bandwidth,
Figure BDA0003702830220000155
the index security level is obtained by inquiring an index security level table according to the obtained average value of the transmission security levels of the transmission channels, and the index security level table takes the average value of the transmission security levels and the transmission security level of the current transmission channel as indexes to record the index security level; the size of the channels themselves is first calculated, as reflected by the R, i.e., the transmission capacity value, since the same type of data uses the same channel, the channels can be mutually reserved.
B3, summing the obtained transmission capacity values to obtain a total transmission capacity value, and calculating the current transmission redundancy when waiting to receive a transmission coordination request, wherein the current transmission redundancy is the difference between the total transmission capacity value and the total current load value;
b4, acquiring a solid state transmission load and a dynamic transmission load table in the transmission coordination request, where the solid state transmission load reflects the transmission load of the fixed task data of the user side corresponding to the scalable service set, the dynamic transmission load table uses the predicted transmission redundancy as an index, each predicted transmission redundancy corresponds to a time-sharing transmission condition, and the predicted transmission redundancy is the difference between the current transmission redundancy and the solid state transmission load; the fixed redundancy is identified and determined by the content of the transmitted data, then a corresponding dynamic load table is generated according to the timing task of the user side, the transmission coordination request is automatically generated or generated according to the request of the user side, the dynamic transmission load table can judge whether the corresponding time-sharing transmission condition is sent in a peak error mode or not according to the predicted transmission redundancy, for example, if the predicted transmission redundancy is higher, the time-sharing transmission condition can be sent in a peak error mode without the peak error mode, otherwise, the peak error sending can be carried out.
And B5, sending the time-sharing transmission condition to an access balancing server, wherein the access balancing server configures transmission time for the dynamic data corresponding to the access user terminal through the time-sharing transmission condition. The time-sharing transmission condition may include a priority sub-condition and a size sub-condition, and when the priority and the size of the data task do not satisfy the conditions, the data is time-sharing transmitted.
The above are only typical examples of the present invention, and besides, the present invention may have other embodiments, and all the technical solutions formed by equivalent substitutions or equivalent changes are within the scope of the present invention as claimed.

Claims (10)

1. The utility model provides an enterprise thing allies oneself with management system based on flexible deployment which characterized in that: the system comprises an access balance server, a front-end service cluster, a middle balance server, an intermediate service cluster, a rear-end balance server, a rear-end service cluster and a deployment server;
the access balancing server is provided with an access acquisition module, and the access acquisition module is used for acquiring data characteristic information of an access user terminal and generating user characteristic information;
the deployment server is configured with a business service model, the business service model comprises a plurality of different user characteristic groups, the user characteristic groups are divided through a preset characteristic division strategy according to user characteristic information, the deployment server is configured with a user solid load calculation module and a division updating module, and the user solid load calculation module calculates the user solid load of each user characteristic group according to the division result; when new user characteristic information is generated, the division updating module executes the characteristic division strategy again to divide the business service model so as to generate a new user characteristic group;
the inter-service cluster is provided with a telescopic group configuration module, the telescopic group configuration module is provided with a telescopic index table, a user characteristic index and corresponding service components are stored in the telescopic index table in advance, the telescopic group configuration module generates corresponding user characteristic indexes according to user characteristic groups to determine corresponding telescopic service groups, each telescopic service group comprises a plurality of service components, and the telescopic group configuration module configures the initial number of the corresponding service modules of each service component according to the solid-state load of a user;
and the inter-service cluster is also provided with a telescopic response module, the telescopic response module obtains a corresponding telescopic target and an adjusting parameter from the received dynamic telescopic instruction, adjusts the number of service modules of a telescopic service group corresponding to the telescopic target according to the adjusting parameter, and updates a service address list corresponding to the service modules to the rear-end balance server.
2. The enterprise internet of things management system based on scalable deployment of claim 1, wherein: the deployment server is configured with a category database, the category database stores a plurality of category information, each category information comprises a plurality of category keywords, and the characteristic division strategy comprises
A1, extracting user keywords from the user characteristic information;
a2, matching the user keywords with the category keywords to obtain a category matching value;
a3, determining an initial group for each user characteristic information according to the category matching value, wherein each category information corresponds to one initial group;
a4, calculating the communication characteristic value of each user characteristic information through a preset communication characteristic algorithm, and calculating the data characteristic value of each user characteristic information through a preset data characteristic algorithm;
a5, calculating the intra-group difference value of the communication characteristic value and the data characteristic value of the user characteristic information in the group through a preset difference evaluation algorithm, and if the intra-group difference value is larger than a preset group compatibility threshold value, entering A6; if the intra-group difference value is smaller than the preset group compatibility threshold, repeating the step A5 for the next group until all groups are compared;
a6, re-dividing the data into new groups according to the communication characteristic value and the data characteristic value by a cluster analysis algorithm, updating the user characteristic information in the groups after waiting for a first preset time, and returning to the step A4.
3. The enterprise internet of things management system based on scalable deployment of claim 1, wherein: the communication characteristic algorithm is
Figure FDA0003702830210000021
Wherein A isCommunication characteristic value, a 2 For the communication protocol solid state redundancy value, a 1 For communication protocol static redundancy values, T c For a data sampling period, T 1 Is a preset lower limit cycle threshold, C i The duty ratio corresponding to the ith sampling data is set, i is the number of the sampling data, and Delta C is the average duty ratio of the sampling data; the user characteristic information comprises communication protocol data, the deployment server is configured with a protocol communication table, each communication protocol data in the protocol communication table has a corresponding communication protocol solid-state redundancy value and a corresponding communication protocol static redundancy value, and the corresponding protocol solid-state redundancy value and the corresponding communication protocol static redundancy value are obtained through the communication protocol data in the user characteristic information.
4. The enterprise internet of things management system based on scalable deployment of claim 3, wherein: the data characteristic algorithm is
Figure FDA0003702830210000022
Wherein B is a data characteristic value, S i For a data priority value corresponding to the ith sampling data, the deployment server is configured with a data type table, each data type index in the data type table corresponds to a data priority value, the data priority value corresponding to the acquired data is called through the data type index in the user characteristic information, and if the data priority value corresponding to the acquired data does not exist in the data type index table, the data type table is used for calculating S through the following formula k ＝b 1 S n +b 2 S m ，S k For the data priority value corresponding to the kth sample data, b 1 Adjusting the weight for a preset length, b 2 Adjust weights for preset features, have b 1 +b 2 ＝1，S n Is the data priority value of the sample data closest to the effective data length of the sample data, S m The data priority value of the sample data closest to the data characteristic of the sample data.
5. The enterprise internet of things management system based on scalable deployment of claim 4, wherein: the discrepancy evaluation algorithm comprises
Figure FDA0003702830210000031
Wherein W is the intra-group difference value, A j Is the communication characteristic value corresponding to the jth user characteristic information, delta A is the average value of the data characteristic values in the group, B j And the data characteristic value corresponding to the jth user characteristic information is delta B, the average value of the data characteristic values in the group is delta B, and j is the number of the user characteristic information in the group.
6. The enterprise internet of things management system based on scalable deployment of claim 1, wherein: the user solid-state load is the sum of the solid-state sub-loads of each user characteristic information, the deployment server is configured with a load index database, the load index database stores load analysis conditions and corresponding solid-state sub-loads, and the deployment server judges the satisfied user characteristic information according to the load analysis conditions to configure the corresponding solid-state sub-loads;
each service component is configured with absolute service weight, when the service components form a telescopic service group, the relative service weight is obtained through calculation according to the absolute service weight corresponding to each service component, and the number of the corresponding service modules is obtained through calculation according to the solid load of the user and the corresponding relative service weight;
the intervenient service cluster is also provided with a load feedback module, the load feedback module comprises a state monitoring unit configured on the service module, the state monitoring unit acquires the load state of the corresponding service module to generate an actual load value, when the actual load value is larger than an upper limit adjusting threshold value, the absolute service weight of the corresponding service assembly is increased, and when the actual load value is smaller than a lower limit adjusting threshold value, the absolute service weight of the corresponding service assembly is reduced.
7. The enterprise internet of things management system based on scalable deployment of claim 1, wherein: the inter-service cluster also comprises a channel construction module and a channel selection module, wherein the channel construction module configures transmission characteristics for each transmission channel, and the transmission characteristics comprise a transmission protocol, a transmission success rate, a transmission safety level and a transmission bandwidth; the channel selection module is used for matching a corresponding transmission channel for the telescopic service group according to the transmission characteristics and service requirement characteristics generated by the telescopic service group in real time, wherein the service requirement characteristics comprise a transmission protocol, a data priority, a data transmission requirement and a data security level;
each transmission channel is provided with a transmission monitoring unit, the transmission monitoring unit monitors the transmission load of the transmission channel in real time, and when the transmission load exceeds a preset load threshold value, the corresponding telescopic service group is accessed to a standby transmission channel.
8. The enterprise internet of things management system based on scalable deployment of claim 7, wherein: the deployment server comprises a transmission balancing module, the transmission balancing module is configured with a load coordination strategy, and the load coordination strategy comprises
B1, acquiring a transmission protocol from the service requirement characteristics;
b2, screening the corresponding transmission channel according to the transmission protocol, and calculating the transmission capacity value of the obtained transmission channel by a channel load formula, wherein the channel load formula is
Figure FDA0003702830210000041
Where R is the transmission capacity value of the transmission channel,
Figure FDA0003702830210000042
in order to have a success rate of the transmission,
Figure FDA0003702830210000043
in order to transmit the security level,
Figure FDA0003702830210000044
in order to be able to transmit the bandwidth,
Figure FDA0003702830210000045
the index security level is obtained by inquiring an index security level table according to the obtained average value of the transmission security levels of the transmission channels, and the index security level table takes the average value of the transmission security levels and the transmission security level of the current transmission channel as indexes to record the index security level;
b3, summing the obtained transmission capacity values to obtain a total transmission capacity value, and calculating the current transmission redundancy when waiting to receive a transmission coordination request, wherein the current transmission redundancy is the difference between the total transmission capacity value and the total current load value;
b4, acquiring a solid state transmission load and a dynamic transmission load table in the transmission coordination request, where the solid state transmission load reflects the transmission load of the fixed task data of the user side corresponding to the scalable service set, the dynamic transmission load table uses the predicted transmission redundancy as an index, each predicted transmission redundancy corresponds to a time-sharing transmission condition, and the predicted transmission redundancy is the difference between the current transmission redundancy and the solid state transmission load;
and B5, sending the time-sharing transmission condition to an access balancing server, wherein the access balancing server configures transmission time for the dynamic data corresponding to the access user terminal through the time-sharing transmission condition.
9. The enterprise internet of things management system based on scalable deployment of claim 1, wherein: the inter-service cluster is provided with a telescopic prediction module, the telescopic prediction module is connected with a historical telescopic database, historical telescopic data of each telescopic service group are stored in the historical telescopic database, the historical telescopic data comprise telescopic targets and corresponding adjusting parameters, and the telescopic prediction module generates corresponding dynamic receiving instructions to the telescopic response module according to the historical telescopic data.
10. The enterprise internet of things management system based on scalable deployment of claim 1, wherein: the intervening service cluster comprises an index marking module, and the index marking module is used for marking a new user characteristic index for the user characteristic group or deleting marks of the user characteristic index from the user characteristic group.
CN202210696552.5A 2022-06-20 2022-06-20 Enterprise Internet of things management system based on telescopic deployment Active CN115102960B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210696552.5A CN115102960B (en) 2022-06-20 2022-06-20 Enterprise Internet of things management system based on telescopic deployment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210696552.5A CN115102960B (en) 2022-06-20 2022-06-20 Enterprise Internet of things management system based on telescopic deployment

Publications (2)

Publication Number Publication Date
CN115102960A true CN115102960A (en) 2022-09-23
CN115102960B CN115102960B (en) 2023-08-22

Family

ID=83291419

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210696552.5A Active CN115102960B (en) 2022-06-20 2022-06-20 Enterprise Internet of things management system based on telescopic deployment

Country Status (1)

Country Link
CN (1) CN115102960B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115695549A (en) * 2022-10-27 2023-02-03 武汉方舟拓海科技有限公司 Software architecture, terminal equipment and application method of software architecture

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101064641A (en) * 2006-04-28 2007-10-31 国际商业机器公司 Method and system for attribute based index and/or query of Web service resource
CN110489579A (en) * 2019-08-21 2019-11-22 深见网络科技(上海)有限公司 Distributed vector index engine
US20200036584A1 (en) * 2015-04-10 2020-01-30 Comcast Cable Communications, Llc Virtual Gateway Control and Management
US10990572B1 (en) * 2018-09-20 2021-04-27 Amazon Technologies, Inc. Scalable indexing service

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101064641A (en) * 2006-04-28 2007-10-31 国际商业机器公司 Method and system for attribute based index and/or query of Web service resource
US20200036584A1 (en) * 2015-04-10 2020-01-30 Comcast Cable Communications, Llc Virtual Gateway Control and Management
US10990572B1 (en) * 2018-09-20 2021-04-27 Amazon Technologies, Inc. Scalable indexing service
CN110489579A (en) * 2019-08-21 2019-11-22 深见网络科技(上海)有限公司 Distributed vector index engine

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
杨忠;: "面向Docker容器的动态负载集群伸缩研究", 舰船电子工程, no. 08 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115695549A (en) * 2022-10-27 2023-02-03 武汉方舟拓海科技有限公司 Software architecture, terminal equipment and application method of software architecture

Also Published As

Publication number Publication date
CN115102960B (en) 2023-08-22

Similar Documents

Publication Publication Date Title
CN102281190B (en) Networking method for load balancing apparatus, server and client access method
US20050131982A1 (en) System, method and program for allocating computer resources
CN103179217B (en) A kind of load-balancing method for WEB application server farm and device
CN109831524B (en) Load balancing processing method and device
CN111027736A (en) Micro-service combined deployment and scheduling method under multi-objective optimization
CN100440891C (en) Method for balancing gridding load
CN100481783C (en) Control system of grid service container
CN101207550A (en) Load balancing system and method for multi business to implement load balancing
Baker et al. GridRM: an extensible resource monitoring system
CN113645304B (en) Data service processing method and related equipment
CN109327335A (en) A kind of cloud monitoring solution system and method
CN107888666A (en) A kind of cross-region data-storage system and method for data synchronization and device
CN103561078A (en) Telecom operation system and service implementation method
JP4839585B2 (en) Resource information collection and distribution method and system
CN113032419A (en) Multi-source data aggregation search method, device, equipment and storage medium
US7721295B2 (en) Execution multiplicity control system, and method and program for controlling the same
CN115102960B (en) Enterprise Internet of things management system based on telescopic deployment
CN113468221A (en) System integration method based on kafka message data bus
CN109783073A (en) Bullet contracting method and device, the micro services, storage medium of application container
CN111324429B (en) Micro-service combination scheduling method based on multi-generation ancestry reference distance
CN110233904B (en) Equipment updating method, device, system, storage medium and computer equipment
CN113965587B (en) Data acquisition method, device, equipment and medium of artificial intelligence platform
CN107180034A (en) The group system of MySQL database
CN112667747A (en) Dynamic configuration multi-database distributed persistence method supporting user-defined plug-in
CN102325098A (en) Group information acquisition method and system

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